Information processing device, information processing method, and information processing program

Abstract

To enable efficient provision of information requested by a user..SOLUTION: An information processing device 100 is provided, comprising an acquisition unit 121 for acquiring multiple words selected by a user, an estimation unit 122 configured to estimate relevance among the multiple words acquired by the acquisition unit 121, a generation unit 124 configured to generate a sentence corresponding to the relevance among the multiple words estimated by the estimation unit 122, and an output unit 125 for outputting the sentence generated by the generation unit 124.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to an information processing device, an information processing method, and an information processing program. [Background technology]

[0002] Conventionally, there are techniques that provide information related to a search query. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 2022-178270 Summary of the Invention [Problem to be solved by the invention]

[0004] However, conventional techniques are unable to efficiently provide information that users desire. [Means for solving the problem]

[0005] In order to solve the above-mentioned problems and achieve the objective, the information processing device is characterized by having an acquisition unit that acquires multiple words selected by a user, an estimation unit that estimates the relationship between the multiple words acquired by the acquisition unit, a generation unit that generates a sentence corresponding to the relationship between the multiple words estimated by the estimation unit, and an output unit that outputs the sentence generated by the generation unit. [Effects of the Invention]

[0006] According to the present invention, it is possible to efficiently provide information desired by a user. [Brief explanation of the drawings]

[0007] [Figure 1] FIG. 1 is a diagram illustrating an overview of an information processing apparatus according to an embodiment. [Figure 2]FIG. 2 is a block diagram illustrating an example of the configuration of the information processing device according to the embodiment. [Figure 3] FIG. 3 is a diagram illustrating an example of processing performed by the information processing apparatus according to the embodiment. [Figure 4] FIG. 4 is a diagram illustrating an example of processing performed by the information processing device according to the embodiment. [Figure 5] FIG. 5 is a diagram illustrating an example of processing performed by the information processing device according to the embodiment. [Figure 6] FIG. 6 is a diagram illustrating an example of processing performed by the information processing device according to the embodiment. [Figure 7] FIG. 7 is a flowchart illustrating an example of processing by the information processing device according to the embodiment. [Figure 8] FIG. 8 is a diagram illustrating an example of a computer that executes an information processing program. DETAILED DESCRIPTION OF THE INVENTION

[0008] Hereinafter, with reference to the drawings, an information processing device, an information processing method, and an information processing program according to the present application will be described in detail. Note that the present invention is not limited to these embodiments. In addition, in the description of the drawings, the same parts are denoted by the same reference numerals, and duplicated explanations will be omitted.

[0009] [Introduction] Conventionally, there are technologies that provide information related to search queries. However, conventional technologies cannot efficiently provide the information that users seek. For example, conventional technologies only provide information related to search queries, and cannot immediately provide information about the relationships between multiple words that appear in content. Furthermore, in order to know the relationships between multiple words that appear in content, users must search for each word and then select and discard information from the search results themselves. In addition, conventional technologies cannot resolve questions that users may have about the relationships between multiple words. As a result, the burden on users to obtain the information they seek remains significant.

[0010] Therefore, the information processing device 100 according to this embodiment performs the following process. First, the information processing device 100 acquires, for example, multiple words selected by the user, such as "○○ Taro" and "○○ Summit," as shown in FIG. 1(1). Next, the information processing device 100 estimates the association between the acquired multiple words "○○ Taro" and "○○ Summit" as, for example, "○○ Taro," "○○ Summit," and "participation." Next, the information processing device 100 generates a sentence corresponding to the estimated association between the multiple words, "○○ Taro," "○○ Summit," and "participation." Then, the information processing device 100 outputs and provides the generated sentence, "○○ Taro has participated in the first to fifth ○○ Summits in the past. ○○ Taro said "~~" about problem △△ at the third ○○ Summit. ○○ Taro plans to participate in the sixth ○○ Summit," as shown in FIG. 1(2), for example, to the user.

[0011] As a result, the information processing device 100 can efficiently provide the information desired by the user by estimating the association between multiple words selected by the user and generating a sentence corresponding to the estimated association.

[0012] [Configuration of information processing device] Next, the configuration of the information processing device 100 will be described with reference to Fig. 2. Fig. 2 is a diagram showing an example of the configuration of the information processing device 100 according to the embodiment. As shown in Fig. 2, the information processing device 100 includes a communication unit 110, a control unit 120, and a storage unit 130. Note that these units may be held in a distributed manner in multiple devices. The processing of these units will be described below.

[0013] The communication unit 110 is realized by a NIC (Network Interface Card) or the like, and enables communication between an external device and the control unit 120 via a telecommunication line such as a LAN (Local Area Network) or the Internet. For example, the communication unit 110 enables communication between the external device and the control unit 120.

[0014] The storage unit 130 is realized by a semiconductor memory element such as a RAM (Random Access Memory) or a flash memory, or a storage device such as a hard disk or an optical disk. Examples of information stored in the storage unit 130 include a user ID, user attributes, search history, browsing history, article information, data related to various machine learning algorithms, learning data for machine learning, and trained models. Here, user attributes include information such as age, gender, birthplace, residence, family structure, occupation, hobbies, and interests. The search history includes information such as search terms, the date and time of the search, and the number of searches. The search history can be displayed in a format tailored to the purpose, such as a search suggestion in which search terms are sorted in order of the number of searches. The article information includes information such as the article title, headline, content, highlights (parts that attracted attention), comments, the creator, ratings for the creator, ratings for the article, and the posting date and time. The article information may also be information that has been analyzed using natural language processing. The information stored in the storage unit 130 is not limited to the above.

[0015] The control unit 120 is realized using a CPU (Central Processing Unit), an NP (Network Processor), an FPGA (Field Programmable Gate Array), or the like, and executes a processing program stored in a memory. As shown in Fig. 2, the control unit 120 has an acquisition unit 121, an estimation unit 122, an identification unit 123, a generation unit 124, and an output unit 125. Each unit of the control unit 120 will be described below.

[0016] The acquisition unit 121 acquires multiple words selected by the user. For example, the acquisition unit 121 acquires multiple words selected from content viewed by the user (any news article, web page, etc.). For example, the acquisition unit 121 acquires multiple words selected from an article previously viewed by the user or an article currently being viewed by the user. Note that the words acquired by the acquisition unit 121 may be selected from different articles. More specifically, the acquisition unit 121 acquires multiple words selected by the user, such as "○○ Taro" and "○○ Summit."

[0017] Furthermore, the acquisition unit 121 acquires a sentence input by the user. For example, the acquisition unit 121 acquires a sentence input by the user into an input form. For example, the acquisition unit 121 acquires a question or request input by the user into a search form. More specifically, the acquisition unit 121 acquires a sentence input by the user, "Please tell me about information on other summits."

[0018] The estimation unit 122 estimates the associations between the multiple words acquired by the acquisition unit 121. For example, the estimation unit 122 estimates the associations between the multiple words acquired by the acquisition unit 121 by using a learning model that takes multiple words as input and outputs the associations between the multiple words using the multiple words and the associations between the multiple words.

[0019] Furthermore, for example, the estimation unit 122 may estimate the association between a plurality of words using the result of classifying words by clustering. For example, the estimation unit 122 estimates the association between a plurality of words acquired by the acquisition unit 121 by using a learning model that receives a plurality of words and word classifications as input and outputs the association between the plurality of words using the plurality of words, the word classifications, and the association between the plurality of words.

[0020] The association between multiple words estimated by the estimation unit 122 may be expressed in the form of only the words, or in a form in which a verb, a noun, an adjective, a particle, a complement, etc. is added to the words. For example, if the multiple words are "○○ Taro" and "○○ Summit," the association between the multiple words may be expressed in the form of "○○ Taro," "○○ Summit," and "participation," or in the form of "○○ Taro's participation in ○○ Summit."

[0021] Furthermore, the estimation unit 122 estimates the relevance between the multiple words acquired by the acquisition unit 121 using the search history. For example, the estimation unit 122 may estimate the relevance between the multiple words by comparing the multiple words selected by the user with the user's search history. For example, the estimation unit 122 estimates the relevance between the multiple words acquired by the acquisition unit 121 by comparing the multiple words selected by the user with search suggestions when a keyword including one or more of the words selected by the user is used as a search query. For example, the estimation unit 122 may estimate, as the relevance between the multiple words, "○○ Taro," "○○ Summit," and "Participation," which are the most frequently searched for words among search suggestions including the multiple words selected by the user, "○○ Taro" and "○○ Summit."

[0022] The search history used by the estimation unit 122 when estimating the relevance between a plurality of words may be for all users or for a specific user.

[0023] Furthermore, the estimation unit 122 estimates the relevance between the plurality of words acquired by the acquisition unit 121 using user attributes. For example, the estimation unit 122 estimates the relevance between the plurality of words acquired by the acquisition unit 121 using the user's interests as the user attributes. For example, the estimation unit 122 estimates the relevance corresponding to the user's interests as the user attributes as the relevance between the plurality of words acquired by the acquisition unit 121. More specifically, the estimation unit 122 may estimate the relevance between the plurality of words "○○ Taro", "○○ Summit", and "environmental issues" from the user's interests (nature and environmental issues).

[0024] The user attributes used by the estimation unit 122 when estimating the association between multiple words can be appropriately selected from one or more of age, sex, birthplace, residence, family structure, occupation, hobbies, interests, etc.

[0025] The identification unit 123 identifies an article related to the association between a plurality of words estimated by the estimation unit 122 and a sentence input by the user. For example, the identification unit 123 identifies an article related to the association between a plurality of words estimated by the estimation unit 122 and the content of a sentence such as a question or request acquired by the acquisition unit 121. More specifically, first, the identification unit 123 analyzes the sentence acquired by the acquisition unit 121 using natural language processing such as morphological analysis, grammatical analysis, semantic analysis, and contextual analysis.

[0026] Thereafter, the identification unit 123 receives as input the correlations between the multiple words estimated by the estimation unit 122 and the analyzed sentence, and uses the correlations between the multiple words, the analyzed sentence, and information on the analyzed article stored in the storage unit 130 to use a learning model that outputs articles that are highly relevant between the correlations between the multiple words estimated by the estimation unit 122 and the sentence acquired by the acquisition unit 121, thereby identifying articles that are highly relevant between the correlations between the multiple words estimated by the estimation unit 122 and the sentence acquired by the acquisition unit 121. Note that conventional technologies according to the purpose can be used for the natural language processing and learning model used in the process of identifying articles that are highly relevant to the sentence acquired by the acquisition unit 121.

[0027] The generation unit 124 generates a sentence corresponding to the relationship between the multiple words estimated by the estimation unit 122. For example, the generation unit 124 uses the relationship between the multiple words estimated by the estimation unit 122 to generate a sentence explaining the relationship between the multiple words estimated by the estimation unit 122. More specifically, the generation unit 124 uses the relationship between the multiple words estimated by the estimation unit 122, "Taro XX's participation in the XX summit," to generate a sentence explaining the relationship between the multiple words, "Taro XX's participation in the XX summit."

[0028] Furthermore, the generation unit 124 generates a sentence that explains the relationship between a plurality of words based on the relationship between the plurality of words, using a model that has learned the relationship between the relationship and the sentence that explains the relationship. Here, when generating the sentence, the generation unit 124 uses a model that has been trained to output an answer sentence that corresponds to the input sentence. For example, this model is stored in a server that processes information and was created independently by a business operator (Yahoo). Note that it is desirable to keep input information, such as personal information, confidential by training it so that it will not be used as a new answer.

[0029] Furthermore, the generation unit 124 uses information about an article in which multiple words are selected to generate a sentence corresponding to the relationship between the multiple words estimated by the estimation unit 122. For example, the generation unit 124 uses information about an article in which each word is selected to generate a sentence corresponding to the relationship between the multiple words estimated by the estimation unit 122. More specifically, the generation unit 124 uses information about the relationship between the multiple words estimated by the estimation unit 122, "Taro XX's participation in the XX Summit," and information about an article in which multiple words are selected, "Main theme of the XXth XX Summit," to generate a sentence explaining the relationship between the multiple words, "Taro XX's participation in the XX Summit."

[0030] Furthermore, the generation unit 124 generates a summary of the article identified by the identification unit 123 as a response to the sentence acquired by the acquisition unit 121. For example, the generation unit 124 receives the sentence acquired by the acquisition unit 121 and information on the article identified by the identification unit 123 as input, and generates a summary of the article identified by the identification unit 123 as a response to the sentence acquired by the acquisition unit 121 using a model that outputs a summary of the article.

[0031] The output unit 125 outputs the sentence generated by the generation unit 124. For example, the output unit 125 displays the sentence generated by the generation unit 124 on a terminal screen. For example, the output unit 125 displays the sentence generated by the generation unit 124 together with search results. Furthermore, for example, the output unit 125 transmits the sentence generated by the generation unit 124 to an external device or an external system.

[0032] [Processing performed by information processing device] Next, an example of processing performed by the information processing device 100 according to the embodiment will be described with reference to FIGS. 3 to 6. FIGS. 3 to 6 are diagrams illustrating an example of processing performed by the information processing device 100 according to the embodiment. Here, FIG. 3(1) is a schematic diagram showing word selection by a user. For example, as shown in FIG. 3(1), when a user selects the words "○○ Taro" and "○○ Summit" from a displayed article, the words selected by the user are displayed in the window at the bottom of FIG. 3(1). Next, for example, the user presses the "Check Relevance" button in the window at the bottom of FIG. 3(1). At this time, the acquisition unit 121 acquires the multiple words "○○ Taro" and "○○ Summit" selected by the user.

[0033] 3(2) is a schematic diagram illustrating a process of estimating the relevance that the user wants to know. Here, for example, the estimation unit 122 estimates the relevance between the multiple words acquired by the acquisition unit 121 by comparing the multiple words selected by the user, "○○ Taro" and "○○ Summit," with search suggestions when a keyword including one or more of the words selected by the user is used as a search query.

[0034] For example, as shown in FIG. 3(2), when a search is performed for "XX Summit," search suggestions such as "XX Summit participants," "XX Summit participating countries," "XX Summit schedule," "XX Summit venue," etc. are displayed. The estimation unit 122 compares such search suggestions with the multiple words selected by the user, and estimates, for example, the suggestion with the most searches among the suggestions containing the multiple words selected by the user as the association between the multiple words.

[0035] At this time, the estimation unit 122 may estimate the relevance between multiple words using user attributes. For example, the estimation unit 122 may estimate "○○ Taro," "○○ Summit," and "environmental issues" as the relevance between multiple words based on the user's interests (nature and environmental issues). Note that when estimating the relevance between multiple words, the estimation unit 122 may use a combination of elements such as search history and user attributes. For example, the estimation unit 122 can estimate, from among suggestions including multiple words selected by the user, those that have been searched for a certain number of times or more and that correspond to the user attributes as the relevance between multiple words.

[0036] Next, FIG. 4(1) is a schematic example of information generated by the generation unit 124. For example, the generation unit 124 inputs the associations between multiple words estimated by the estimation unit 122 into a large-scale pre-trained language model independently created by the business (Yahoo!) to generate a sentence explaining the associations between multiple words estimated by the estimation unit 122. For example, the generation unit 124 generates information such as "○○ Taro has participated in the 1st to 5th ○○ Summits in the past. ○○ Taro said "~~" about problem △△ at the 3rd ○○ Summit. ○○ Taro is planning to participate in the 6th ○○ Summit." as a sentence corresponding to the associations between multiple words "○○ Taro," "○○ Summit," and "participation" estimated by the estimation unit 122.

[0037] 4(2) is a schematic diagram showing an example of a screen display of the sentence generated by the generation unit 124. As shown in FIG. 4(2), the output unit 125 displays the sentence generated by the generation unit 124 together with the search results.

[0038] Next, Fig. 5(1) is a schematic diagram showing an example of a sentence input by a user. For example, as shown in Fig. 5(1), the acquisition unit 121 acquires a sentence expressing a request input by the user for the generated sentence, such as "Please tell me about information on other summits."

[0039] 5(2) is a schematic example of article information stored in the storage unit 130. For example, the identification unit 123 performs natural language processing such as morphological analysis on the information of the sentence acquired by the acquisition unit 121, inputs the analyzed sentence and the associations between multiple words, compares the information of the analyzed sentence and the associations between multiple words with the information of each analyzed article stored in the storage unit 130, and uses a model that outputs articles highly related to the associations between multiple words estimated by the estimation unit 122 and the sentence acquired by the acquisition unit 121, thereby identifying articles related to the associations between multiple words estimated by the estimation unit 122 and the sentence acquired by the acquisition unit 121. For example, the identification unit 123 identifies article A and article D as articles related to the associations between multiple words estimated by the estimation unit 122, "Taro XX," "XX Summit," and "Participation," and the sentence acquired by the acquisition unit 121, "Please tell me about other summits."

[0040] Next, the generation unit 124 generates a summary of the article identified by the identification unit 123 as a response to the sentence acquired by the acquisition unit 121. For example, the generation unit 124 inputs the sentence acquired by the acquisition unit 121 and information on the article identified by the identification unit 123 into a trained large-scale language model independently created by a business operator (Yahoo), thereby obtaining a result that summarizes the information on the article identified by the identification unit 123 to be a response to the sentence acquired by the acquisition unit 121.

[0041] For example, the generation unit 124 receives as input the sentence "Tell me about the achievements of player XX" acquired by the acquisition unit 121, information on the article "XX Taro mentions ~~ at the XX Summit" identified by the identification unit 123, and information on the article "At the △△ Summit, cooperation with allied nations is key," and generates information such as "XX Taro participated in the 4th XX Summit and the 6th △△ Summit. At the 4th XX Summit, XX Taro expressed his opinion by saying "..." about the problematic ~~. At the meeting held at the 6th △△ Summit, XX Taro said, "Cooperation with allied nations □□ and □□ is the key to ~~."

[0042] 6 is a schematic diagram showing an example in which the text generated by the generation unit 124 and the article title (including URL information) are displayed on the screen. In this way, when the user inputs a text such as a question or request regarding the output result by the output unit 125, the information processing device 100 again acquires the input text, identifies related articles, summarizes the articles so as to answer the input text, and outputs the summary content, thereby repeating the process.

[0043] 〔flowchart〕 Next, the flow of processing by the information processing device 100 will be described with reference to Fig. 7. Fig. 7 is a flowchart showing an example of the flow of processing according to this embodiment.

[0044] First, the acquiring unit 121 acquires a plurality of words selected by the user (step S101). For example, the acquiring unit 121 acquires a plurality of words selected from content browsed by the user.

[0045] Next, the estimation unit 122 estimates the relevance between the multiple words acquired by the acquisition unit 121 (step S102). For example, the estimation unit 122 estimates the relevance between the multiple words acquired by the acquisition unit 121 by using a search history.

[0046] Then, the generation unit 124 generates a sentence corresponding to the association between the multiple words estimated by the estimation unit 122 (step S103). The generation unit 124 uses the association between the multiple words estimated by the estimation unit 122 to generate a sentence explaining the association between the multiple words estimated by the estimation unit 122, using a model trained to output an answer sentence corresponding to the input sentence.

[0047] Then, the output unit 125 outputs the sentence generated by the generation unit 124 (step S104). For example, the output unit 125 displays the sentence generated by the generation unit 124 together with the search results.

[0048] Thereafter, the acquisition unit 121 determines whether or not a sentence has been input by the user (step S105). If the acquisition unit 121 determines that a sentence has been input by the user (step S105 "YES"), the acquisition unit 121 acquires the sentence input by the user (step S106). For example, the acquisition unit 121 acquires a question or request input by the user.

[0049] Thereafter, the identifying unit 123 identifies an article related to the association between the plurality of words estimated by the estimating unit 122 and the sentence input by the user (step S107). For example, the identifying unit 123 identifies an article related to the association between the plurality of words estimated by the estimating unit 122 and the content of the sentence such as a question or request acquired by the acquiring unit 121.

[0050] Then, the generation unit 124 generates a summary of the article identified by the identification unit 123 as a response to the sentence acquired by the acquisition unit 121 (step S108). For example, the generation unit 124 receives the sentence acquired by the acquisition unit 121 and information on the article identified by the identification unit 123 as input, and generates a summary of the article identified by the identification unit 123 as a response to the sentence acquired by the acquisition unit 121 using a model that outputs an article summary.

[0051] On the other hand, if the acquiring unit 121 determines that no text has been input by the user (step S105 "NO"), the information processing device 100 ends the process.

[0052] 〔effect〕 The information processing device 100 according to the embodiment includes an acquisition unit 121 that acquires multiple words selected by a user, an estimation unit 122 that estimates the association between the multiple words acquired by the acquisition unit 121, a generation unit 124 that generates a sentence corresponding to the association between the multiple words estimated by the estimation unit 122, and an output unit 125 that outputs the sentence generated by the generation unit 124.

[0053] As a result, the information processing device 100 can efficiently provide the information the user is looking for by estimating the relationship between multiple words selected by the user and generating a sentence that explains the relationship between the words that the user wants to know.

[0054] Furthermore, the estimation unit 122 of the information processing device 100 according to the embodiment uses the search history to estimate the relevance between the multiple words acquired by the acquisition unit 121. As a result, the information processing device 100 uses the information of the search history to estimate the relevance between the multiple words selected by the user, and generates a sentence explaining the relevance between the words that the user wants to know, thereby making it possible to efficiently provide the information desired by the user.

[0055] Furthermore, the estimation unit 122 of the information processing device 100 according to the embodiment uses user attributes to estimate the association between the plurality of words acquired by the acquisition unit 121. As a result, the information processing device 100 uses information on the user attributes to estimate the association between the plurality of words selected by the user, and generates a sentence explaining the association between the words that the user wants to know, thereby making it possible to efficiently provide information desired by the user.

[0056] Furthermore, the generation unit 124 of the information processing device 100 according to the embodiment generates sentences explaining the associations between a plurality of words from the associations between the plurality of words using a model that has learned the relationship between the associations and sentences explaining the associations. As a result, the information processing device 100 estimates the associations between a plurality of words selected by a user and generates sentences explaining the associations between the words that the user wants to know using the learned model, thereby efficiently providing the information desired by the user.

[0057] Furthermore, the generation unit 124 of the information processing device 100 according to the embodiment uses information about an article from which a plurality of words are selected to generate a sentence corresponding to the association between the plurality of words estimated by the estimation unit 122. In this way, the information processing device 100 can efficiently provide information desired by a user, taking into account information such as the article from which the word was selected, and the context in which the user sought information about the word.

[0058] Moreover, the acquisition unit 121 of the information processing device 100 according to the embodiment further includes an identification unit 123 that acquires a sentence input by a user and identifies an article related to the association between a plurality of words estimated by the estimation unit 122 and the sentence input by the user acquired by the acquisition unit 121, and the generation unit 124 summarizes the article identified by the identification unit 123 as an answer to the sentence acquired by the acquisition unit 121. In this way, the information processing device 100 identifies an article related to a question, request, etc. from the user regarding the generated sentence and summarizes the content of the identified article as an answer, thereby resolving any doubts the user has about the generated sentence and efficiently providing the information the user desires.

[0059] 〔program〕 It is also possible to create a program written in a computer-executable language that executes the processes executed by the information processing device 100 described in the above embodiment. In this case, the same effects as those of the above embodiment can be achieved by having a computer execute the program. Furthermore, such a program may be recorded on a computer-readable recording medium, and the program recorded on the recording medium may be read and executed by a computer to achieve the same processes as those of the above embodiment.

[0060] Fig. 8 is a diagram showing an example of a computer that executes an information processing program. As shown in Fig. 8, the controller 200 includes a computer 1000, which includes, for example, a memory 1010, a CPU 1020, a hard disk drive interface 1030, a disk drive interface 1040, a serial port interface 1050, a video adapter 1060, and a network interface 1070. These components are connected by a bus 1080.

[0061] The memory 1010 includes a ROM (Read Only Memory) 1011 and a RAM 1012. The ROM 1011 stores, for example, a boot program such as a BIOS (Basic Input Output System). The hard disk drive interface 1030 is connected to a hard disk drive 1090. The disk drive interface 1040 is connected to a disk drive 1041. A removable storage medium such as a magnetic disk or optical disk is inserted into the disk drive 1041. The serial port interface 1050 is connected to, for example, a mouse 1110 and a keyboard 1120. The video adapter 1060 is connected to, for example, a display 1130.

[0062] 8, the hard disk drive 1090 stores, for example, an OS 1091, an application program 1092, a program module 1093, and program data 1094. The tables described in the above embodiments are stored in the hard disk drive 1090 or the memory 1010, for example.

[0063] The control program is stored in the hard disk drive 1090 as a program module in which instructions to be executed by the computer 1000 are written. Specifically, the hard disk drive 1090 stores a program module 1093 in which each process executed by the controller 2000 described in the above embodiment is written.

[0064] Furthermore, data used for information processing by the control program is stored as program data, for example, in the hard disk drive 1090. Then, the CPU 1020 reads out the program module 1093 and program data 1094 stored in the hard disk drive 1090 into the RAM 1012 as necessary, and executes each of the above-described procedures.

[0065] The program module 1093 and program data 1094 related to the control program are not limited to being stored in the hard disk drive 1090, but may be stored in a removable storage medium and read by the CPU 1020 via the disk drive 1041, etc. Alternatively, the program module 1093 and program data 1094 related to the control program may be stored in another computer connected via a network such as a LAN (Local Area Network) or a WAN (Wide Area Network), and read by the CPU 1020 via the network interface 1070.

[0066] 〔others〕 Although various embodiments have been described in detail herein with reference to the drawings, these embodiments are merely examples and are not intended to limit the present invention. In other words, the present invention makes it possible to learn, build, and update various multi-class classification models and estimate desired probabilities by changing the features used as input. The features described herein can be realized in various ways, including various modifications and improvements based on the knowledge of those skilled in the art.

[0067] Furthermore, the above-mentioned "module (-er suffix, -or suffix)" can be read as a unit, means, circuit, etc. For example, a communication module, a control module, and a storage module can be read as a communication unit, a control unit, and a storage unit, respectively. [Explanation of symbols]

[0068] 100 Information processing device 110 Communications Department 120 control section 121 Acquisition Department 122 Estimation Department 123 Specific part 124 Generation part 125 Output section 130 Storage section

Claims

1. an acquisition unit that acquires a plurality of words selected by a user; an estimation unit that refers to a search history and estimates a relationship between a plurality of words that are search suggestions obtained when the plurality of words acquired by the acquisition unit are used as a search query; a generation unit that inputs the associations between the plurality of words estimated by the estimation unit into a large-scale language model and generates a sentence that explains the associations between the plurality of words; an output unit that outputs the sentence generated by the generation unit; An information processing device comprising:

2. An information processing method executed by an information processing device, an acquisition step of acquiring a plurality of words selected by a user; an estimation step of estimating, with reference to a search history, relevance between a plurality of words that are search suggestions obtained when the plurality of words acquired in the acquisition step are used as a search query; a generation step of inputting the associations between the plurality of words estimated by the estimation step into a large-scale language model to generate a sentence explaining the associations between the plurality of words; an output step of outputting the sentence generated by the generation step; An information processing method comprising:

3. an acquiring step of acquiring a plurality of words selected by a user; an estimation step of estimating, with reference to a search history, relevance between a plurality of words that are search suggestions obtained when the plurality of words acquired by the acquisition step are used as a search query; a generation step of inputting the associations between the plurality of words estimated by the estimation step into a large-scale language model to generate sentences that explain the associations between the plurality of words; an output step of outputting the sentence generated by the generation step; An information processing program characterized by causing a computer to execute the above.

Images