JP7836283B2 - Information processing device, information processing method, and information processing program - Google Patents

Description

Embodiments of the present invention relate to an information processing apparatus, an information processing method, and an information processing program.

In recent years, artificial intelligence (AI) has been utilized in various fields, and the use of generative AI, which can automatically generate content, is rapidly expanding. Some generative AIs use large-scale learning models that have learned a vast amount of natural language and programming languages to perform tasks such as generating and summarizing text in response to user instructions with high accuracy. By using generative AI, users can accurately obtain the information they need when searching for information on the internet.

Conventionally, in AI-based dialogue systems that automatically respond to user questions, systems have been disclosed that efficiently learn question-and-answer patterns by calculating the similarity of questions from multiple users and using similar questions for training, thereby improving the accuracy of responses to user questions.

Japanese Patent Publication No. 2019-219737

AI-powered dialogue systems offer high convenience because they can automatically generate text and other responses based on user instructions and allow users to select from past instructions in their history. However, when users perform searches or other actions, even though their daily routines often involve repetition, they need to input information into the AI each time to request a response, which can be cumbersome.

This invention was made in consideration of these circumstances, and aims to provide an information processing device, an information processing method, and an information processing program that can provide appropriate action suggestions in line with the user's behavioral tendencies.

An information processing apparatus according to an embodiment of the present invention comprises: an access log storage unit that stores input information including information on access to an external network by a user terminal; a log analysis unit that classifies the input information according to categories and extracts categories that satisfy predetermined conditions as the user's behavioral tendencies; and an action suggestion unit that makes action suggestions to the user terminal based on the input information corresponding to the extracted behavioral tendency categories, wherein the access log storage unit stores location information of the user terminal when the user terminal accesses an external network, and the action suggestion unit sets the location of the user terminal for which action suggestions are made based on the location information, and when the user changes the proposed action suggestion, it inputs information including the link destination received from the user to a generating AI server and makes a new action suggestion based on a summary obtained from the generating AI server .

Embodiments of the present invention provide an information processing device, an information processing method, and an information processing program capable of providing appropriate action suggestions in line with the user's behavioral tendencies.

A configuration diagram showing an example of the configuration of the information processing device according to this embodiment. An explanatory diagram showing an example of input information, including user access information, that is stored in the access log storage section. An explanatory diagram showing an example of user behavior trends extracted based on input information. An explanatory diagram showing an example of action suggestions that are proposed to the user based on analyzed behavioral trends. An explanatory diagram showing an example of a screen used when modifying an action proposal submitted by the Action Proposal Department. A flowchart showing an example of the information processing method according to this embodiment.

Hereinafter, embodiments of the present invention will be described based on the attached drawings.
Figure 1 shows an example of the configuration of the information processing device 10 according to this embodiment.

The information processing device 10 according to this embodiment extracts behavioral tendencies that are considered to be of high interest to the user from input information such as access information to an external network by the user of the user terminal 20, and based on these behavioral tendencies, it suggests actions such as access destinations to the user.

The user terminal 20 refers to a device that each user owns, such as a PC, smartphone, or tablet, that is capable of communicating with an external network and can communicate with the information processing device 10 via the network. Note that while Figure 1 shows three user terminals 20 (20a, 20b, 20c) for each user, the configuration is not limited to this. The information processing device 10 is connected to each of the user terminals 20 for communication, and various types of data are sent and received.

The AI generation server 100 is a server that automatically generates responses to user input, equipped with a learning model that has learned information and programming languages on the network. It is connected to the information processing device 10 via a network for communication, and automatically generates responses based on the learning model according to the conditions specified by the input, in response to any input from the user or the information processing device 10.

The specific configuration of the information processing device 10 according to this embodiment will be described below.
The information processing device 10 includes an input information receiving unit 11, an access log storage unit 12, a log analysis unit 13, an action suggestion unit 14, and a suggestion generation unit 15.

Furthermore, the functions of each unit constituting the information processing device 10 may be realized by executing predetermined program code using a processor. However, this is not limited to such software processing; for example, it may also be realized by hardware processing using an ASIC, or by a combination of software and hardware processing.

The input information receiving unit 11 receives user input information from the user terminal 20. Input information refers to information related to user operations on the user terminal 20, including search terms used for internet access using the user terminal 20's web browser, URL information of the accessed website, and applications running on the user terminal. Access time and location information of the user terminal 20 at the time of access may also be received.

The access log storage unit 12 stores input information, including access information by the user of the user terminal 20 to an external network. When the user of the user terminal 20 accesses an external network, the access log storage unit 12 continuously stores access information (search words related to the access, accessed URL), applications being used on the terminal, etc., as input information. It also stores the access time and the location information of the user terminal 20 at the time of external access. The access log storage unit 12 may also arrange the input information chronologically and store it as daily data. Input information is collected and recorded for each user as it occurs.

Furthermore, the access log storage unit 12 uses a database pre-registered with website categories corresponding to each URL to set and save the category corresponding to the accessed URL received from the input information receiving unit 11. A category indicates a type according to the content of the website, meaning classification information according to content such as sports, history, or education. If a category is not registered for the user's accessed URL, the system may query the generation AI server 100 for the category (website type) of that accessed URL, obtain the category, and save it.

Furthermore, the access log storage unit 12 may delete access to search site URLs from the input information, as this is considered unrelated to the user's behavioral trends.

Figure 2 is an explanatory diagram showing an example of input information, including user access information, that is stored in the access log storage unit 12.

As shown in Figure 2, the user's input information for August 15, 2023, for user Yamada Taro, includes the search term used during external access, the URL of the accessed site, and the application used on the terminal, all of which are saved sequentially. The location information of the user's terminal 20 is also recorded along with the time of access. The accessed URL is compared with a pre-registered database of website categories, and the category of the accessed URL is saved.

The log analysis unit 13 classifies the input information according to categories and extracts categories that meet predetermined conditions as user behavior trends. Specifically, it classifies each user access information stored in the access log storage unit 12 according to the set categories and determines the frequency of access for each category. Then, it extracts categories whose access frequency meets predetermined conditions (for example, 5 or more accesses) as behavior trends. Behavioral trends refer to categories that are considered to be of high user interest. The predetermined conditions could, for example, be a condition such as arranging categories in descending order of access frequency and extracting the top four categories.

Figure 3 is an explanatory diagram illustrating an example of user behavioral trends extracted based on input information. As shown in Figure 3, user Yamada Taro's access information is classified by category, and the frequency of access for each category is determined. Then, the categories are sorted in descending order of frequency, and four categories are extracted as behavioral trends. The classification results of Yamada Taro's access information categories show a tendency towards high interest in the fields of sports, history, education (university), and business; these four categories represent his behavioral trends.

The action suggestion unit 14 provides action suggestions to the user terminal 20 based on the input information corresponding to the extracted action trend categories. Specifically, for each category extracted by the log analysis unit 13, it sends links to the access URLs of the input information as action suggestions to the user terminal 20. The user of the user terminal 20 can then view the action suggestions displayed on the terminal screen. For example, the action suggestions could be based on the user's access logs for one week to determine their action trends, and then suggested when the user launches their browser.

Furthermore, the action suggestion unit 14 may check the access times for the extracted action trend categories (if there is variation in access times, the time of day with the most access) and set a time (for example, after 7 PM) for suggesting actions based on the access times.

Furthermore, for the extracted behavioral trend categories, the access locations of the user terminal 20 may be checked (if there is variation in access locations, the range of locations with the most frequent access) and the location of the terminal that will suggest actions based on the access location (for example, within a 10m radius of the user's home location information) may be set.

The suggestion generation unit 15 inputs input information corresponding to the extracted behavioral tendency categories into the generation AI server 100 under predetermined conditions and obtains the output results from the learning model. The behavior suggestion unit 14 may use the information obtained by the suggestion generation unit 15 to make behavior suggestions. For example, regarding a link to an accessed URL, the generation AI server 100 is input to request a summary of the content on the site, the summary results from the learning model are obtained, and this summary is used as a behavior suggestion.

Figure 4 is an explanatory diagram illustrating an example of behavioral suggestions provided to the user based on analyzed behavioral trends. Based on Taro Yamada's access information, behavioral suggestions are displayed on the user's terminal screen for categories extracted. For the "Sports" category, information on the accessed URL and a summary of that URL are displayed. For the "History" category, information on the accessed URL and a summary of the search terms are displayed. The behavioral suggestion display screen includes buttons for "Change Suggestion" to modify the suggestion content and "Delete" to remove the suggestion. If a suggestion is deleted, it may be possible to configure the system to prevent suggestions for that category from being provided again in the future.

Figure 5 is an explanatory diagram showing an example of a screen when modifying an action suggestion proposed by the action suggestion unit 14. If you wish to change the proposed content, you can specify the link destination and other details you would like to be included in future suggestions. When a suggestion change is set, the AI server 100 receives input requesting a summary of the content on the entered access URL link. The AI server obtains the summary result from the learning model and modifies the action suggestion to use this summary.

In this way, the user of user terminal 20 can automatically obtain appropriate information based on their own daily behavioral patterns.

Next, the operation of the information processing device 10 according to this embodiment will be described. Figure 6 is a flowchart showing an example of the information processing method according to this embodiment (refer to Figure 1 as appropriate).

The access log storage unit 12 stores input information, including information about the user's access to the external network from the user terminal 20 (S10).

The log analysis unit 13 classifies the input information according to category and extracts categories that meet predetermined conditions as user behavior trends (S11, S12).

The log analysis unit 13 extracts categories that meet predetermined conditions as behavioral trends (S12).

The action suggestion unit 14 makes action suggestions to the user terminal 20 based on the input information corresponding to the extracted behavioral tendency categories (S13).

According to the information processing device of the embodiment described above, it is possible to extract user behavioral trends from daily input information, including user access to external networks from the user terminal, and to suggest actions such as access destinations to the user based on these behavioral trends, thereby providing appropriate action suggestions that align with the user's behavioral trends.

The program executed by the information processing device 10 is provided pre-installed in a storage circuit such as ROM. Alternatively, this program may be provided as an installable or executable file stored on a computer-readable storage medium such as a CD-ROM, CD-R, memory card, DVD, or flexible disk. Furthermore, the program executed by the information processing device 10 may be stored on a computer connected to a network such as the Internet and provided via network download.

While several embodiments of the present invention have been described, these embodiments are presented as examples only and are not intended to limit the scope of the invention. These novel embodiments can be implemented in various other forms, and various omissions, substitutions, and modifications are possible without departing from the spirit of the invention. These embodiments and their variations are included within the scope and spirit of the invention, as well as within the scope of the invention and its equivalents as described in the claims.

10…Information processing device, 11…Input information receiving unit, 12…Access log storage unit, 13…Log analysis unit, 14…Action suggestion unit, 15…Suggestion generation unit, 20…User terminal, 100…Generating AI server.

Claims (4)

An access log storage unit that stores input information including information on the user's access to an external network on the user terminal,
A log analysis unit classifies the input information according to categories and extracts categories that meet predetermined conditions as the user's behavioral tendencies.
The system includes an action suggestion unit that makes action suggestions to the user terminal based on the input information corresponding to the extracted categories of behavioral tendencies,
The access log storage unit stores the location information of the user terminal when the user terminal accesses an external network.
The action suggestion unit sets the location of the user terminal that makes the action suggestion based on the location information ,
When the user modifies the proposed action, the AI server generates information including the link received from the user, and based on the summary obtained from the AI server, it generates a new action proposal.
An information processing device characterized by the following: A proposal generation unit inputs the input information corresponding to the extracted behavioral tendency categories into a learning model that automatically generates responses according to conditions specified by the user, and makes behavioral suggestions based on the output results of the learning model.
The information processing apparatus according to claim 1, further comprising: A step of saving input information, including information about the user's access to an external network on the user terminal,
The steps include classifying the input information according to categories and extracting the categories that satisfy predetermined conditions as the user's behavioral tendencies,
A step of making an action suggestion to the user terminal based on the input information corresponding to the extracted category of behavioral tendencies,
A step of saving the location information of the user terminal when the user terminal accesses an external network,
The steps include setting the location of the user terminal that will make action suggestions based on the aforementioned location information,
When the user modifies the proposed action, the process involves inputting information including the link received from the user into the generating AI server, and generating a new action based on the summary obtained from the generating AI server.
An information processing method characterized by including On the computer,
A function that saves input information, including information about the user's access to external networks on the user terminal.
A function that classifies the input information according to categories and extracts the categories that meet predetermined conditions as the user's behavioral tendencies.
A function that provides action suggestions to the user terminal based on the input information corresponding to the extracted categories of behavioral tendencies.
A function that saves the location information of the user terminal when the user terminal accesses an external network.
A function to set the location of the user terminal that makes action suggestions based on the aforementioned location information.
When the user modifies the proposed action, the AI server receives information including the link destination from the user and inputs it into the AI server to generate a new action proposal based on the summary obtained from the AI server.
An information processing program characterized by achieving this.