JP7654599B2 - Information processing device, information processing method, and information processing program - Google Patents

Description

The present invention relates to an information processing device, an information processing method, and an information processing program.

Conventionally, various technologies related to advertising have been provided. For example, technologies related to the submission of content such as advertisements have been provided.

JP 2022-003474 A

However, there is room for improvement in the above conventional technology. For example, the above conventional technology merely accepts submissions of content such as advertisements from the source, and when the source selects a category for the destination of the advertisement, the source is not necessarily able to select an appropriate destination category. Therefore, it is desirable to enable appropriate suggestions.

The present application has been made in view of the above, and aims to provide an information processing device, an information processing method, and an information processing program that enable appropriate suggestions.

The information processing device according to the present application is characterized by comprising an extraction unit that extracts a search query that satisfies a predetermined condition as a first keyword from a group of search queries used in a search, and a determination unit that determines a proposed category, which is a category to be proposed as a destination for the first keyword, based on the first keyword extracted by the extraction unit and a second keyword corresponding to each category related to the destination for advertisement submission.

According to one aspect of the embodiment, it is possible to provide appropriate suggestions.

FIG. 1 is a diagram illustrating an example of information processing according to an embodiment. FIG. 2 is a diagram showing a specific example of information processing. FIG. 3 is a diagram illustrating an example of the configuration of an information processing system according to the embodiment. FIG. 4 is a diagram illustrating an example of the configuration of an information processing device according to an embodiment. FIG. 5 is a diagram illustrating an example of a model information storage unit according to the embodiment. FIG. 6 is a flowchart showing an example of the flow of information processing executed by the information processing device. FIG. 7 is a hardware configuration diagram illustrating an example of a computer that realizes the functions of the information processing device.

Below, the information processing device, information processing method, and information processing program according to the present application will be described in detail with reference to the drawings. Note that the information processing device, information processing method, and information processing program according to the present application are not limited to these embodiments. In addition, the same parts in the following embodiments will be denoted by the same reference numerals, and duplicated descriptions will be omitted.

(Embodiment)
[1. Information Processing]
First, an example of information processing executed by the information processing device 100 will be described with reference to Fig. 1. Fig. 1 is a diagram showing an example of information processing according to an embodiment. In the following, a case will be described as an example in which the information processing device 100 determines a category (also called a "proposed category") to be proposed as a submission destination of the first keyword based on a search query (also called a "first keyword") extracted by the information processing device 100 from a search query group and a keyword (also called a "second keyword") corresponding to each category (also simply called a "category") related to the submission destination of the advertisement.

Classifications related to ad submission destinations are, for example, classifications set up when placing ads. Below, campaigns and ad groups are described as examples of classifications related to ad submission destinations. Campaigns and ad groups are concepts of classifications set up when submitting (placing) search ads. For example, an ad group is a role (component) of registering keywords (queries) in association with ads, and when a search is performed using a keyword as a query, an ad associated with that keyword is delivered to the user who performed the search using the keyword. For example, a campaign is a component unit that includes ad groups, and the same advertising target can be divided and set up according to budget and targeting.

1-1. Configuration of Information Processing System
Prior to the description of the process shown in FIG. 1, the configuration of the information processing system 1 will be described with reference to FIG. 3. FIG. 3 is a diagram showing a configuration example of the information processing system according to the embodiment. As shown in FIG. 3, the information processing system 1 includes a terminal device 10, a user terminal 20, an information providing device 50, and an information processing device 100. The terminal device 10, the user terminal 20, the information providing device 50, and the information processing device 100 are connected to each other via a predetermined network N so as to be able to communicate with each other by wire or wirelessly. Note that the information processing system 1 shown in FIG. 3 may include a plurality of terminal devices 10, a plurality of user terminals 20, a plurality of information providing devices 50, and a plurality of information processing devices 100.

The information processing device 100 is a computer that determines a proposed category to be proposed as a destination for the first keyword based on a first keyword extracted from a group of search queries and a second keyword corresponding to each of the categories related to the destination of advertisements. For example, the information processing device 100 uses a vector conversion model that receives a character string such as a keyword as input and outputs vector information (also simply called a "vector") corresponding to the character string, and determines the proposed category using the vectorized information.

The information processing device 100 proposes the determined proposal category to the advertiser. For example, the information processing device 100 provides information indicating the determined proposal category to the advertiser that is the source of the information request. The information processing device 100 provides the first keyword as a proposed keyword to the terminal device 10 used by the advertiser (also referred to as the "target advertiser") to which the information is to be provided. The information processing device 100 provides the proposed category of the first keyword together with the proposed keyword to the terminal device 10.

The terminal device 10 is a computer (advertiser device) used by an advertiser. The advertiser may be a corporation or the like, and the user of the terminal device 10 may be, for example, a publicity officer of the advertiser. For example, the terminal device 10 is realized by a smartphone, a tablet terminal, a notebook PC, a desktop PC, a mobile phone, a PDA, or the like. Note that the example in FIG. 1 shows an example in which the terminal device 10 is a notebook PC. The advertiser uses the terminal device 10 to send and receive information to and from the information processing device 100. For example, the terminal device 10 displays various information provided by the information processing device 100.

The terminal device 10 receives information provided from the information processing device 100 and transmits information to the information processing device 100. For example, the terminal device 10 receives content for accepting designation by an advertiser (also referred to as "designation content") from the information processing device 100. For example, the terminal device 10 receives designation content for accepting designation of keywords by an advertiser from the information processing device 100. For example, the terminal device 10 transmits information indicating keywords (designated keywords) designated by the advertiser using the designation content to the information processing device 100. The terminal device 10 transmits information indicating the keywords designated by the advertiser to the information processing device 100. For example, the terminal device 10 transmits information indicating the keywords designated by the advertiser to the information processing device 100.

In addition, an advertiser may use the terminal device 10 to request an agency to submit advertising content (hereinafter also simply referred to as "advertisement") to the information processing device 100, rather than submitting the advertisement. In this case, it is the agency that submits the advertisement to the information processing device 100. The term "advertiser" is intended to be a concept that includes not only advertisers but also agencies, and the term "advertiser device" is intended to be a concept that includes not only advertiser devices but also agency devices used by agencies.

The user terminal 20 is a computer used by a user. The user terminal 20 is a device (terminal device) that can be carried by a user. The user terminal 20 is realized, for example, by a smartphone, a tablet terminal, a notebook PC (Personal Computer), a mobile phone, a PDA (Personal Digital Assistant), or the like. FIG. 1 shows a case where the user terminal 20 is a smartphone. Note that, hereinafter, the user terminal 20 may be referred to as a user. In other words, hereinafter, the user may also be read as the user terminal 20. For example, the user terminal 20 is a computer that displays advertisements provided by the information processing device 100. For example, the user terminal 20 displays various information such as advertisements using a display app.

The information providing device 50 is a computer that stores information for providing various information related to a service. For example, the information providing device 50 collects information from various external devices such as a web server, and provides the collected information to the information processing device 100. For example, the information providing device 50 is realized by, for example, a server device or a cloud system. For example, the information providing device 50 provides various types of information, such as information related to the service (service information) and information related to users who use the service (user information).

For example, the information providing device 50 may provide a service of providing the information processing device 100 with data used for learning (also referred to as "learning data"). In this case, the information providing device 50 provides the information processing device 100 with data (learning data) used by the information processing device 100 to learn a model. For example, the information providing device 50 provides the information processing device 100 with data (learning data) used by the information processing device 100 to learn a generative model. The information providing device 50 may also provide a service of providing the information processing device 100 with a learned model. For example, the information providing device 50 provides the information processing device 100 with a model (e.g., model M1) used for vector conversion. For example, the information providing device 50 provides the information processing device 100 with a model (e.g., prediction model M2) used to predict an advertising index.

[1-2. Overview of processing in information processing system]
From here, an example of information processing performed by the information processing system 1 will be described with reference to Fig. 1. Note that detailed description of the information processing performed by the information processing system 1 that is the same as conventional processing will be omitted as appropriate.

In FIG. 1, the information processing device 100 acquires information used to execute a determination process for determining a proposed category from the information providing device 50 (step S1). The information processing device 100 receives information used to execute the determination process from the information providing device 50. For example, the information processing device 100 acquires a vector conversion model M1 (also simply referred to as "model M1") that converts a character string into a vector from the information providing device 50. The model M1 is a vector conversion model that inputs a character string and outputs a vector corresponding to the character string. The model M1 outputs a vector obtained by converting a character string into a vector using a word embedding technique such as word2vec. For example, the model M1 outputs a multi-dimensional (e.g., 200-dimensional) vector corresponding to the input character string. Note that the above is merely an example, and any model can be adopted as the model M1 as long as it outputs a vector corresponding to the characteristics of the input character string. In addition, when learning the model M1, the information processing device 100 may acquire data (learning data) used in the learning process of the model M1 from the information providing device 50.

The information processing device 100 stores the information acquired from the information providing device 50 in the storage unit 120 (see FIG. 6) (step S2). The information processing device 100 stores the information received from the information providing device 50 in the storage unit 120. For example, when the information processing device 100 acquires the model M1 from the information providing device 50, the information processing device 100 stores the model M1 in the model information storage unit 121 (see FIG. 4). When the information processing device 100 acquires learning data to be used in the learning process from the information providing device 50, the information processing device 100 stores the acquired learning data to be used in the learning process of the model M1 in the model information storage unit 121 (see FIG. 4). Note that, when the information processing device 100 has already acquired information to be used for executing the determination process to determine the proposed category, it does not need to perform the processes of steps S1 and S2.

The information processing device 100 extracts a first keyword from the search query group CD used in the search (step S3). The information processing device 100 extracts a first keyword from the search query group CD stored in the search query information storage unit 122 (see FIG. 4). The information processing device 100 extracts, from the search query group CD, a search query that a user who used that search query (also referred to as a "query") to access a specific site as a first keyword. In FIG. 1, the information processing device 100 extracts keyword KW1, keyword KW2, etc. as first keywords from the search query group CD, as shown in the extracted query list CL.

In FIG. 1, the keywords are shown as abstract character strings such as KW1, KW2, etc., but each keyword is a character string indicating specific content such as "Naha Hotel" or "Kyoto Sightseeing." In this way, each keyword may be a combination of multiple keywords. When multiple keywords are combined, each keyword may be vectorized individually, or each combination of multiple keywords may be vectorized.

Then, the information processing device 100 executes a process of vectorizing each primary keyword. The information processing device 100 executes a process of generating a vector (also called a "first vector") corresponding to each primary keyword. In FIG. 1, the information processing device 100 generates a first vector corresponding to each primary keyword, as shown in the first vector list FV.

First, the information processing device 100 inputs each of the primary keywords as input information to the model M1 (step S4-1). In FIG. 1, the information processing device 100 inputs each of the primary keywords, such as keywords KW1, KW2, etc., as input information to the model M1.

The model M1 to which the input information has been input outputs a vector (step S4-2). In FIG. 1, the model M1 to which the keyword KW1 has been input outputs a vector KV1. Moreover, the model M1 to which the keyword KW2 has been input outputs a vector KV2. In this way, the information processing device 100 uses the model M1 to generate vectors KV1, KV2, etc., which are first vectors corresponding to each of the first keywords, such as the keywords KW1, KW2, etc.

The information processing device 100 also executes a prediction process to predict advertising indicators corresponding to each first keyword. In FIG. 1, the information processing device 100 predicts advertising indicators corresponding to each first keyword as shown in the predicted value list IL.

The information processing device 100 predicts the predicted CVR of each first keyword using a prediction model M2 that takes as input a vector obtained by vectorizing a keyword and outputs a predicted value of the CVR (conversion rate) when a keyword corresponding to the vector is used as a search query. For example, in response to an input of a vector, the prediction model M2 outputs a predicted CVR value (0 to 100 (%)) when a keyword corresponding to the vector is used as a search query. The information processing device 100 predicts the predicted CVR of the first keyword by inputting the vector of the first keyword into the prediction model M2. Note that the prediction model M2 may be generated for each advertiser, but this will be described later.

In FIG. 1, CVR (conversion rate) is described as an example of an advertising index, but the advertising index is not limited to CVR and any index can be adopted. For example, in addition to CVR, any index related to conversion, such as the number of times conversion was achieved, can be adopted as the advertising index. The conversion here may be any user behavior. For example, the conversion may be a user behavior that leads to the purchase of a product (service use) for the advertised target of the advertisement or the benefit of member registration. Note that purchase (use), member registration, etc. are only examples, and the conversion may be a request for information, viewing (display) a site corresponding to the advertisement, visiting a physical store, etc. In this way, any user behavior that is considered to be a conversion can be set. For example, for each advertisement, the administrator of the information processing device 100, the advertiser, etc. may individually set what user behavior is considered to be a conversion. In addition to the index related to conversion, any index related to clicks, such as the number of clicks (number of clicks) and CTR (click-through rate), can be adopted as the advertising index. Please note that the above advertising metrics are merely examples, and any advertising metrics can be used.

First, the information processing device 100 inputs each of the first vectors as input information to the prediction model M2 (step S5-1). In FIG. 1, the information processing device 100 inputs each of the first vectors, such as vectors KV1, KV2, etc., as input information to the prediction model M2.

The prediction model M2 to which the input information has been input outputs a predicted CVR of the keyword corresponding to the input vector (step S5-2). In FIG. 1, the prediction model M2 to which the vector KV1 has been input outputs a predicted CVR value VL1. Moreover, the prediction model M2 to which the vector KV2 has been input outputs a predicted CVR value VL2. In this way, the information processing device 100 uses the prediction model M2 to predict the predicted CVR corresponding to each first keyword, such as keywords KW1, KW2, etc., which correspond to each of the vectors KV1, KV2, etc.

The information processing device 100 extracts second keywords from the advertising information group AD (step S6). The information processing device 100 extracts second keywords from the advertising information group AD stored in the advertising information storage unit 123 (see FIG. 4). The information processing device 100 extracts keywords associated with each advertising campaign in the advertising information group AD as second keywords. In FIG. 1, the information processing device 100 extracts keywords KW11, KW12, etc. as second keywords for category CP1 (e.g., advertising campaign A) as shown in the category list DL. The information processing device 100 also extracts keywords KW21, KW22, etc. as second keywords for category CP2 (e.g., advertising campaign B).

Then, the information processing device 100 executes a process of generating a vector (also called a "second vector") corresponding to each category using the second keyword of each category. The information processing device 100 executes a process of generating a second vector based on a vector (also called a "sub-vector") obtained by vectorizing the second keyword corresponding to each category. The information processing device 100 calculates the average of the sub-vectors of the second keyword corresponding to each category, and sets the calculated average as the second vector. In FIG. 1, the information processing device 100 generates a second vector corresponding to each category, as shown in the second vector list SV.

First, the information processing device 100 inputs each of the secondary keywords corresponding to each category as input information to the model M1 (step S7-1). In FIG. 1, the information processing device 100 inputs each of the secondary keywords, such as keywords KW11 and KW12, for category CP1 as input information to the model M1. Also, the information processing device 100 inputs each of the secondary keywords, such as keywords KW21 and KW22, for category CP2 as input information to the model M1.

The model M1 to which the input information is input outputs a vector (step S7-2). In FIG. 1, the model M1 to which the keyword KW11 is input outputs a vector KV11. Moreover, the model M1 to which the keyword KW12 is input outputs a vector KV12. The information processing device 100 calculates a vector KV10, which is the average of the vectors KV11, KV12, etc., as the second vector of the section CP1. Moreover, the model M1 to which the keyword KW21 is input outputs a vector KV21. Moreover, the model M1 to which the keyword KW22 is input outputs a vector KV22. The information processing device 100 calculates a vector KV20, which is the average of the vectors KV21, KV22, etc., as the second vector of the section CP1. In this way, the information processing device 100 generates vectors KV1, KV2, etc., which are second vectors corresponding to each section, by using the model M1.

The information processing device 100 determines the proposed category of the first keyword based on the first vector of the first keyword and the second vector corresponding to each of the categories (step S8). The information processing device 100 determines the proposed category of each first keyword based on a comparison between the first vector of each first keyword and the second vector of each of the categories. The information processing device 100 determines the proposed category of the first keyword based on the cosine similarity between the first vector of the first keyword and the second vector of each of the categories. In FIG. 1, the information processing device 100 determines the similar category having the maximum cosine similarity of the second vector with the first vector of the first keyword as the proposed category among the categories shown in the category list DL.

In FIG. 1, as shown in the proposal list PK, for keyword KW1, which is the first keyword, the information processing device 100 determines the proposed category to be category CP2 corresponding to vector KV20, which has the maximum cosine similarity with vector KV1. Also, for keyword KW2, which is the first keyword, the information processing device 100 determines the proposed category to be category CP1 corresponding to vector KV10, which has the maximum cosine similarity with vector KV2.

Then, the information processing device 100 provides the information generated by the above process to the advertiser (step S9). For example, the information processing device 100 transmits the first keyword (proposed keyword) together with information indicating the proposed category corresponding to the provided keyword to the terminal device 10 used by the advertiser. The information processing device 100 transmits the first keyword (proposed keyword) together with information indicating the predicted CVR corresponding to the provided keyword to the terminal device 10 used by the advertiser. For example, the information processing device 100 transmits potential keyword list information that displays a list of potential keywords, with the first keyword as a potential keyword, together with information corresponding to the potential keyword, to the terminal device 10. The potential keyword list information includes information indicating, for each potential keyword, various performance values, predicted CVR, proposed categories such as recommended campaigns and recommended ad groups, and the presence or absence of inflow via search ads. The terminal device 10 displays the potential keyword list information.

In this way, the information processing device 100 can make appropriate proposals by extracting first keywords from a group of search queries and determining the proposed category of the first keywords based on the extracted first keywords and second keywords corresponding to each category related to the advertisement submission destination. Furthermore, the information processing device 100 can provide the advertiser with an appropriately determined proposed category by providing the first keywords as proposed keywords to the terminal device 10 used by the target advertiser, thereby making appropriate proposals.

[1-2-1. Specific examples of processing]
On the premise of the information processing performed by the information processing system 1 described above, a specific example of processing by the information processing device 100 will be described with reference to Fig. 2. Fig. 2 is a diagram showing a specific example of information processing. Note that the same points as those described above will not be described as appropriate.

In FIG. 2, the keyword KW2, which is the first keyword, is "Naha Hotel", and the first vector corresponding to the keyword KW2 is the vector KV2, which is "0.6, 0.3". The information processing device 100 converts "Naha Hotel" to "0.6, 0.3" using the model M1.

2, the keyword KW11, which is "Okinawa", and the keyword KW12, which is "Churaumi", are associated with the category CP1, which is the advertising campaign A. The information processing device 100 converts "Okinawa" into the vector KV11, "0.7, 0.2", using the model M1. The information processing device 100 also converts "Churaumi" into the vector KV12, "0.5, 0.2", using the model M1. The information processing device 100 also calculates the average of the sub-vectors "0.7, 0.2" and "0.5, 0.2" to generate the vector KV10, "0.6, 0.2". As shown in the similarity information DS1, the information processing device 100 calculates the cosine similarity between the vector KV2, which is the first vector ("0.6, 0.3"), and the vector KV10, which is the second vector ("0.6, 0.2"), to be "0.9".

2, the keyword KW21, which is "Hokkaido", and the keyword KW22, which is "Shiretoko Peninsula", are associated with the category CP2, which is the advertising campaign B. The information processing device 100 converts "Hokkaido" into the vector KV21, "0.2, 0.8", using the model M1. The information processing device 100 also converts "Shiretoko Peninsula" into the vector KV22, "0.4, 0.4", using the model M1. The information processing device 100 also calculates the average of the sub-vectors "0.2, 0.8" and "0.4, 0.4", to generate the vector KV20, "0.3, 0.6". As shown in the similarity information DS2, the information processing device 100 calculates the cosine similarity between the vector KV2 ("0.6, 0.3"), which is the first vector, and the vector KV20 ("0.3, 0.6"), which is the second vector, to be "0.5".

In FIG. 2, for the first keyword (keyword KW2) "Okinawa", the information processing device 100 determines advertising campaign A, which is category CP1 corresponding to vector KV10 that has the maximum cosine similarity with vector KV2, as the proposed category. That is, in FIG. 2, the information processing device 100 determines that for the first keyword (keyword KW2) "Okinawa", advertising campaign A is to be recommended to the advertiser.

Below, a detailed specific example will be briefly described, taking as an example a case where the target of information provision is advertiser X. The information processing device 100 generates a prediction model M2 that predicts the likelihood of conversion from the registered keywords of advertiser X. For example, the information processing device 100 calculates the registered keywords and CVR performance values for the search advertising account of advertiser X.

The information processing device 100 performs the following processing as preprocessing for registered keywords. For example, the information processing device 100 may refer to a configuration file, etc., correct spelling variations in brand names, etc., and then exclude brand names, etc. in a partial match search. Note that brand names and other nomination keywords may have a high CVR and become outliers, so the information processing device 100 can exclude abnormal values of keywords by this exclusion and suppress a decrease in prediction accuracy. The information processing device 100 also excludes outliers in CVR and performs normalization processing to keep the CVR between 0 and 1. Note that the information processing device 100 may perform random sampling when the learning data is above a certain threshold in order to reduce the processing load.

The information processing device 100 performs morphological analysis on the registered keywords. The information processing device 100 vectorizes the words after the morphological analysis. For example, the information processing device 100 averages and aggregates each vector for each keyword to create a 200-dimensional vector. The information processing device 100 uses the CVR as the objective variable and learns a prediction model M2 from the 200-dimensional vector of each word.

The information processing device 100 scores the probability of conversion for the search query. The information processing device 100 extracts search queries that flow into advertiser X and its competitors, and generates a list of suggested keywords. The information processing device 100 performs the above-mentioned pre-processing on the suggested keywords.

The information processing device 100 performs morphological analysis on the search query. The information processing device 100 vectorizes the words after the morphological analysis. The information processing device 100 inputs the vectors to a prediction model M2 and generates a score that predicts the CVR (predicted CVR). For example, the information processing device 100 normalizes the score so that the minimum value is 0 and the maximum value is 1.

The information processing device 100 extracts keywords for each category, such as campaigns and ad groups, registered in search ads. The information processing device 100 extracts keywords registered in advertiser X's search ad account and the associated ad campaigns and ad groups.

Then, the information processing device 100 performs pre-processing, excluding the removal of CVR outliers, on the keywords registered in the search advertising account of advertiser X. The information processing device 100 performs morphological analysis on the registered keywords. The information processing device 100 vectorizes the words of the campaign and the advertising group, calculates the average value, and calculates the cosine similarity between the calculated average value and the vector of each proposed keyword for all cases. As shown in FIG. 2, the information processing device 100 determines the information to be proposed to advertiser X based on the calculated cosine similarity.

As described above, the information processing device 100 discovers keywords (queries) that have not been reached in the operation of search advertising, and proposes them together with numerical values that serve as operational indicators. The information processing device 100 extracts unsubmitted queries as extracted queries from among queries that have flowed into the site of the advertiser (business owner) to which the advertisement is proposed and the advertiser's competing sites. The information processing device 100 assigns a predicted score (predicted CVR, etc.) to the extracted queries. For example, the information processing device 100 breaks down the query into parts of speech, vector-converts and quantifies the meaning, generates a machine learning model, and predicts a score (predicted CVR, etc.). The information processing device 100 proposes optimal campaign or ad group divisions, etc., based on, for example, the advertiser's account structure and submitted keywords.

When input, the information processing device 100 vectorizes the query candidate that is the target of delivery of a specified advertisement and the advertisement, and if the vector of the query candidate is similar to the vector of the advertisement, it proposes the query candidate as a new query. For example, the information processing device 100 uses a model that converts text with a similar intent into a similar vector. Conventionally, in an advertising campaign, etc., an advertisement is displayed when a registered keyword is input, and since the keywords are registered manually, keywords are missed. Conventionally, it has not been possible to respond when a new word is generated.

Therefore, the information processing device 100 vectorizes the keywords related to the advertisement. Specifically, the information processing device 100 extracts keywords from any information such as the advertisement text, the set advertisement keywords, and various other metadata related to the advertisement. The information processing device 100 breaks down the extracted keywords into parts of speech. The information processing device 100 vectorizes the keywords for each part of speech using a model. The information processing device 100 links the average value of the vectors to the advertisement. The information processing device 100 may generate a compound query or a long query consisting of multiple words at a predetermined time interval, or may use the history of search queries. The information processing device 100 breaks down and vectorizes the keywords for each part of speech, and obtains the average. The information processing device 100 compares the vector of the search query with the vector of the advertisement to identify a search query with a high similarity. The information processing device 100 proposes a search query that has not been set among the identified search queries. The information processing device 100 may preferentially propose a query with a higher similarity. Note that the above-described process is merely an example, and the information processing device 100 may use various information as appropriate to determine the information to be proposed and make the proposal to the advertiser.

2. Configuration of information processing device
Next, the configuration of the information processing device 100 according to the embodiment will be described with reference to Fig. 4. Fig. 4 is a diagram showing an example of the configuration of the information processing device 100 according to the embodiment. As shown in Fig. 4, the information processing device 100 has a communication unit 110, a storage unit 120, and a control unit 130. Note that the information processing device 100 may have an input unit (e.g., a keyboard, a mouse, etc.) that accepts various operations from an administrator of the information processing device 100, and a display unit (e.g., a liquid crystal display, etc.) that displays various information.

(Communication unit 110)
The communication unit 110 is realized by, for example, a network interface card (NIC) etc. The communication unit 110 is connected to a network via a wired or wireless connection, and transmits and receives information between the terminal device 10, the user terminal 20, and the information providing device 50.

(Memory unit 120)
The storage unit 120 is realized by, for example, a semiconductor memory element such as a random access memory (RAM) or a flash memory, or a storage device such as a hard disk or an optical disk. As shown in Fig. 4, the storage unit 120 according to the embodiment has a model information storage unit 121, a search query information storage unit 122, and an advertisement information storage unit 123. The storage unit 120 may store various information, not limited to the above.

(Model information storage unit 121)
The model information storage unit 121 according to the embodiment stores information about the model. For example, the model information storage unit 121 stores information (model data) of a trained model (model) trained (generated) by a learning process. The model information storage unit 121 shown in FIG. 5 stores data used in the learning (trained data) in association with the trained model (model). FIG. 5 is a diagram showing an example of a model information storage unit according to the embodiment. In the example shown in FIG. 5, the model information storage unit 121 includes items such as "model ID", "purpose", "model data", and "trained data". In the example of FIG. 5, the model information storage unit 121 stores data used in the learning (trained data) in association with the trained model (model).

"Model ID" indicates identification information for identifying a model. "Use" indicates the use of the corresponding model. "Model data" indicates the data of the model. Figure 5 etc. shows an example in which conceptual information such as "MDT1" is stored in "model data", but in reality, various information that constitutes the model is included, such as information on the model configuration (network configuration) and information on parameters. For example, "model data" includes information including the nodes in each layer of the network, the functions employed by each node, the connection relationships between the nodes, and the connection coefficients set for the connections between the nodes.

"Learning data" refers to data used to train a trained model (model). In "Learning data", information indicating the data set used to train the corresponding model is stored. For example, "Learning data" associates data (input information) with the correct answer information (output information) corresponding to that data, and stores the data as learning data (also called "data for learning"). Figure 5 shows an example in which conceptual information such as "LDT1" is stored in "Learning data", but in reality, various information related to the data used to train the corresponding model, such as data (input information) and the correct answer information (output information) corresponding to that data, is included.

In FIG. 5, the model (model M1) identified by the model ID "M1" indicates that its application is "vector transformation." In other words, model M1 indicates that it is a model that outputs (generates) a vector corresponding to an input character string (keyword). Also, it indicates that the model data of model M1 is model data MDT1. Also, it indicates that the learning data used to learn model M1 is learning data LDT1.

Also, the model (prediction model M2) identified by the model ID "M2" indicates that its use is "CVR prediction." In other words, the prediction model M2 indicates that it is a model that outputs information (score) indicating the predicted CVR for a keyword corresponding to an input vector. For example, the prediction model M2 outputs a value from 0 to 100 (%). For example, the prediction model M2 outputs a predicted value of the CVR for a keyword corresponding to an input vector. This indicates that the model data of the prediction model M2 is model data MDT2. This also indicates that the learning data used to train the prediction model M2 is learning data LDT2. The learning data LDT2 includes multiple teacher data in which the CVR value (actual value, etc.) of an advertising copy is associated with the advertising copy as correct answer information.

The model information storage unit 121 may store various information according to the purpose, not limited to the above. For example, the model information storage unit 121 may store a model that predicts other advertising indicators, not limited to CVR. For example, the model information storage unit 121 may store model M3 that predicts CTR. In this case, the model information storage unit 121 stores model M3 that predicts CTR in association with the data used for learning (a combination of CTR, which is correct information, and the advertising copy from which the actual CTR was obtained).

(Search query information storage unit 122)
The search query information storage unit 122 according to the embodiment stores various pieces of information related to a search query The search query information storage unit 122 stores information related to a keyword used as a search query.

For example, the search query information storage unit 122 stores a keyword used as a search query in association with information identifying the user who used the keyword (user ID). For example, the search query information storage unit 122 stores a keyword used as a search query in association with behavioral information indicating the post-search behavior of the user who used the keyword. For example, the search query information storage unit 122 stores a keyword used as a search query in association with behavioral information indicating that the user who used the keyword accessed a specific site.

The search query information storage unit 122 may store various types of information depending on the purpose, not limited to the above.

(Advertisement information storage unit 123)
The advertisement information storage unit 123 according to the embodiment stores various information related to advertisements. For example, the advertisement information storage unit 123 stores various information related to search advertisements. For example, the advertisement information storage unit 123 stores various information related to categories related to advertisement submission destinations, such as campaigns and ad groups.

For example, the advertisement information storage unit 123 stores various information regarding the classification of advertisement submission destinations for each account (advertiser). For example, the advertisement information storage unit 123 stores various information regarding the classification of advertisement submission destinations, such as campaigns and ad groups set by each advertiser, in association with information that identifies each advertiser (such as an advertiser ID). Note that advertisers may request submission work, etc., from an agency. The term "advertiser" is intended to be a concept that includes not only advertisers but also agencies.

For example, the advertising information storage unit 123 stores, for each campaign, the budget, period, and targeting (targeted user attributes, etc.) set for that campaign. For each campaign, the advertising information storage unit 123 stores the advertising group set for that campaign. For example, the advertising information storage unit 123 stores, for each campaign, the delivery performance for that campaign.

For example, the advertisement information storage unit 123 stores, for each advertisement group, the keywords and advertisements set for that advertisement group. For example, the advertisements are keywords specified by the advertiser. For example, the advertisements are advertisements submitted by the advertiser. For example, the advertisements store text used as advertisements, or file path names indicating the storage locations of these, etc. Note that the advertisements may include images, and may be data such as video images and audio, so long as they include advertisement text.

For example, the advertising information storage unit 123 stores, for each ad group, the delivery performance of that ad group. For example, the advertising information storage unit 123 stores, for each ad group, advertising indicators based on the delivery performance of that ad group. For example, the advertising information storage unit 123 stores, for each ad group, advertising indicators based on the delivery performance collected for that ad group, or file path names indicating the storage locations of these, etc. For example, the advertising information storage unit 123 stores, for each ad group, the performance values of various advertising indicators such as CTR and CVR corresponding to the ads in the ad group.

The advertisement information storage unit 123 may store various information according to the purpose, not limited to the above. For example, the advertisement information storage unit 123 may store information related to the bid price, etc.

(Control unit 130)
Returning to the explanation of Fig. 4, the control unit 130 is a controller, and is realized, for example, by a CPU (Central Processing Unit), an MPU (Micro Processing Unit), or the like, executing various programs (corresponding to an example of an information processing program) stored in a storage device inside the information processing device 100 using a RAM as a working area. The control unit 130 is also a controller, and is realized, for example, by an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array).

As shown in FIG. 4, the control unit 130 has an acquisition unit 131, a learning unit 132, an extraction unit 133, a determination unit 134, and a provision unit 135, and realizes or executes the functions and actions of the information processing described below. Note that the internal configuration of the control unit 130 is not limited to the configuration shown in FIG. 4, and may be other configurations as long as they perform the information processing described below.

(Acquisition unit 131)
The acquisition unit 131 acquires various pieces of information. The acquisition unit 131 acquires various pieces of information from external devices such as the terminal device 10, the user terminal 20, and the information providing device 50. The acquisition unit 131 also acquires various pieces of information from the storage unit 120. The acquisition unit 131 also acquires various pieces of information from the model information storage unit 121, the search query information storage unit 122, the advertisement information storage unit 123, and the like.

The acquisition unit 131 receives various information from an external information processing device via the communication unit 110. The acquisition unit 131 stores the received information in the storage unit 120. The acquisition unit 131 receives various information from the terminal device 10, the user terminal 20, or the information providing device 50. The acquisition unit 131 acquires information from the terminal device 10 used by the user. For example, the acquisition unit 131 acquires various information from the information providing device 50. For example, the acquisition unit 131 receives learning data used for learning the model from the information providing device 50. For example, the acquisition unit 131 receives the model from the information providing device 50.

The acquisition unit 131 acquires a group of search queries. The acquisition unit 131 acquires information indicating second keywords corresponding to each of the categories.

(Learning Unit 132)
The learning unit 132 executes a learning process to learn a learning model (model). Note that, when the information processing device 100 acquires learning models such as the models M1 and M2 from the information providing device 50, the information processing device 100 does not need to include the learning unit 132.

For example, the learning unit 132 executes a learning process based on various information acquired by the acquisition unit 131. The learning unit 132 executes a learning process based on information from an external information processing device and information stored in the storage unit 120. The learning unit 132 executes a learning process based on information stored in the model information storage unit 121. The learning unit 132 stores a model generated by learning in the model information storage unit 121.

The learning unit 132 performs a learning process. The learning unit 132 performs various types of learning. The learning unit 132 learns various types of information based on the information acquired by the acquisition unit 131. The learning unit 132 learns (generates) a model. The learning unit 132 learns various types of information such as a model. The learning unit 132 generates a model through learning. The learning unit 132 learns the model using various machine learning techniques. For example, the learning unit 132 learns the parameters of the model (network). The learning unit 132 learns the model using various machine learning techniques.

The learning unit 132 generates various learning models such as model M1. The learning unit 132 learns network parameters. For example, the learning unit 132 learns network parameters of various learning models such as model M1. The learning unit 132 generates various learning models such as model M1 by performing a learning process using the learning data stored in the model information storage unit 121. For example, the learning unit 132 generates a model used for vector conversion. The learning unit 132 generates various learning models such as model M1 by learning network parameters of various learning models such as model M1.

The learning unit 132 performs a learning process based on the learning data (teacher data) stored in the model information storage unit 121. The learning unit 132 performs a learning process using the learning data stored in the model information storage unit 121 to generate various learning models such as model M1.

For example, the learning unit 132 performs a learning process using a method such as backpropagation so that the vector output by the model M1 approaches correct answer information (desired vector) associated with the character string (input information) input to the model M1. For example, the learning unit 132 performs a learning process so that the vector output by the model M1 to which input information, which is a character string such as a keyword, is input, approaches correct answer information associated with the input information.

For example, the learning unit 132 adjusts the value of the weight (i.e., the connection coefficient) that is taken into account when values are transmitted between nodes by the learning process. In this way, the learning unit 132 learns the model M1 by processing such as backpropagation that corrects the parameters (connection coefficients) so as to reduce the error between the output in the model M1 and the correct answer information corresponding to the input. For example, the learning unit 132 generates the model M1 by performing processing such as backpropagation to minimize a predetermined loss function. This allows the learning unit 132 to perform a learning process that learns the parameters of the model M1.

The model learning method is not limited to the above-mentioned method, and any known technology can be applied. Each model may be generated using various conventional techniques related to machine learning as appropriate. For example, the model may be generated using a supervised learning machine learning technique such as SVM (Support Vector Machine). Also, for example, the model may be generated using an unsupervised learning machine learning technique. For example, the model may be generated using a deep learning technique. For example, the model may be generated using various deep learning techniques such as DNN (Deep Neural Network), RNN (Recurrent Neural Network), and CNN (Convolutional Neural Network). Note that the above description of the model generation is an example, and the model may be generated by a learning method appropriately selected according to the obtainable information, etc. That is, the learning unit 132 may generate the model M1 by any method as long as the model M1 can be trained to output a score corresponding to the correct answer information when input information included in the learning data is input.

As described above, the learning method used by the learning unit 132 is not particularly limited, but for example, learning data that links data (input information) with its correct answer information (output information) may be prepared, and the learning data may be input to a computational model based on a multi-layer neural network for learning. In addition, a method based on DNN, such as CNN or 3D-CNN, may be used. When targeting time-series data such as voice, the learning unit 132 may use a method based on a recurrent neural network (RNN) or LSTM (Long Short-Term Memory units), which is an extension of RNN.

The learning unit 132 may also learn a prediction model M2 that is a model used to predict advertising indicators. For example, the learning unit 132 may learn a prediction model M2 that is a model used to predict CVR. For example, the learning unit 132 learns the prediction model M2 using learning data LDT2 that includes multiple teacher data in which the CVR value of the advertising copy (such as an actual value of 0 to 100 (%)) is associated with the advertising copy as correct answer information. For example, the learning unit 132 learns the prediction model M2 so that when a vector is input, correct answer information (value) associated with a keyword corresponding to the vector is output. For example, the learning unit 132 learns a prediction model M2 that outputs a predicted CVR value (0 to 100 (%)) when a keyword corresponding to the vector is used as a search query in response to the input of the vector. Note that when the prediction model M2 is acquired from the information providing device 50, the learning unit 132 does not need to learn the prediction model M2.

(Extraction Unit 133)
The extraction unit 133 extracts various information. For example, the extraction unit 133 extracts various information from various information acquired from an external information processing device such as the terminal device 10, the user terminal 20, or the information providing device 50. For example, the extraction unit 133 extracts various information from various information stored in the storage unit 120. For example, the acquisition unit 131 extracts various information from various information stored in the model information storage unit 121, the search query information storage unit 122, the advertisement information storage unit 123, etc. The acquisition unit 131 extracts a query that satisfies a condition from the search query information storage unit 122.

For example, the extraction unit 133 extracts various pieces of information from the various pieces of information acquired by the acquisition unit 131. The extraction unit 133 generates various pieces of information using the extracted information. The extraction unit 133 calculates various pieces of information using the extracted information. The extraction unit 133 estimates various pieces of information using the extracted information.

The extraction unit 133 extracts, as a first keyword, a search query that satisfies a predetermined condition from the group of search queries used in the search. The extraction unit 133 extracts, from the group of search queries, a search query in which the user who performed the search performed a predetermined action as a first keyword. The extraction unit 133 extracts, from the group of search queries, a search query in which the user accessed a predetermined site as a first keyword. The extraction unit 133 determines a proposal category for the first keyword based on the first keyword and the second keyword acquired by the acquisition unit 131. The extraction unit 133 extracts the first keyword from the group of search queries acquired by the acquisition unit 131.

(Determination Unit 134)
The determination unit 134 executes a determination process to determine various pieces of information. The determination unit 134 stores the information determined by the determination process in the storage unit 120. For example, the determination unit 134 executes the determination process based on various pieces of information acquired by the acquisition unit 131. For example, the determination unit 134 executes the determination process based on various pieces of information extracted by the extraction unit 133. The determination unit 134 executes the determination process based on various pieces of information stored in the storage unit 120. For example, the determination unit 134 executes the determination process based on various pieces of information received from an external information processing device.

The determination unit 134 executes the determination process using the model learned by the learning unit 132. The determination unit 134 executes the determination process using the model stored in the model information storage unit 121.

The determination unit 134 determines, from the second advertising copy generated by the learning model, an advertising candidate copy that is a candidate for use in an advertisement. The determination unit 134 determines, as the advertising candidate copy, the second advertising copy whose advertising indicators satisfy a predetermined condition. The determination unit 134 determines, as the advertising candidate copy, the second advertising copy whose language-likeness condition is satisfied.

The determination unit 134 determines a proposed category, which is a category proposed as a destination for the first keyword, based on the first keyword extracted by the extraction unit 133 and the second keyword corresponding to each of the categories related to the destination of advertisement submission. The determination unit 134 determines the proposed category for the first keyword based on first vector information in which the first keyword is converted into a vector, and second vector information for each of the categories based on the vector conversion of the second keyword corresponding to each of the categories.

The determination unit 134 determines a proposed category for the first keyword based on a comparison between first vector information of the first keyword generated using a vector conversion model M1 that receives a character string and outputs a vector corresponding to the character string, and second vector information of each of the categories generated using the vector conversion model M1. The determination unit 134 determines a proposed category for the first keyword based on a comparison between the first vector information of the first keyword and the second vector information of each of the categories.

The determination unit 134 determines a proposed category for the first keyword based on the similarity between the first vector information of the first keyword and the second vector information of each of the categories. The determination unit 134 determines a proposed category for the first keyword based on the similarity, which is the cosine similarity, between the first vector information of the first keyword and the second vector information of each of the categories. The determination unit 134 determines, among the categories, the similar category in which the similarity of the second vector information to the first vector information of the first keyword is the greatest, as the proposed category.

The determination unit 134 receives a vector as input, and uses a prediction model M2 that outputs a predicted value of an advertising index related to a search advertisement for a keyword corresponding to the vector to determine a predicted value of an advertising index for the first keyword. The determination unit 134 uses a prediction model M2 that outputs a predicted value of an advertising index related to a conversion to determine a predicted value of an advertising index related to a conversion by a search using the first keyword. The determination unit 134 uses a prediction model M2 that outputs a predicted value of an advertising index related to the number of times a conversion has been reached to determine a predicted value of an advertising index related to the number of times a conversion has been reached to a search using the first keyword. The determination unit 134 uses a prediction model M2 that outputs a predicted value of a conversion rate to determine a predicted value of a conversion rate by a search using the first keyword.

The determination unit 134 determines a proposed category for the first keyword based on the first keyword and a second keyword corresponding to each of the categories, which are campaigns or ad groups.

The determination unit 134 executes a generation process to generate various information. For example, the determination unit 134 executes the generation process based on various information acquired by the acquisition unit 131. For example, the determination unit 134 executes the generation process based on information stored in the storage unit 120. The determination unit 134 executes the generation process based on the determined information.

The determination unit 134 generates content. For example, the determination unit 134 generates a screen (content) to be provided to the terminal device 10 by appropriately using various technologies such as Java (registered trademark). Note that the determination unit 134 may generate the screen (content) to be provided to the terminal device 10 based on the format of CSS, JavaScript (registered trademark), or HTML. Also, for example, the determination unit 134 may generate the screen (content) in various formats such as JPEG (Joint Photographic Experts Group), GIF (Graphics Interchange Format), or PNG (Portable Network Graphics). The determination unit 134 generates content for receiving evaluations.

(Providing Unit 135)
The providing unit 135 provides various information. The providing unit 135 transmits the various information to an external information processing device via the communication unit 110. The providing unit 135 transmits the various information to the terminal device 10, the user terminal 20, and the information providing device 50. The providing unit 135 transmits a result of the determination process to the terminal device 10.

The providing unit 135 provides the information extracted by the extracting unit 133. The providing unit 135 provides the information determined by the determining unit 134. The providing unit 135 transmits the determined advertisement candidate text to the terminal device 10. The providing unit 135 provides the information generated by the determining unit 134. The providing unit 135 transmits the determined advertisement to the user terminal 20.

The providing unit 135 provides the first keyword as a suggested keyword to the terminal device 10 used by the target advertiser, who is the advertiser to which information is to be provided. The providing unit 135 provides the proposed category of the first keyword determined by the determining unit 134 to the terminal device 10 together with the suggested keyword.

3. Processing Procedure
Next, the procedure of information processing executed by the information processing device 100 according to the embodiment will be described with reference to Fig. 6. Fig. 6 is a flowchart showing an example of the flow of information processing executed by the information processing device.

In FIG. 6, the information processing device 100 extracts a search query that satisfies a predetermined condition from a group of search queries used in a search as a first keyword (step S101). For example, the information processing device 100 extracts a search query that a user used to access a predetermined site from the group of search queries as a first keyword.

The information processing device 100 also determines a proposed category, which is a category proposed as a destination for the first keyword, based on the first keyword and the second keyword corresponding to each category related to the destination of the advertisement (step S102). For example, the information processing device 100 determines a campaign or an ad group as the proposed category for the first keyword based on the first keyword and the second keyword.

4. Effects
As described above, the information processing device 100 according to the embodiment includes the extraction unit 133 and the determination unit 134. The extraction unit 133 extracts a search query that satisfies a predetermined condition from a group of search queries used in a search as a first keyword. The determination unit 134 determines a proposed category that is a category proposed as a submission destination of the first keyword, based on the first keyword extracted by the extraction unit 133 and second keywords corresponding to each category related to the submission destination of the advertisement.

In this way, the information processing device 100 according to the embodiment can make appropriate suggestions by extracting first keywords from a group of search queries and determining the suggested category of the first keywords based on the extracted first keywords and second keywords corresponding to each of the categories related to the advertisement submission destinations.

In addition, in the information processing device 100 according to the embodiment, the determination unit 134 determines the proposed category of the first keyword based on the first vector information in which the first keyword is converted into a vector and the second vector information of each category based on the vector conversion of the second keyword corresponding to each category.

In this way, the information processing device 100 according to the embodiment can enable appropriate suggestions by determining the suggested category of the primary keyword based on the first vector information of the primary keyword and the second vector information of each category.

In addition, in the information processing device 100 according to the embodiment, the determination unit 134 determines a proposed category for the first keyword based on a comparison between first vector information of the first keyword generated using a vector conversion model M1 that receives a character string as input and outputs a vector corresponding to the character string, and second vector information of each of the categories generated using the vector conversion model M1.

In this way, the information processing device 100 according to the embodiment can make appropriate suggestions by determining the suggested category of the first keyword based on the first vector information and the second vector information generated using the vector conversion model M1, which takes a character string as input and outputs a vector corresponding to the character string.

In addition, in the information processing device 100 according to the embodiment, the determination unit 134 determines the proposed category for the primary keyword based on a comparison between the first vector information of the primary keyword and the second vector information of each of the categories.

In this way, the information processing device 100 according to the embodiment can enable appropriate suggestions by determining the suggested category for the first keyword based on a comparison between the first vector information of the first keyword and the second vector information of each category.

In addition, in the information processing device 100 according to the embodiment, the determination unit 134 determines the proposed category for the primary keyword based on the similarity between the first vector information of the primary keyword and the second vector information of each of the categories.

In this way, the information processing device 100 according to the embodiment can enable appropriate suggestions by determining the suggested category of the first keyword based on the similarity between the first vector information of the first keyword and the second vector information of each category.

In addition, in the information processing device 100 according to the embodiment, the determination unit 134 determines the proposed category for the primary keyword based on the similarity, which is the cosine similarity between the first vector information of the primary keyword and the second vector information of each of the categories.

In this way, the information processing device 100 according to the embodiment can enable appropriate suggestions by determining the proposed category of the first keyword based on the similarity, which is the cosine similarity between the first vector information of the first keyword and the second vector information of each category.

In addition, in the information processing device 100 according to the embodiment, the determination unit 134 determines, among the categories, the similar category in which the similarity of the second vector information to the first vector information of the first keyword is the greatest, as the proposed category.

In this way, the information processing device 100 according to the embodiment can make appropriate suggestions by determining, among the categories, the similar category in which the similarity of the second vector information to the first vector information of the first keyword is the greatest as the suggested category.

In addition, in the information processing device 100 according to the embodiment, the determination unit 134 determines a predicted value of the advertising index for the first keyword using a prediction model M2 that receives a vector as input and outputs a predicted value of the advertising index for the search advertisement of the keyword corresponding to the vector.

In this way, the information processing device 100 according to the embodiment can appropriately predict the predicted value of the advertising index of the first keyword by using the prediction model M2 to determine the predicted value of the advertising index of the first keyword.

In addition, in the information processing device 100 according to the embodiment, the determination unit 134 determines a predicted value of an advertising index related to a conversion resulting from a search using the first keyword, using a prediction model M2 that outputs a predicted value of an advertising index related to a conversion.

In this way, the information processing device 100 according to the embodiment can appropriately predict the predicted value of the advertising index for the first keyword by using the prediction model M2 to determine the predicted value of the advertising index for the conversion of the first keyword.

In addition, in the information processing device 100 according to the embodiment, the determination unit 134 determines a predicted value of an advertising indicator related to the number of times a search using the first keyword leads to a conversion, using a prediction model M2 that outputs a predicted value of an advertising indicator related to the number of times a conversion is reached.

In this way, the information processing device 100 according to the embodiment can appropriately predict the predicted value of the advertising indicator for the first keyword by using the prediction model M2 to determine the predicted value of the advertising indicator related to the number of times the first keyword has led to conversion.

In addition, in the information processing device 100 according to the embodiment, the determination unit 134 determines a predicted value of the conversion rate resulting from a search using the first keyword using a prediction model M2 that outputs a predicted value of the conversion rate.

In this way, the information processing device 100 according to the embodiment can appropriately predict the conversion rate of the primary keyword by using the prediction model M2 to determine a predicted value of the conversion rate of the primary keyword.

In addition, in the information processing device 100 according to the embodiment, the extraction unit 133 extracts, from the group of search queries, a search query in which the user who performed the search performed a predetermined action as a first keyword.

In this way, the information processing device 100 according to the embodiment can make appropriate suggestions by extracting, from among a group of search queries, a search query in which the user who performed the search performed a predetermined action as a first keyword.

In addition, in the information processing device 100 according to the embodiment, the extraction unit 133 extracts, from the group of search queries, a search query that caused a user to access a specific site as a first keyword.

In this way, the information processing device 100 according to the embodiment can make appropriate suggestions by extracting, from among a group of search queries, a search query that a user used to access a specific site as a first keyword.

In addition, in the information processing device 100 according to the embodiment, the determination unit 134 determines the proposed category of the first keyword based on the first keyword and the second keyword corresponding to each of the categories, which are campaigns or ad groups.

In this way, the information processing device 100 according to the embodiment can make appropriate suggestions by determining the first keyword and the proposal category, which is a campaign or an ad group.

The information processing device 100 according to the embodiment also includes a providing unit 135. The providing unit 135 provides the first keyword as a suggested keyword to the terminal device 10 used by the target advertiser, who is the advertiser to whom the information is to be provided.

In this way, the information processing device 100 according to the embodiment can provide the first keyword as a suggested keyword to the terminal device 10 used by the target advertiser, who is the advertiser to whom the information is to be provided, thereby providing the advertiser with an appropriately determined proposed category and making an appropriate proposal.

In addition, in the information processing device 100 according to the embodiment, the providing unit 135 provides the proposed category of the first keyword determined by the determining unit 134 to the terminal device 10 together with the proposed keyword.

In this way, the information processing device 100 according to the embodiment can provide the proposed category of the first keyword together with the proposed keyword to the terminal device 10, thereby providing the advertiser with an appropriately determined proposed category together with the first keyword, and making an appropriate proposal.

The information processing device 100 according to the embodiment also includes an acquisition unit 131. The acquisition unit 131 acquires a group of search queries. The extraction unit 133 extracts a first keyword from the group of search queries acquired by the acquisition unit 131.

In this way, the information processing device 100 according to the embodiment can make appropriate suggestions by extracting first keywords from the acquired group of search queries and determining the suggested category of the first keywords based on the extracted first keywords and second keywords corresponding to each of the categories related to the advertisement submission destinations.

In addition, in the information processing device 100 according to the embodiment, the acquisition unit 131 acquires information indicating secondary keywords corresponding to each of the categories. The extraction unit 133 determines the proposed category of the primary keyword based on the primary keyword and the secondary keyword acquired by the acquisition unit 131.

In this way, the information processing device 100 according to the embodiment can make appropriate suggestions by determining the suggestion category of the primary keyword based on the primary keyword and the acquired secondary keyword.

5. Hardware Configuration
The information processing device 100 according to the embodiment described above is realized by a computer 1000 having a configuration as shown in Fig. 7. Fig. 7 is a hardware configuration diagram showing an example of a computer that realizes the functions of the information processing device. The computer 1000 has a CPU 1100, a RAM 1200, a ROM 1300, a HDD (Hard Disk Drive) 1400, a communication interface (I/F) 1500, an input/output interface (I/F) 1600, and a media interface (I/F) 1700.

The CPU 1100 operates based on a program stored in the ROM 1300 or the HDD 1400, and controls each component. The ROM 1300 stores a boot program executed by the CPU 1100 when the computer 1000 is started, and programs that depend on the hardware of the computer 1000, etc.

HDD 1400 stores programs executed by CPU 1100 and data used by such programs. Communication interface 1500 receives data from other devices via network N and sends it to CPU 1100, and transmits data generated by CPU 1100 to other devices via network N.

The CPU 1100 controls output devices such as a display and a printer, and input devices such as a keyboard and a mouse, via the input/output interface 1600. The CPU 1100 acquires data from the input devices via the input/output interface 1600. The CPU 1100 also outputs generated data to the output devices via the input/output interface 1600.

The media interface 1700 reads a program or data stored in the recording medium 1800 and provides it to the CPU 1100 via the RAM 1200. The CPU 1100 loads the program from the recording medium 1800 onto the RAM 1200 via the media interface 1700 and executes the loaded program. The recording medium 1800 is, for example, an optical recording medium such as a DVD (Digital Versatile Disc) or a PD (Phase change rewritable Disc), a magneto-optical recording medium such as an MO (Magneto-Optical disk), a tape medium, a magnetic recording medium, or a semiconductor memory.

For example, when the computer 1000 functions as the information processing device 100 according to the embodiment, the CPU 1100 of the computer 1000 executes programs loaded onto the RAM 1200 to realize the functions of the control unit 130. The CPU 1100 of the computer 1000 reads and executes these programs from the recording medium 1800, but as another example, the CPU 1100 may obtain these programs from another device via the network N.

Although some of the embodiments and variations of the present application have been described in detail above with reference to the drawings, these are merely examples, and the present invention can be implemented in other forms that incorporate various modifications and improvements based on the knowledge of those skilled in the art, including the aspects described in the lines of the disclosure of the invention.

[6. Other]
Furthermore, among the processes described in the above embodiments and modifications, all or part of the processes described as being performed automatically can be performed manually, or all or part of the processes described as being performed manually can be performed automatically by a known method. In addition, the information including the processing procedures, specific names, various data and parameters shown in the above documents and drawings can be changed arbitrarily unless otherwise specified. For example, the various information shown in each drawing is not limited to the illustrated information.

In addition, each component of each device shown in the figure is a functional concept, and does not necessarily have to be physically configured as shown in the figure. In other words, the specific form of distribution and integration of each device is not limited to that shown in the figure, and all or part of them can be functionally or physically distributed and integrated in any unit depending on various loads, usage conditions, etc.

The above-described embodiments and variations can be combined as appropriate to the extent that they do not cause inconsistencies in the processing content.

The above-mentioned "section, module, unit" can be read as "means" or "circuit." For example, an acquisition unit can be read as an acquisition means or an acquisition circuit.

REFERENCE SIGNS LIST 1 Information processing system 100 Information processing device 121 Model information storage unit 122 Search query information storage unit 123 Advertisement information storage unit 130 Control unit 131 Acquisition unit 132 Learning unit 133 Extraction unit 134 Determination unit 135 Provision unit 10 Terminal device 20 User terminal 50 Information provision device N Network

Claims (20)

an extraction unit that extracts, from a group of search queries used in a search, a search query that satisfies a predetermined condition as a first keyword;
a determination unit that determines a proposed category, which is a category to be proposed as a destination of advertisement submission, based on the first keyword extracted by the extraction unit and second keywords corresponding to each category related to the destination of advertisement submission;
An information processing device comprising: The determination unit is
The information processing device according to claim 1, characterized in that the proposed category of the first keyword is determined based on first vector information in which the first keyword is converted into a vector and second vector information of each of the categories based on the vector conversion of the second keyword corresponding to each of the categories. The determination unit is
The information processing device according to claim 2, characterized in that the proposed category of the first keyword is determined based on a comparison between the first vector information of the first keyword generated using a vector conversion model that takes a character string as input and outputs a vector corresponding to the character string, and the second vector information of each of the categories generated using the vector conversion model. The determination unit is
The information processing apparatus according to claim 2 , further comprising: determining the proposed category for the first keyword based on a comparison of the first vector information for the first keyword with the second vector information for each of the categories. The determination unit is
The information processing apparatus according to claim 2 , further comprising: determining the proposed category of the first keyword based on a similarity between the first vector information of the first keyword and the second vector information of each of the categories. The determination unit is
The information processing apparatus according to claim 5 , further comprising: determining the proposed category of the first keyword based on the similarity, which is a cosine similarity between the first vector information of the first keyword and the second vector information of each of the categories. The determination unit is
The information processing apparatus according to claim 4 , wherein, of the categories, a similar category in which the degree of similarity of the second vector information to the first vector information of the first keyword is maximum is determined as the proposed category. The determination unit is
The information processing device according to claim 2 , further comprising: determining a predicted value of the advertising index of the first keyword using a prediction model that receives a vector as input and outputs a predicted value of an advertising index related to a search advertisement of a keyword corresponding to the vector. The determination unit is
The information processing device according to claim 8 , further comprising: determining a predicted value of the advertising index regarding the conversion resulting from a search using the first keyword, using a prediction model that outputs a predicted value of the advertising index regarding the conversion. The determination unit is
The information processing device according to claim 9 , further comprising: determining a predicted value of the advertising index relating to the number of times a search using the first keyword has led to the conversion, using a prediction model that outputs a predicted value of the advertising index relating to the number of times a search has led to the conversion. The determination unit is
The information processing apparatus according to claim 10 , further comprising: determining a predicted value of a conversion rate resulting from a search using the first keyword, using a prediction model that outputs a predicted value of a conversion rate. The extraction unit is
The information processing apparatus according to claim 1 , further comprising: extracting, from the group of search queries, a search query in which a user who performed a search performed a predetermined action as the first keyword. The extraction unit is
The information processing apparatus according to claim 12 , wherein a search query for which the user accessed a predetermined site is extracted as the first keyword from the group of search queries. The determination unit is
The information processing apparatus according to claim 1 , further comprising: determining the proposed category of the first keyword based on the first keyword and the second keyword corresponding to each of the categories, which are a campaign or an advertisement group. a providing unit that provides the first keyword as a suggested keyword to a terminal device used by a target advertiser who is an advertiser to which information is to be provided;
The information processing apparatus according to claim 1 , further comprising: The providing unit is
The information processing apparatus according to claim 15 , further comprising: providing the proposed category of the first keyword determined by the determination unit to the terminal device together with the proposed keyword. an acquisition unit that acquires the group of search queries;
Further equipped with
The extraction unit is
The information processing apparatus according to claim 1 , further comprising: extracting the first keyword from the group of search queries acquired by the acquisition unit. The acquisition unit is
acquiring information indicating the second keywords corresponding to each of the categories;
The determination unit is
The information processing apparatus according to claim 17 , further comprising: determining the proposed category of the first keyword based on the first keyword and the second keyword acquired by the acquisition unit. 1. A computer-implemented information processing method, comprising:
an extraction step of extracting search queries that satisfy a predetermined condition as first keywords from a group of search queries used in the search;
a determination step of determining a proposed category, which is a category to be proposed as a destination of advertisement submission, based on the first keyword extracted by the extraction step and second keywords corresponding to each category related to the destination of advertisement submission;
13. An information processing method comprising: an extraction step of extracting a search query that satisfies a predetermined condition as a first keyword from a group of search queries used in the search;
a determination step of determining a proposed category, which is a category to be proposed as a destination of advertisement submission, based on the first keyword extracted by the extraction step and second keywords corresponding to each category related to the destination of advertisement submission;
An information processing program characterized by causing a computer to execute the above.

Images