JP7286133B2 - Computer program, information processing device, and trademark determination method - Google Patents

Description

The present invention relates to a computer program, an information processing apparatus, and a trademark determination method for determining candidate trademarks included in content.

For example, content such as proposals, press releases, or meeting materials created within a company may include characters, logos, or the like that are candidates for trademarks. If the provision of products or services based on such content is started and the trademark attached to the product or service is already registered trademark of another party, it will infringe the rights of the other party. It is important to conduct a trademark search in advance.

Patent Literature 1 proposes a trademark server device that extracts from a database registered trademarks that are judged to be dissimilar by one syllable. This trademark server device generates extraction pronounciation data in which one sound in the pronounciation data is replaced with a dummy, records it in association with the registered trademark, extracts a set of trademarks having the same extraction pronounciation data, and The set of different one-sounds for the trademark is extracted and recorded in the database. The trademark server device extracts from the database a set of trademarks having the same one as the set of different one-sounds in the extraction conditions from the user, and transmits them to the user terminal device.

JP-A-2005-339374

However, trademark research is difficult for non-experts.

SUMMARY OF THE INVENTION The present invention has been made in view of such circumstances, and an object thereof is to provide a computer program, an information processing apparatus, and a trademark determination method for determining candidate trademarks included in content. .

A computer program according to the present invention causes a computer to extract candidate trademarks included in input content, refer to a database in which trademarks are stored, and match or match the extracted candidate trademarks with trademarks stored in the database. A determination is made as to whether or not the candidate trademark is similar to the trademark stored in the database.

Further, the computer program according to the present invention stores registered trademarks in the database, and performs a process of notifying the computer when it is determined that the candidate trademark matches or is similar to the registered trademark stored in the database. let it run.

Further, the computer program according to the present invention, wherein the database stores trademarks at the application stage, notifies the computer when the trademark candidate is determined to match or resemble the trademark at the application stage stored in the database. Let the process run.

Further, the computer program according to the present invention performs a process of notifying the computer of the trademark candidate determined to match or resemble the registered trademark and the trademark candidate determined to match or resemble the application-stage trademark. let it run.

Further, the computer program according to the present invention causes the computer to execute processing of notifying a substitute trademark candidate when it is determined that the trademark candidate matches or resembles a trademark stored in the database.

Further, the computer program according to the present invention causes the computer to execute a process of displaying an advertisement regarding an agent who acts as an agent for trademark application procedures when notifying the alternative trademark candidate.

Further, in the computer program according to the present invention, the candidate trademarks include character or logo trademarks.

A computer program according to the present invention causes a computer to execute a process of determining classification of extracted trademark candidates.

Further, the computer program according to the present invention uses a trained model that accepts content as an input and outputs a trademark classification, and determines the classification of the candidate trademark.

Further, the information processing apparatus according to the present invention includes an extraction unit for extracting trademark candidates included in input content, and referring to a database in which trademarks are stored, and storing the trademark candidates extracted by the extraction unit in the database. a determination unit that determines whether or not the candidate trademark matches or is similar to the trademark stored in the database; Prepare.

In addition, the trademark determination method according to the present invention extracts trademark candidates included in input content, refers to a database in which trademarks are stored, and extracts trademark candidates that match or match trademarks stored in the database. A decision is made as to whether or not the candidate trademark is similar, and if it is decided that the candidate trademark matches or is similar to the trademark stored in the database, a notification is given.

In the present invention, a computer extracts candidate trademarks included in the input content, refers to a database in which trademarks are stored, and determines whether the candidate trademarks match or are similar to the trademarks stored in the database. do. If it determines that the candidate trademark extracted from the content matches or is similar to the trademark in the database, the computer will notify you accordingly. Thereby, the user can grasp the trademark candidate included in the created content and the determination result of whether or not the trademark candidate matches or is similar to an existing trademark.

In addition, in the present invention, the database stores registered trademarks, and when the computer determines that the trademark candidates extracted match or are similar to the registered trademarks, the information is notified. This allows the user to know that the content includes a trademark candidate that may infringe on another person's registered trademark.

Further, in the present invention, the database stores the trademarks at the application stage, and when the trademark candidate extracted by the computer is determined to match or resemble the trademark at the application stage, notification is made. This allows the user to understand that the content includes trademark candidates that match or are similar to trademarks that may be registered in the future.

In addition, in the present invention, a trademark candidate judged to match or resemble a registered trademark and a trademark candidate judged to match or resemble a trademark at the application stage are separately notified. This allows the user to easily determine whether the candidate trademark included in the content matches or resembles the registered trademark or the trademark in the application stage.

Also in accordance with the present invention, if the computer determines that the candidate trademark matches or is similar to a trademark stored in the database, the computer notifies the candidate of the trademark. Alternative trademark candidates are trademark candidates that do not match or resemble trademarks stored in the database. This allows the computer to suggest another trademark candidate to a user who has created content that includes a trademark candidate that matches or is similar to an existing trademark.

Computers can generate alternative trademark candidates by using search term prediction functions, text analysis and prediction learning models of Internet search engines. This allows the computer to easily generate alternate trademark candidates.

In addition, in the present invention, when notifying alternative trademark candidates, the computer displays an advertisement regarding an agent who acts on behalf of the trademark filing procedure. As a result, the user can easily request the agent to carry out application procedures, investigations, or the like for alternative trademark candidates.

Further, in the present invention, trademark candidates to be extracted from contents include character or logo trademarks. This allows the user to determine whether the content contains trademark candidates that match or are similar to the character or logo trademark that can be registered as a trademark.

Further, in the present invention, the computer determines the classification of trademark candidates extracted from the content. As a result, it is possible to entrust the judgment of trademark classification, which is difficult for inexperienced users, to the computer.

Further, in the present invention, a trained model that accepts content as an input and outputs a trademark classification is used to determine the trademark candidate classification. As a result, it is possible to accurately determine the classification of trademark candidates included in the content.

According to the present invention, the user can easily determine whether or not the content contains candidate trademarks that match or are similar to the trademarks stored in the database.

1 is a schematic diagram for explaining a configuration example of a trademark judgment system according to an embodiment; FIG. 1 is a block diagram showing the configuration of a trademark determination device according to this embodiment; FIG. 2 is a block diagram showing the configuration of a learning server according to this embodiment; FIG. FIG. 4 is a schematic diagram for explaining an example of trademark determination processing performed by the trademark determination device according to the present embodiment; FIG. 4 is a schematic diagram for explaining an example of trademark determination processing performed by the trademark determination device according to the present embodiment; 4 is a flow chart showing a procedure of trademark determination processing performed by the trademark determination device according to the present embodiment;

A specific example of the trademark determination method according to the embodiment of the present invention will be described below with reference to the drawings. The present invention is not limited to these examples, but is indicated by the scope of the claims, and is intended to include all modifications within the scope and meaning equivalent to the scope of the claims.

<System Overview>
FIG. 1 is a schematic diagram for explaining a configuration example of a trademark determination system according to this embodiment. The trademark determination system according to this embodiment includes a trademark determination device 1, a learning server 3, a trademark management server 5, and the like. The trademark determination device 1 is a device implemented by installing a computer program for performing trademark determination in an information processing device such as a personal computer or a server computer. In the present embodiment, the following description will be made on the assumption that the trademark determination apparatus 1 is implemented using a personal computer operated by a user. The trademark determination device 1 can communicate with the learning server 3, the trademark management server 5, and the like via a network such as the Internet.

The learning server 3 is a server computer that learns the trademark classification determination model 100 used in the trademark determination system according to the present embodiment. The learning server 3 periodically learns the trademark classification determination model 100 and transmits the trained trademark classification determination model 100 to the trademark determination apparatus 1 . The trademark determination device 1 receives the trademark classification determination model 100 transmitted from the learning server 3, stores it in its own storage unit or the like, and uses the stored trademark classification determination model 100 to perform processing related to trademark determination. Note that the trademark determination device 1 may be configured to learn the trademark classification determination model 100, and in this case, the learning server 3 need not be provided separately.

The trademark management server 5 is a server computer that has a trademark DB (database) 5a that stores trademarks such as registered trademarks and trademarks in the application stage, and that manages trademarks. The trademark management server 5 can be, for example, a server computer that operates and manages the patent information platform of the Patent Office. However, for example, the trademark determination device 1 may be configured to include the trademark DB 5a, or, for example, a unique trademark management server 5 including the trademark DB 5a for this trademark determination system may be provided separately from the one of the Patent Office. good. The trademark management server 5 provides, for example, a pronounciation search service. Information on trademarks in the application stage is output as search results.

A user who uses the trademark determination system according to the present embodiment can, for example, use a personal computer that constitutes the trademark determination apparatus 1 or a personal computer separate from the trademark determination apparatus 1 to process document files such as proposals or meeting materials. to create A document file may contain information such as strings and images. By inputting the created document file into the trademark determination program of the trademark determination apparatus 1, the user can determine whether or not the candidates for the trademark contained in this document file match or are similar to the registered trademark or the trademark at the application stage. It can be carried out.

Upon receiving the input of the document file, the trademark determination device 1 extracts trademark candidates included in the input document file. The trademark determination device 1 also inputs this document file to the trademark classification determination model 100 that has been learned in advance. Based on the input document file, the trademark classification determination model 100 determines the classification of trademark candidates contained in this document file, and outputs the classification of the trademark candidates as a determination result. The trademark determination device 1 determines whether the trademark candidate extracted from the document file matches or resembles the trademark stored in the trademark DB 5 a of the trademark management server 5 . The trademark determination device 1 transmits trademark candidate character strings to the trademark management server 5 and receives trademark search results from the trademark management server 5 . At this time, the trademark determination device 1 may transmit information on the classification determined by the trademark classification determination model 100 to the trademark management server 5, and may receive trademark search results in this classification. Based on the search results obtained from the trademark management server 5, the trademark determination device 1 can determine whether the trademark candidates match or are similar to the registered trademarks or trademarks in the application stage stored in the trademark DB 5a. If the trademark candidate contained in the document file is determined to match or resemble the registered trademark or the trademark in the application stage, the trademark determination device 1 outputs an alert notifying that fact to call the user's attention.

Further, when the trademark determination device 1 determines that a trademark candidate matches or resembles a registered trademark or an application-stage trademark, it generates a substitute trademark candidate to replace this trademark candidate. Internet search engines are equipped with a function of predicting terms following an input search term. there is The trademark determination device 1 uses part or all of the character strings forming the trademark candidates as search terms, and predicts terms that follow the search terms. The trademark determination device 1 generates an alternative trademark candidate by appropriately combining a trademark candidate character string extracted from a document file and a character string obtained as a result of prediction of terms following this character string, and outputs an alert. along with alternative trademark candidates.

Further, when the trademark determination apparatus 1 according to the present embodiment outputs the alternative trademark candidate together with the alert output, it also outputs the advertisement regarding the agent who represents the trademark application. At this time, the trademark determination device 1 appropriately acquires an advertisement regarding the agent from an advertisement distribution server (not shown) and displays the advertisement in a corner of the alert display screen or the like.

As a result, the trademark determination apparatus 1 according to the present embodiment extracts trademark candidates included in a document file such as a proposal or meeting materials created by the user, and the extracted trademark candidates are registered trademarks or registered trademarks stored in the trademark DB 5a. The user can be notified when it is determined that the trademark matches or is similar to the trademark at the application stage. This allows the user to know that the document file contains trademark candidates that may infringe on the trademarks of others.

<Device configuration>
FIG. 2 is a block diagram showing the configuration of the trademark determination device 1 according to this embodiment. The trademark determination device 1 according to the present embodiment is configured including a processing unit (processor) 11, a storage unit (storage) 12, a communication unit (transceiver) 13, an operation unit 14, a display unit (display) 15, and the like. there is The processing unit 11 is configured using an arithmetic processing unit such as a CPU (Central Processing Unit) or an MPU (Micro-Processing Unit), and reads out and executes a trademark determination program 12a stored in the storage unit 12. , various arithmetic processing and control processing. In the present embodiment, the processing unit 11 performs a process of extracting trademark candidates from a document file, a process of determining whether or not the trademark candidate matches or resembles a registered trademark or an application-stage trademark, and a process of notifying the determination results. etc.

The storage unit 12 is configured using, for example, a large-capacity storage device such as a hard disk drive. The storage unit 12 stores a trademark determination program 12a executed by the processing unit 11, data necessary for executing this program, and the like. For example, the storage unit 12 is provided with a term DB (database) 12b in which a large number of terms used for prediction of search terms are accumulated. Further, in the present embodiment, the storage unit 12 stores a trained model of the trademark classification determination model 100 . The trademark classification determination model 100 is a trained model that has been trained in the learning server 3 in advance. The trained model performs a predetermined calculation on the input value and outputs the calculation result. In the storage unit 12, data such as coefficients and threshold values of functions that define this calculation are stored as the trademark classification determination model 100. remembered. The processing unit 11 that executes the trademark determination program 12a reads the data stored as the trademark classification determination model 100, so that the processing unit 11 can execute calculations for determining the classification of trademark candidates from the document file. becomes.

The trademark determination program 12a, the term DB 12b, and the trademark classification determination model 100 may be written in the storage unit 12, for example, at the manufacturing stage of the trademark determination device 1, or may be distributed by another device via a network, for example, for trademark determination. The apparatus 1 may acquire the information, or the trademark determination apparatus 1 may read out the information recorded in the recording medium 99 such as a memory card or an optical disk and store the information in the storage unit 12 . The trademark determination program 12a, the term DB 12b, and the trademark classification determination model 100 may be provided in the form of distribution via a network, or may be provided in the form of being recorded on the recording medium 99. FIG. Also, the trademark determination program 12a, the term DB 12b, and the trademark classification determination model 100 may be communicated periodically.

The communication unit 13 communicates with other devices such as the learning server 3 and the trademark management server 5 via a network NW including a LAN (Local Area Network), the Internet, a mobile phone communication network, and the like. The communication unit 13 provides the processing unit 11 with information received from another device via the network NW, and transmits information for transmission provided from the processing unit 11 to the other device.

The operation unit 14 is configured using input devices such as a keyboard and a mouse. The operation unit 14 gives the processing unit 11 a signal or information according to the operation performed by the user. The display unit 15 is configured using, for example, a liquid crystal panel or the like. The display unit 15 displays the image given from the processing unit 11 . In the present embodiment, the display unit 15 is used to display the determination results regarding trademark candidates included in the document file.

By executing the trademark determination program 12a stored in the storage unit 12, the processing unit 11 includes a trademark candidate extraction unit 11a, a classification determination unit 11b, a trademark determination unit 11c, a substitute trademark candidate generation unit 11d, and a notification process. The unit 11e, the advertisement display processing unit 11f, and the like are implemented as software functional blocks. The trademark candidate extraction unit 11a receives an input of a document file from a user and performs a process of extracting a trademark candidate included in the input content, that is, a character string or the like that can be a trademark. For example, the trademark candidate extraction unit 11a extracts nouns and the like contained in the sentences by performing processing such as lexical analysis and syntactic analysis on the sentences contained in the document file, and extracts common nouns from among the extracted nouns. Anything other than nouns can be used as a trademark candidate. Whether or not the extracted nouns and the like are common can be determined by storing a database or the like in which common nouns are registered in the trademark determination device 1 together with the trademark determination program 12a. Further, for example, the trademark candidate extraction unit 11a can extract nouns enclosed in parentheses, square brackets, quotation marks, or the like in sentences included in the document file as trademark candidates.

The classification determination unit 11b uses the trademark classification determination model 100 stored in the storage unit 12 to perform processing for determining the classification of trademark candidates included in the document file. The trademark classification determination model 100 is a trained model that receives sentences contained in a document file as input and outputs one or more classifications suitable for trademark candidates contained in this document file. The trademark classification determination model 100 may adopt a deep learning model such as RNN (Recurrent Neural Network), LSTM (Long Short Term Memory) network, or Seq2Seq (Sequence to Sequence). The trademark classification determination model 100 outputs class 25, which is the classification of clothing and footwear, for example, when sentences relating to clothes or fashion are input. The classification determination unit 11b inputs sentences included in the input document file to the trademark classification determination model 100, and acquires classification determination results of trademark candidates included in the sentences as outputs of the trademark classification determination model 100. The text input to the trademark classification determination model 100 does not have to be the entire text of the document file. It can be extracted from a file.

The trademark determination unit 11c determines whether or not the trademark candidate extracted from the document file by the trademark candidate extraction unit 11a matches or resembles a registered trademark or an application stage trademark. The trademark determination unit 11c searches for trademark candidates by transmitting the trademark candidates extracted by the trademark candidate extraction unit 11a and the classifications determined by the classification determination unit 11b to the trademark management server 5 through the communication unit 13 . In the present embodiment, the trademark determination unit 11c transmits the pronounciation of the trademark candidate to the trademark management server 5 and requests the trademark management server 5 to search for the pronounciation. The trademark management server 5 searches for registered trademarks or trademarks in the application stage stored in the trademark DB 5a based on the trademark candidates and classifications sent from the trademark determination apparatus 1. FIG. If there are registered trademarks or trademarks at the application stage that match or are similar to the candidate trademark, the trademark management server 5 transmits a list of these registered trademarks or trademarks at the application stage as search results to the trademark determination apparatus 1 . The information transmitted as search results from the trademark management server 5 to the trademark determination apparatus 1 includes, for example, the trademark found by the search, pronunciation, classification, distinction between registered trademarks and trademarks in the application stage, and matching or similar determination results. and may also include information such as the applicant for this trademark, filing date, registration date, application number or registration number. If there is no registered trademark or application-stage trademark that matches or is similar to the trademark candidate, the trademark management server 5 notifies the trademark determination device 1 that there is no registered trademark or application-stage trademark as a search result. The trademark determination unit 11c receives search results from the trademark management server 5, and determines whether or not there is a registered trademark or an application-stage trademark that matches or is similar to the trademark candidate based on the received search results. If the search results received from the trademark management server 5 include, for example, a trademark for which a decision of refusal has been finalized or a trademark that has not survived, the trademark determination unit 11c may perform processing for excluding these.

The substitute trademark candidate generation unit 11d performs processing for generating substitute trademark candidates that can substitute for the trademark candidate when the trademark determination unit 11c determines that there is a registered trademark or application-stage trademark that matches or is similar to the trademark candidate. In this embodiment, the term DB 12b provided in the storage unit 12 of the trademark determination device 1 stores information about search terms input by an unspecified number of users to search engines on the Internet. The alternative trademark candidate generation unit 11d uses the term DB 12 of the storage unit 12 and predicts another term following a certain term by performing the same processing as a search term prediction function of a search engine. For example, the term DB 12b stores a combination of a term and another term following this term. make a prediction of The substitute trademark candidate generation unit 11d extracts from the document file and extracts a part or all of the character string that constitutes the trademark candidate for which it is determined that there is a matching or similar registered trademark or application-stage trademark. Predict the string of . The substitute trademark candidate generating unit 11d generates one or more substitute trademark candidates by combining the character string obtained as a result of prediction and the character strings that constitute the trademark candidates. Information stored in the terminology DB 12b is periodically updated. The trademark determination device 1 acquires information periodically transmitted from, for example, a server device equipped with a search engine, and updates the information in the term DB 12b.

The notification processing unit 11e notifies the user of the presence of registered trademarks or trademarks in the application stage that match or are similar to trademark candidates included in the document file by displaying the result of determination by the trademark determination unit 11c on the display unit 15. process. Further, the notification processing unit 11e notifies the classification of trademark candidates determined by the classification determination unit 11b or the substitute trademark candidates generated by the substitute trademark candidate generation unit 11d as necessary.

The advertisement display processing unit 11f performs a process of displaying an advertisement regarding an agent acting as an agent for trademark application procedures on the display unit 15 when the notification processing unit 11e notifies a substitute trademark candidate. The advertisement display processing unit 11 f acquires information about advertisements from an advertisement distribution server device (not shown) and displays the information on the display unit 15 . The type of advertisement to be displayed is entrusted to the processing of advertisement distribution by the advertisement distribution server device.

FIG. 3 is a block diagram showing the configuration of the learning server 3 according to this embodiment. The learning server 3 according to this embodiment includes a processing unit (processor) 31, a storage unit (storage) 32, a communication unit (transceiver) 33, and the like. The processing unit 31 is configured using an arithmetic processing device such as a CPU, MPU or (Graphics Processing Unit), and reads and executes a server program 32a stored in the storage unit 32 to perform various arithmetic processing and Performs control processing, etc. In the present embodiment, the processing unit 31 performs a process of learning the trademark classification determination model 100, a process of distributing the trained trademark classification determination model 100 to the trademark determination apparatus 1, and the like.

The storage unit 32 is configured using a large-capacity storage device such as a hard disk drive. The storage unit 32 stores a server program 32a executed by the processing unit 31, data necessary for executing this program, and the like. The server program 32a may be written in the storage unit 32, for example, at the manufacturing stage of the learning server 3, or may be distributed by another device via a network, for example, and may be acquired by the learning server 3. The information recorded on the recording medium 98 such as a card or optical disk may be read by the learning server 3 and stored in the storage unit 32 . The server program 32 a may be provided in the form of distribution via a network, or may be provided in the form of being recorded on the recording medium 98 .

Further, in the present embodiment, the storage unit 32 stores a trademark classification determination model 100 and learning data 101 . The trademark classification determination model 100 is a learning model of deep learning which the learning server 3 makes to learn. The learning model performs a predetermined calculation on an input value and outputs the calculation result, and data such as coefficients and thresholds of functions that define this calculation are stored as a trademark classification determination model 100 in the storage unit 32. be done. In the present embodiment, trademark classification determination model 100 employs, for example, a deep learning model such as RNN, LSTM network, or Seq2Seq. The learning data 101 is data used for learning the trademark classification determination model 100. For example, data of a set of inputs and outputs to the trademark classification determination model 100, that is, a sentence containing a trademark candidate and classification of the trademark candidate. It is a collection of tuple data.

The communication unit 33 communicates with other devices via a network NW including LAN, Internet, mobile phone communication network, and the like. The communication unit 33 communicates with the trademark determination device 1 via a network NW such as the Internet. The communication unit 33 provides the processing unit 31 with information received from another device via the network NW, and transmits information for transmission provided from the processing unit 31 to the other device.

In the processing unit 31, a learning processing unit 31a, a model distribution unit 31b, and the like are implemented as software functional blocks by executing a server program 32a stored in the storage unit 32. FIG. The learning processing unit 31 a performs deep learning processing for the trademark classification determination model 100 using the learning data 101 stored in the storage unit 32 . The trademark classification determination model 100 is, for example, a deep learning model that constitutes a neural network such as RNN. The learning data 101 is, for example, a set of teacher data in which sentences containing trademarks are associated with classifications of the trademarks. The learning processing unit 31a uses the sentences of the learning data 101 as input information for the trademark classification determination model 100, and uses the corresponding trademark classification as output information that the trademark classification determination model 100 should output. The trademark classification determination model 100 is learned by adjusting the coefficients and threshold values of a plurality of neurons that constitute the .

The model distribution unit 31b performs processing for transmitting the trained trademark classification determination model 100, which has undergone deep learning by the learning processing unit 31a, to the trademark determination apparatus 1 through the communication unit 33. FIG. Transmission of the trademark classification determination model 100 by the model distribution unit 31b may be performed, for example, in response to a request from the trademark determination apparatus 1, or may be performed spontaneously, for example, in response to completion of learning by the learning processing unit 31a. good too.

<Trademark judgment processing>
4 and 5 are schematic diagrams for explaining an example of trademark determination processing performed by the trademark determination apparatus 1 according to the present embodiment. The upper part of FIG. 4 shows an example of a document file created by the user. The document file of this example includes information of character strings such as sentences, and the sentences are indicated by horizontal lines in the figure. A user inputs a document file to be subjected to trademark determination processing to the trademark determination apparatus 1 . Upon receiving the input of the document file, the trademark determination device 1 extracts trademark candidates contained in the document file. For example, the trademark determination device 1 extracts nouns and the like contained in the text by performing language processing such as lexical analysis and syntactic analysis on the text included in the document file. The trademark determination device 1 eliminates pre-registered general nouns from the extracted nouns, etc., and uses the remaining nouns, etc. as trademark candidates. In addition, for example, the trademark determination device 1 can detect character strings enclosed in parentheses, square brackets, quotation marks, or the like in sentences included in a document file, or character strings emphasized in a color or font that is different from other portions. It may be extracted as a trademark candidate.

The trademark determination device 1 also uses the trademark classification determination model 100 stored in the storage unit 12 to perform processing for determining the classification of the candidate trademark. The trademark determination apparatus 1 inputs the entire text of the input document file, or part of the entire text including the trademark candidate, to the trademark classification determination model 100, and the trademark classification determination model 100 outputs one or A trademark classification is determined by acquiring a plurality of trademark classifications.

Next, the trademark determination device 1 determines whether or not the trademark candidate extracted from the document file matches or resembles the registered trademark or the trademark in the application stage stored in the trademark DB 5 a of the trademark management server 5 . The trademark determination device 1 transmits information on the trademark candidate extracted from the document file and the classification determined by the trademark classification determination model 100 to the trademark management server 5 to request a trademark search. As a result, the trademark management server 5 searches the trademark DB 5 a for matching or similar registered trademarks or trademarks in the application stage using the trademark candidates and classifications received from the trademark determination device 1 as search conditions, and transmits the search results to the trademark determination device 1 . do. Upon receiving the search results from the trademark management server 5, the trademark determination device 1 determines whether or not there is a registered trademark or application-stage trademark that matches or is similar to the trademark candidate based on the received search results. Based on information included in the received search results, the trademark determination device 1 determines whether the retrieved trademark is a registered trademark or an application-stage trademark, and whether the retrieved trademark matches or is similar to the trademark candidate. It is possible to determine which is which.

The trademark determination device 1 displays an alert indicating the determination result as to whether or not there is a registered trademark or application-stage trademark that matches or is similar to the trademark candidate, according to the determination result based on the search result. An example of an alert display is shown in the lower part of FIG. In this example, three trademark candidates extracted from the document file are notified that they match or are similar to the registered trademark or the trademark in the application stage. In the figure, trademark candidates determined to be identical or similar to the registered trademark or the trademark at the application stage are highlighted by enclosing them with dashed oval circles. Furthermore, in the present embodiment, the trademark candidate determined to match or resemble the registered trademark is highlighted so as to be distinguishable from the trademark candidate determined to match or resemble the application-stage trademark. In the drawing, trademark candidates determined to match or resemble registered trademarks are further highlighted by hatching.

If a document file contains a well-known registered trademark that anyone can generally recognize as a registered trademark, the trademark determination apparatus 1 may exclude this registered trademark from the alert target. The trademark determination apparatus 1 can exclude well-known registered trademarks from alerts by providing a list of famous registered trademarks in advance, for example.

In response to the alert display displayed on the display unit 15, the user can instruct the trademark determination apparatus 1 to display details regarding the trademark candidates. In response to this instruction, the trademark determination device 1 displays detailed information as shown in FIG. 5 for one trademark candidate selected from the document file. In FIG. 5, characters constituting the trademark are represented by hatched figures such as circles, squares, triangles and rhombuses.

In the detailed information display screen of the trademark candidate by the trademark determination device 1, for example, the trademark candidate extracted from the document file is displayed at the top. The trademark determination device 1 displays matching or similar registered trademarks or application-stage trademarks found by searching, for example, below trademark candidates. In the illustrated example, the message "A matching or similar trademark exists" and the character strings of registered trademarks and trademarks in the application stage found by the search are displayed.

Further, the trademark determination apparatus 1 displays alternative trademark candidates to replace the trademark candidate below the display of the matching or similar trademark. In the illustrated example, two trademarks are displayed as alternative trademark candidates. If the trademark determination device 1 determines that the trademark candidate matches or resembles the registered trademark or the trademark in the application stage, it generates a substitute trademark candidate. At this time, the trademark determination device 1 predicts another character string following the trademark candidate character string based on the information stored in the term DB 12 b of the storage unit 12 . The trademark determination apparatus 1 generates a substitute trademark candidate by appropriately combining a character string obtained by appropriately dividing a character string of a trademark candidate and a character string obtained as a prediction result. Note that the prediction result may include a plurality of character strings, and the trademark determination apparatus 1 may generate a plurality of alternative trademark candidates. Further, the trademark determination apparatus 1 may confirm that there is no matching or similar registered trademark or application-stage trademark for the generated substitute trademark candidate by the same procedure as the trademark candidate determination described above.

Further, when the trademark determination apparatus 1 generates and displays a substitute trademark candidate for a trademark candidate, it displays an advertisement regarding an agent who acts as an agent for trademark application procedures. In the illustrated example, two advertisements are displayed at the bottom. The trademark determination device 1 appropriately acquires one or more advertisements relating to the agent from the advertisement distribution server and displays them.

FIG. 6 is a flow chart showing a procedure of trademark determination processing performed by the trademark determination apparatus 1 according to the present embodiment. The processing unit 11 of the trademark determination device 1 according to the present embodiment receives an input of a document file, for example, based on a user's operation on the operation unit 14 (step S1). The trademark candidate extraction unit 11a of the processing unit 11 extracts trademark candidates included in the input document file (step S2). The classification determination unit 11b of the processing unit 11 inputs part or all of the sentences included in the document file received in step S1 to the trademark classification determination model 100 stored in the storage unit 12, and determines the trademark classification determination model. The classification of the trademark candidate is determined by obtaining the classification information output by 100 (step S3).

Next, the trademark determination unit 11c of the processing unit 11 transmits information on the trademark candidates extracted in step S2 and the classification determined in step S3 to the trademark management server 5, thereby identifying the trademark stored in the trademark DB 5a. A request is made to search for trademarks that match or are similar to the candidate trademark (step S4). In response to this request, the trademark management server 5 searches for a trademark, and the search result is transmitted from the trademark management server 5 to the trademark determination device 1 . The trademark determination unit 11c of the trademark determination device 1 determines whether or not the search result has been received from the trademark management server 5 (step S5). If the search result has not been received ( S<b>5 : NO), the trademark determination unit 11 c waits until the search result is received from the trademark management server 5 .

When the search result from the trademark management server 5 is received (S5: YES), the trademark determination unit 11c determines whether the trademark matches or resembles the trademark candidate according to whether or not the search result includes a matching or similar trademark. It is determined whether or not there is a registered trademark or an application-stage trademark (step S6). If there is no registered trademark or application-stage trademark that matches or is similar to the trademark candidate (S6: NO), the notification processing unit 11e of the processing unit 11 notifies that there is no registered trademark or application-stage trademark that matches or is similar to the trademark candidate. is displayed on the display unit 15 (step S8), and the process ends.

If there is a registered trademark or an application-stage trademark that matches or is similar to the trademark candidate (S6: YES), the alternative trademark candidate generation unit 11d of the processing unit 11 regards the character string of the trademark candidate as a search term, The following character string is predicted based on the information stored in the term DB 12 b of the storage unit 12 . The substitute trademark candidate generation unit 11d generates a substitute trademark candidate by appropriately combining the character string of the trademark candidate and the character string obtained by prediction (step S7). After that, the notification processing unit 11e of the processing unit 11 displays on the display unit 15 the determination result notifying that there is a registered trademark or a trademark in the application stage that matches or is similar to the trademark candidate (step S8), and ends the process. . At this time, the notification processing unit 11e may display an advertisement regarding an agent who acts as an agent for trademark application procedures together with the alternative trademark candidates.

<Summary>
The trademark determination device 1 according to the present embodiment configured as described above extracts trademark candidates included in an input document file (content), refers to the trademark DB 5a in which trademarks are stored, and extracts trademark candidates from the trademark DB 5a. determines whether it matches or is similar to the trademark stored in the If it is determined that the trademark candidate extracted from the document file matches or resembles the trademark stored in the trademark DB 5a, the trademark determination device 1 notifies the user to that effect. Thereby, the user can grasp the trademark candidate included in the created document file and the determination result as to whether or not the trademark candidate matches or is similar to an existing trademark.

Further, in the present embodiment, the trademark management server 5 stores registered trademarks in the trademark DB 5a. The trademark determination device 1 notifies when it determines that the extracted trademark candidate matches or resembles a registered trademark. This allows the user to know that the document file contains a trademark candidate that may infringe on another person's registered trademark.

Further, in the present embodiment, the trademark management server 5 stores trademarks in the application stage in the trademark DB 5a. The trademark determination device 1 notifies when it determines that the extracted trademark candidates match or are similar to the trademark at the application stage. This allows the user to know that the document file contains trademark candidates that match or are similar to trademarks that can be registered in the future.

Further, the trademark determination apparatus 1 according to the present embodiment distinguishes between a trademark candidate determined to match or resemble a registered trademark and a trademark candidate determined to match or resemble an application-stage trademark and notifies them. This allows the user to easily determine whether the candidate trademark included in the document file matches or resembles the registered trademark or the trademark at the application stage.

Further, when the trademark determination apparatus 1 according to the present embodiment determines that the trademark candidate matches or resembles the trademark stored in the trademark DB 5a, it generates and notifies a substitute trademark candidate. A substitute trademark candidate is a trademark candidate that does not match or resemble the trademark stored in the trademark DB 5a. As a result, the trademark determination apparatus 1 can propose another trademark candidate to the user who has created a document file containing trademark candidates that match or are similar to an existing trademark.

Further, the trademark determination apparatus 1 according to the present embodiment displays an advertisement regarding an agent who acts as an agent for trademark application procedures when notifying a substitute trademark candidate. As a result, the user can easily request the agent to carry out application procedures, investigations, or the like for alternative trademark candidates.

Further, the trademark determination device 1 according to the present embodiment determines classification of trademark candidates extracted from a document file. As a result, it is possible to entrust the judgment of trademark classification, which is difficult for users who are unfamiliar with trademarks, to the trademark determination device 1 .

Further, the trademark determination device 1 according to the present embodiment uses the trademark classification determination model 100 to determine classification of trademark candidates. The trademark classification determination model 100 receives sentences in a document file as an input, and outputs classifications of candidate trademarks contained in the sentences. As a result, the trademark determination device 1 can accurately determine the classification of trademark candidates included in the document file.

In the present embodiment, the trademark determination apparatus 1 uses the trademark classification determination model 100 to determine the classification of the candidate trademark, but the invention is not limited to this. The trademark determination device 1 may, for example, extract keywords included in the document file and determine classification based on the extracted keywords. In this case, the trademark determination device 1 is provided with a table or database in which keywords and categories are associated, for example, and can determine categories based on whether or not the relevant keyword is included in the document file. can. Furthermore, the trademark determination device 1 may be configured not to determine classification. In this case, the trademark judging device 1 transmits the extracted trademark candidate to the trademark management server 5 to request a search, and causes the trademark management server 5 to search for trademarks that do not designate a category. Alternatively, it is also possible to adopt a configuration in which the user sets in advance the categories for searching trademark candidates. In these cases, the trademark classification determination system may not include the trademark classification determination model 100, the learning server 3, and the like.

In the present embodiment, the trademark determination device 1 is the user's personal computer, but the invention is not limited to this, and the trademark determination device 1 may be a server computer. In this case, the user transmits a document file created by the personal computer to the server computer and requests judgment of the trademark. The server computer judges the received document file, and transmits the judgment result to the user's personal computer to notify it. The personal computer outputs an alert based on the determination result from the server computer.

Further, in the present embodiment, the trademark judging apparatus 1 receives a document file created by a user as an input, and judges trademark candidates contained in this document file. However, the content input to the trademark determination apparatus 1 may be content other than document files.

Further, in the present embodiment, the trademark determination apparatus 1 predicts a character string following the search term when a trademark candidate character string is regarded as a search term based on the term DB 12b of the storage unit 12, and predicts a character string following the search term. We generated alternative trademark candidates based on, but not limited to, For example, the character string following the search term may be predicted by a server device or the like separate from the trademark determination device 1 . In this configuration, the trademark determination device 1 transmits trademark candidate character strings to the server device, acquires the prediction result character strings from the server device, and generates alternative trademark candidates. Further, for example, the trademark determination device 1 is provided with a trained model of artificial intelligence that predicts the character string following the input of a character string of a search term, and the trademark determination device 1 assigns trademark candidates to this trained model. A configuration may be adopted in which a character string as a prediction result is acquired by inputting.

(Modification)
The above-described trademark determination apparatus 1 processes character trademarks as trademark candidates included in a document file. However, the trademark determination device 1 may be configured to handle trademarks in the form of graphics such as logos. The trademark determination device 1 according to the modification extracts image data included in, for example, a document file, and uses this image data as a graphic trademark candidate. The trademark determination device 1 transmits to the trademark management server 5 the image data of the extracted trademark candidate and the classification determined by the trademark classification determination model 100 based on the text included in the document file.

The trademark management server 5 according to the modification stores registered trademarks or trademarks in the application stage of the figure in the trademark DB 5a, and based on the given image data and category, the registered trademarks or trademarks in the application stage of the matching or similar figure. can be searched. The trademark management server 5 extracts and compares the provided image data and the image data of the registered trademark or the trademark in the application stage by, for example, extracting the feature values of the images, and determines whether or not they are similar. do. The trademark management server 5 transmits the search result to the trademark determination device 1 . The trademark determination device 1 receives the search results from the trademark management server 5 and determines whether or not there is a registered trademark or an application-stage trademark that matches or is similar to the trademark candidates of the graphics included in the document file. The trademark determination device 1 outputs an alert for a trademark candidate with a graphic that matches or resembles a registered trademark or a trademark in the application stage.

When using a server device that manages and operates a patent information platform of the Patent Office as the trademark management server 5, it is necessary to input a figure classification when searching for a figure trademark. In this case, for example, the trademark determination device 1 may acquire the figure classification of the trademark candidate of the figure using artificial intelligence that receives image data and outputs the figure classification. This artificial intelligence is provided, for example, as a deep learning model. The deep learning model is trained in advance using teacher data in which, for example, images and graphic classifications are associated with each other, and the trained model is provided in the trademark management server 5 .

It should be noted that the trademark determination apparatus 1 does not need to generate alternative trademark candidates for graphic trademark candidates, and may display an advertisement regarding an agent when outputting an alert. Alternatively, the trademark determination device 1 may perform a process of generating a graphic as a substitute trademark candidate based on the trademark candidate graphic. In this case, for example, the trademark determination apparatus 1 takes image data as an input, applies various processing to the input image, and uses artificial intelligence to output the image as the alternative trademark candidate from the image data of the trademark candidate. data can be generated. This artificial intelligence is provided, for example, as a deep learning model. The deep learning model is trained in advance using, for example, teacher data in which an image before processing and an image after processing are associated with each other, and the trained model is provided in the trademark determination device 1 .

The embodiments disclosed this time are illustrative in all respects and should be considered not restrictive. The scope of the present invention is indicated by the scope of the claims rather than the meaning described above, and is intended to include all modifications within the scope and meaning equivalent to the scope of the claims.

1 Trademark determination device (computer, information processing device)
3 learning server 5 trademark management server 5a trademark DB
11 processing unit 11a trademark candidate extraction unit (extraction unit)
11b classification determination unit 11c trademark determination unit (determination unit)
11d Alternative trademark candidate generation unit 11e Notification processing unit (notification unit)
11f advertisement display processing unit 12 storage unit 12a trademark determination program (computer program)
13 communication unit 14 operation unit 15 display unit 31 processing unit 31a learning processing unit 31b model distribution unit 32 storage unit 32a server program 33 communication unit 98, 99 recording medium 100 trademark classification determination model 101 data for learning

Claims (10)

to the computer,
Extract trademark candidates included in the input content,
determining the classification of the candidate trademark based on the content;
referring to a database in which trademarks are stored based on the trademark candidate and the classification , and determining whether the extracted trademark candidate matches or is similar to the trademark stored in the database;
A computer program for executing a process of notifying when it is determined that the candidate trademark matches or is similar to a trademark stored in the database. The database stores registered trademarks,
to the computer,
2. The computer program according to claim 1, which executes a process of notifying when it is determined that the candidate trademark matches or is similar to a registered trademark stored in the database. The database stores trademarks in the application stage,
to the computer,
3. The computer program according to claim 2, which causes execution of a process of notifying when it is determined that the candidate trademark matches or is similar to a filing-stage trademark stored in the database. to the computer,
4. The computer program according to claim 3, which executes a process of distinguishing and notifying the trademark candidate determined to match or resemble the registered trademark from the trademark candidate determined to match or resemble the application-stage trademark. to the computer,
5. The computer program according to any one of claims 1 to 4, which executes a process of notifying a substitute trademark candidate when it is determined that the trademark candidate matches or resembles a trademark stored in the database. . to the computer,
6. The computer program according to claim 5, causing execution of a process of displaying an advertisement about an agent who represents a trademark filing procedure when notifying the substitute trademark candidate. 7. The computer program product of any one of claims 1 to 6, wherein the candidate trademarks include character or logo trademarks. 8. The computer program according to any one of claims 1 to 7 , wherein a trained model that accepts content as input and outputs a trademark classification is used to determine the classification of the candidate trademark. an extraction unit that extracts trademark candidates included in the input content;
a classification determination unit that determines classification of the candidate trademark based on the content;
Based on the trademark candidate and the classification, a database in which trademarks are stored is referenced, and trademark determination is performed to determine whether the trademark candidate extracted by the extraction unit matches or is similar to the trademark stored in the database. Department and
an information processing apparatus, comprising: a notification unit that notifies when the trademark determination unit determines that the trademark candidate matches or resembles a trademark stored in the database. A computer-implemented trademark judgment method comprising:
an extraction step of extracting trademark candidates included in the input content;
a classification determination step of determining classification of the candidate trademark based on the content;
a trademark determination step of referring to a database in which trademarks are stored based on the trademark candidate and the classification , and determining whether the extracted trademark candidate matches or is similar to the trademark stored in the database;
a notification step of notifying when it is determined that the candidate trademark matches or is similar to the trademark stored in the database;
Trademark judgment method , including

Images