JP7041639B2 - Selection device, selection method and selection program - Google Patents

Description

The present invention relates to a selection device, a selection method and a selection program.

Conventionally, information providing technology for providing various information to users has been known. As an example of such information providing technology, a technology for selecting information to be provided according to the attributes of a user who is an information providing destination is known. For example, in a so-called Q & A (question answering) site, there is known a technique of estimating an ideal respondent based on the attributes of a questioner and notifying the estimated respondent of the question.

Japanese Unexamined Patent Publication No. 2012-033076

However, it may not be possible to say that the above-mentioned prior art provides appropriate information to the user.

For example, in the above-mentioned Q & A site, if the expression of the document posted as an answer by the respondent does not suit the questioner's preference, the questioner may cancel the use of the Q & A site. In addition, if the expression of the text posted as a question by the questioner does not suit the respondent's preference, there is a risk that an appropriate answer will not be posted.

The present application has been made in view of the above, and an object thereof is to provide more appropriate information to the user.

The selection device according to the present application is provided by an estimation unit that estimates the properties of the user based on the information input by the user, and a provider having properties corresponding to the properties of the user estimated by the estimation unit. It is characterized by having a selection unit for selecting the provided target to be provided as the provided target to be provided to the user.

According to one aspect of the embodiment, more appropriate information can be provided to the user.

FIG. 1 is a diagram showing an example of an information providing device according to the first embodiment. FIG. 2 is a diagram showing a configuration example of the information providing device according to the first embodiment. FIG. 3 is a diagram showing an example of information registered in the user database according to the first embodiment. FIG. 4 is a diagram showing an example of information registered in the posting database according to the first embodiment. FIG. 5 is a flowchart showing an example of the flow of the learning process according to the first embodiment. FIG. 6 is a flowchart showing an example of the flow of the selection process according to the first embodiment. FIG. 7 is a diagram showing an example of a selection process executed by the information providing device according to the second embodiment when delivering an advertisement. FIG. 8 is a diagram showing a configuration example of the information providing device according to the second embodiment. FIG. 9 is a diagram showing an example of information registered in the user database according to the second embodiment. FIG. 10 is a diagram showing an example of information registered in the advertisement database according to the second embodiment. FIG. 11 is a diagram showing an example of a selection process executed when the information providing device according to the third embodiment receives an inquiry. FIG. 12 is a diagram showing a configuration example of the information providing device according to the third embodiment. FIG. 13 is a diagram showing an example of information registered in the operator database according to the third embodiment. FIG. 14 is a diagram showing an example of a hardware configuration.

Hereinafter, a selection device, a selection method, and a mode for carrying out the selection program (hereinafter, referred to as “embodiment”) according to the present application will be described in detail with reference to the drawings. It should be noted that this embodiment does not limit the selection device, selection method and selection program according to the present application. Further, in each of the following embodiments, the same parts are designated by the same reference numerals, and duplicate description will be omitted.

[First Embodiment]
[1-1. About an example of information providing device]
First, an example of the information providing device 10 which is an example of the selection device according to the embodiment will be described with reference to FIG. In the following description, as an example of the selection process executed by the information providing device 10, an example of the selection process executed in the service related to the so-called Q & A site will be described, but the embodiment is not limited to this. As will be clarified in the explanation described later, the selection process can be applied to any service provided to the user, such as various community sites, in addition to the service related to the Q & A site.

Further, in the following explanation, the user who posts the text or the like that becomes a question to the Q & A site is described as the questioner, and the user who posts the text or the like that becomes the answer to the question is described as the respondent. Further, in the following explanation, information received from each user, such as a question sentence and an answer sentence, may be collectively referred to as posted information posted by each user.

FIG. 1 is a diagram showing an example of an information providing device according to the first embodiment. The information providing device 10 shown in FIG. 1 is an information processing device that performs selection processing, and is realized by, for example, a server device, a cloud system, or the like. For example, the information providing device 10 is a network such as various wireless communication networks such as 4G (Generation), 5G, LTE (Long Term Evolution), Wifi (registered trademark), wireless LAN (Local Area Network), or various wired communication networks. Communication is performed between the terminal devices QT1 and RT1 to RT3 used by each user via N (for example, see FIG. 2).

The terminal devices QT1 and RT1 to RT3 (hereinafter, may be collectively referred to as "terminal device T") are realized by a smart device such as a PC (Personal Computer), a server device, a smart television, a smartphone or a tablet, and the like. It is a portable terminal device capable of communicating with the information providing device 10 via the network N. Further, the terminal device T is a screen of a liquid crystal display or the like and has a screen having a touch panel function, and is distributed from the information providing device 10 such as a tap operation, a slide operation, a scroll operation, etc. from the user with a finger, a stylus, or the like. It may have a function capable of accepting various operations on the content to be performed.

In the example shown in FIG. 1, the terminal device QT1 is a terminal device used by the questioner Q1, and the terminal devices RT1 to RT3 are terminal devices used by different respondents R1 to R3. do. Further, in the example shown in FIG. 1, one questioner Q1 and three respondents R1 to R3 are described, but the embodiment is not limited to this. The information providing device 10 may accept an arbitrary number of questions from an arbitrary number of questioners, and may accept an arbitrary number of answers from an arbitrary number of respondents.

[1-2. About selection process]
Here, on the Q & A site, by improving the quality of the answers to each question, it is possible to encourage the posting of new questions, and as a result, the number of views of the questions and answers should be improved and the entire site should be activated. Is thought to be possible. On the other hand, even if the content of the answer is correct, depending on the sentence expression, as a result of offending the questioner, the frequency of questions may be reduced and the entire site may not be activated.

Therefore, the information providing device 10 corresponds not to the characteristics of the posted information itself such as the content and expression of the posted information such as a question or an answer, but to the nature of the poster (that is, the questioner or the respondent) who posted the posted information. Ask the respondents questions based on whether or not they are. For example, the information providing device 10 estimates the nature of the respondent based on the history of the responses posted by the respondent in the past. For example, the information providing device 10 estimates the personality of the respondent, such as whether he / she is impatient or patient. The information providing device 10 has a tendency of information that the respondent prefers, such as whether or not the respondent prefers a long sentence, whether or not he / she prefers a direct expression, and whether or not he / she frequently uses a critical expression. May be estimated as the nature of the respondent.

Then, the information providing device 10 selects, among the questions posted on the Q & A site, the question posted by the questioner having the property corresponding to the nature of the respondent as the provision target. For example, when the respondent is presumed to have an impatient personality, the information providing device 10 selects a question posted by a similarly impatient questioner and provides the respondent with information on the selected question. do. Further, in another example, when the respondent is presumed to prefer a long sentence, the information providing device 10 selects a question posted by a questioner who also prefers a long sentence, and inputs the information of the selected question. Provide to respondents.

As described above, the information providing device 10 selects the information to be provided to each user based on the similarity of the nature of the poster who posted the posted information, not the similarity between the posted information. Specifically, the information providing device 10 selects the information provided by the information provider having the property corresponding to the property of the user as the information to be provided to the user who is the information providing destination. As a result of such processing, the information providing device 10 can, for example, prompt the user to post an answer that the questioner is likely to like, so that appropriate information can be provided to the user. To give a more specific example, the information providing device 10 encourages the posting of new questions as a result of increasing the frequency with which the questioner likes the answer of the respondent, so that the entire site can be activated. can.

Hereinafter, an example of a process in which the personality of the user is estimated as the nature of the user and the information provided by the provider of the estimated personality and the corresponding personality is provided to the user will be described. Here, the nature of the user is not limited to the so-called various personalities, but is a concept including feelings about the preference of what kind of information is preferred.

[1-3. About an example of learning process]
For example, the information providing device 10 executes a learning process for generating a learning model for estimating the properties of the user from the posted information input by each user prior to the selection process. For example, in the example shown in FIG. 1, the information providing device 10 causes the learning model to learn the relationship between the past logs of the users U1 and U2 and the personalities of the users U1 and U2 (step S1).

For example, when the information providing device 10 inputs the feature of the past log # 1 of the user U1 into the learning model, the personality diagnosis result # U1 of the user U1 is output, and the past log # 2 of the user U2 is output. The learning model is trained so that the personality diagnosis result # U2 of the user U2 is output when the characteristics to be possessed are input to the learning model.

It should be noted that such a learning model can be realized by various classifiers such as SVM (Support Vector Machine) and DNN (Deep Neural Network). As the DNN, for example, a neural network having an arbitrary configuration such as RNN (Recurrent Neural Network), CNN (Convolution Neural Network), LSTM (Long short-term memory), etc. can be adopted.

Hereinafter, a specific example of the learning process will be described. For example, the information providing device 10 has a personality that is pre-diagnosed by a predetermined diagnostic method as a property of an arbitrary number of arbitrary users U1 and U2 (hereinafter, simply collectively referred to as "user U"). Get the diagnosis result. For example, the information providing device 10 acquires a score indicating each personality characteristic such as openness, seriousness, extroversion, cooperation, and mental stability of each user U diagnosed by a personality diagnosis method called Big5. .. The information providing device 10 assigns a score indicating each personality characteristic such as strictness, kindness, calmness, free-spiritedness, and obedience of each user U diagnosed by a personality diagnosis method called egogram. You may get it. Further, the information providing device 10 has a score indicating each personality characteristic such as other-negativeness, other-affirmation, self-affirmation, and self-negativeness of each user U diagnosed by a personality diagnosis method called OK gram. May be obtained.

The personality diagnosis result may be, for example, information obtained by asking each user U some questions in advance and answering the questions. Further, such a personality diagnosis result may be estimated from the usage history of various services provided to the user U, such as a fortune-telling site and a matchmaking site.

Further, such a personality diagnosis result may be estimated based on the relationship between the preset personality and the characteristic word. For example, the information providing device 10 holds in advance the correspondence between each personality and a keyword that is easy for a user having that personality to input, and extracts the keyword from various character strings posted in the past by the user U. .. Then, the information providing device 10 may estimate the personality of the user U based on the personality associated with the extracted keyword. Further, the information providing device 10 may specify a keyword that co-occurs with the search query input by the user U, and estimate the personality of the user U based on the personality corresponding to the specified keyword.

In addition to the personality traits described above, the information providing device 10 also includes, for example, whether or not a long sentence is preferred, whether or not a direct expression is preferred, whether or not a redundant expression is preferred, and the like. Information indicating the preference for the text may be acquired as a personality diagnosis result. That is, the information providing device 10 is information that can be an index of posted information preferred by the user U, and if it is information about the user U itself, the information providing device 10 acquires arbitrary information as information indicating the nature of the user U. good.

Subsequently, the information providing device 10 classifies each user U into categories based on the personality diagnosis result. For example, the information providing device 10 classifies each user U into categories according to the personality diagnosis result on an arbitrary scale such as an extroverted user, a cooperative user, and a user who prefers long sentences. .. Then, the information providing device 10 extracts feature information from the input information input by the user U belonging to the same category for each category by using the mutual information amount. For example, the information providing device 10 is used for various UGCs (User Generated) such as search queries, questions, answers, information posted on SNS (Social Networking Service), profiles, various comments, etc., which are input to users U belonging to the same category. Contents) is collected. The information providing device 10 collects, for example, the answer selected by the user U as the best answer among the answers to the questions posted by the user U, that is, the answer that the user U likes most as UGC information. May be good.

Subsequently, the information providing device 10 extracts a character string such as a word or a frequently appearing expression from the input information as a feature word, and calculates a mutual information amount (Mutual information) between the extracted feature word and the category. Then, the information providing device 10 selects a predetermined number of feature words as the feature words in the category in order from the one with the highest mutual information amount value. By executing such processing for each category, the information providing device 10 can obtain the feature words corresponding to each category as the features of the past log.

Subsequently, the information providing device 10 learns so that when a feature of the past log corresponding to a certain category, that is, a feature word corresponding to the category is input to the learning model, the feature word and the corresponding category are output. Train the model. For example, when the feature word is input to the learning model, the information providing device 10 learns the learning model by using a technique such as backpropagation so as to output information indicating the category corresponding to the feature word. .. The information providing device 10 may learn the learning model so as to output a score indicating the accuracy corresponding to the input feature word and the category for each category. Further, the information providing device 10 may learn the learning model so as to output the information indicating the category corresponding to the feature word included in the UGC information when the UGC information is input. As a result of such processing, the information providing device 10 can obtain a learning model that outputs information indicating the character of the user U when the information input by the user U is input.

For example, the information providing device 10 provides information indicating the character of the user U when the history of the text input by the user U as UGC information or the text of the answer selected by the user U as the best answer is input. The training model may be trained so as to output. When such learning is performed, the information providing device 10 learns to estimate the personality of the user U based on, for example, the characteristics of the text created by the user U and the characteristics of the text preferred by the user U. It is possible to realize the learning of the model.

[1-4. About an example of selection processing]
Subsequently, the information providing device 10 estimates the personality of the user from the UGC information of the user to whom the information is provided, using a learning model, and is provided by the provider of the personality corresponding to the estimated personality. Provide information. For example, the information providing device 10 estimates the personality of the respondent and provides the respondent with a question posted by the questioner having the personality corresponding to the estimated personality. That is, the information providing device 10 does not simply perform matching according to the similarity and correspondence between the questioner's question and the respondent's answer, but the respondent's personality and the questioner can be glimpsed from the respondent's answer. Matches with the personality of the respondent and provides the respondent with a question posted by a questioner with a personality that matches the personality of the respondent.

For example, in the example shown in FIG. 1, the information providing device 10 receives a question from the questioner Q1 via the terminal device QT1 used by the questioner Q1 (step S2). In such a case, the information providing device 10 estimates the character of the questioner from the text of the questioner or the text selected by the questioner (step S3). For example, in the example shown in FIG. 1, the information providing device 10 has a question history # Q1 which is a history of questions posted by the questioner Q1 and a selection history # Q1 which is a history of answers selected by the questioner Q1 as the best answer. Etc. are collected. Then, the information providing device 10 estimates the character of the questioner Q1 by inputting a character string such as a word included in the collected question history # Q1 and the selection history # Q1 into the learning model.

Subsequently, the information providing device 10 collects the answer history of each respondent R1 to R3 (step S4). Then, the information providing device 10 estimates the personality of the respondent from the response history of each respondent R1 to R3 (step S5). For example, the information providing device 10 estimates the character of each respondent R1 to R3 by inputting the sentences posted as answers by each respondent R1 to R3 into the learning model.

Subsequently, the information providing device 10 selects a respondent having a personality that matches the personality of the questioner (step S6). Here, the information providing device 10 may match a questioner and a respondent having the same or similar personality, or may match a questioner and a respondent with a previously associated personality. For example, the information providing device 10 may match a questioner who likes a long sentence with a respondent who likes a long sentence. Further, the information providing device 10 may match a long-tempered answerer with a short-tempered questioner, for example.

Then, the information providing device 10 presents to each respondent a question posted by a questioner who exactly matches. For example, when the personality of the questioner Q1 matches the personality of the respondent R3, the information providing device 10 provides information on the question posted by the questioner Q1 (step S7). For example, the information providing device 10 may send information such as a question posted by the questioner Q1 and whether or not to answer it to the terminal device RT3 by various push notifications such as e-mail.

As a result of such processing, the information providing device 10 can prompt the questioner Q1 to register an answer by the respondent R3 having a personality that matches the questioner Q1. As a result, the information providing device 10 can improve the possibility that the questioner Q1 will like the answer given by the respondent R3.

For example, the text of the question posted by the questioner Q1 is considered to indicate the character of the questioner Q1. Therefore, the information providing device 10 estimates the personality of the questioner Q1 from the question history # Q1 of the questioner Q1, and among the respondents R1 to R3, the personality estimated from the answer is the same as or similar to the personality of the questioner Q1. Select a respondent with a personality to do. The information providing device 10 may associate the user R with the questioner Q having a different personality, which is associated in advance, such as matching the patient R with the impatient questioner Q1. As a result of such processing, the information providing device 10 can have the same personality as the questioner Q1 and the respondent who answers the question of the questioner Q1. You can encourage the posting of appropriate answers.

Further, for example, the text of the answer selected by the questioner Q1 as the best answer is considered to indicate the preference of the questioner Q1 in the text. Therefore, the information providing device 10 estimates the personality preferred by the questioner Q1 from the text of the answer that is the best answer selected by the questioner Q1 by using the learning model. That is, the information providing device 10 estimates the personality of the respondent preferred by the questioner Q1. Then, the information providing device 10 estimates the personality of each respondent R1 to R3 from the history of the responses of the respondents R1 to R3, and the estimated personality is the same as or similar to the personality of the respondent preferred by the questioner Q1. Select the respondents to do. Then, the information providing device 10 provides the question of the questioner R1 to the selected respondent. As a result of such processing, the information providing device 10 can post the answer of the sentence expression preferred by the questioner Q1, so that the questioner Q1 can be encouraged to post a new question.

[2. Configuration of information providing device]
Hereinafter, an example of the functional configuration of the above-mentioned information providing device 10 will be described. In the following description, an example of the functional configuration of the information providing device 10 that executes the selection process at the Q & A site will be shown. FIG. 2 is a diagram showing a configuration example of the information providing device according to the first embodiment. As shown in FIG. 2, the information providing device 10 includes a communication unit 20, a storage unit 30, and a control unit 40.

The communication unit 20 is realized by, for example, a NIC (Network Interface Card) or the like. Then, the communication unit 20 is connected to the network N by wire or wirelessly, and transmits / receives information to / from the terminal device T.

The storage unit 30 is realized by, for example, a semiconductor memory element such as a RAM (Random Access Memory) or a flash memory (Flash Memory), or a storage device such as a hard disk or an optical disk. Further, the storage unit 30 stores the user database 31, the posting database 32, and the learning model 33. Hereinafter, an example of information registered in the user database 31 and the posting database 32 will be described with reference to FIGS. 3 and 4.

Information of a user who is a questioner or a respondent is registered in the user database 31. For example, FIG. 3 is a diagram showing an example of information registered in the user database according to the first embodiment. As shown in FIG. 3, information having items such as "user ID (Identifier)", "question history", "answer history", and "personality information" is registered in the user database 31.

Here, the "user ID" is an identifier for identifying a user. The "question history" is a history of questions posted by users. Further, the "answer history" is a history of answers posted by the user. Further, the "personality information" is information indicating the personality of the user diagnosed by various personality diagnosis methods or the personality estimated by using a learning model.

For example, in the example shown in FIG. 3, information such as a user ID "U1", a question history "question history # U1", an answer history "answer history # U1", and personality information "personality # U1" is registered. For such information, the user indicated by the user ID "U1" has posted the question indicated by the question history "question history # U1" and posted the answer indicated by the answer history "answer history # U1". Personality information Indicates that the person has the personality indicated by "personality # U1". In the example shown in FIG. 3, conceptual values such as question history # U1, answer history # U1, and personality # U1 are described, but in reality, the posted question, the character string that becomes the answer, and the personality are described. The indicated character string, numerical value, etc. will be registered. Further, in addition to the information shown in FIG. 3, arbitrary information about the user may be registered in the user database 31.

Various posting information posted by the questioner or the respondent is registered in the posting database 32. For example, FIG. 4 is a diagram showing an example of information registered in the posting database according to the first embodiment. As shown in FIG. 4, information having items such as "post ID", "post type", "title", "post content", "selection information", and "evaluation information" is registered in the post database 32. To.

Here, the "post ID" is an identifier that identifies the post information. Further, the "post type" is information indicating whether the posted information is a question or an answer. Further, the "title" is information indicating a title that becomes a thumbnail of the question. The "posted content" is the text of the posted question or answer. Further, the "selection information" is information indicating the posting ID of the answer selected by the questioner as the best answer among the answers corresponding to the associated questions. Further, "evaluation information" is information indicating the evaluation of each user for a question or an answer.

For example, in the example shown in FIG. 4, the post ID "C1", the post type "question", the title "title # C1", the post content "content # C1", the selection information "C2", and the evaluation information "evaluation # C1". Information such as is registered in association with each other. Such information is the post information that the post indicated by the post ID "C1" is posted as a question, as indicated by the post type "question", the title is "Title # C1", and the post content is "". Content # C1 ”indicates that. Further, as for such information, the answer selected as the best answer in the question indicated by the posting ID "C1" is the answer indicated by the posting ID "C2", and the evaluation information "evaluation # C1" is provided by each user. Indicates that the evaluation has been received.

In the example shown in FIG. 4, conceptual values such as "Title # C1", "Content # C1", and "Evaluation # C1" are described, but in reality, the title character string, question, and answer are described. A character string, a numerical value indicating the number of times a favorable evaluation such as "like" has been performed, etc. will be registered. Further, in addition to the information shown in FIG. 4, for example, the user ID of the user who made the posting and arbitrary information regarding various postings may be registered in the posting database 32.

The explanation will be continued by returning to FIG. The control unit 40 is, for example, a controller, and various programs stored in a storage device inside the information providing device 10 by a CPU (Central Processing Unit), an MPU (Micro Processing Unit), or the like have a RAM as a work area. It is realized by executing as. Further, the control unit 40 is a controller, and is realized by, for example, an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array).

As shown in FIG. 2, the control unit 40 includes a collection unit 41, a learning unit 42, a reception unit 43, an estimation unit 44, a selection unit 45, and a providing unit 46. The collecting unit 41 collects learning data for learning the learning model.

For example, the collection unit 41 includes search queries entered by each user in the past, question texts and answer texts posted by users, answer texts selected by users as the best answer, and user-friendly evaluations. Collect the texts of the questions and answers that you asked as a past log. In addition, the collection unit 41 collects various UGC information as a past log.

Subsequently, the collection unit 41 collects the personality diagnosis results of each user. For example, the collection unit 41 proposes to each user to perform various personality diagnoses such as Big5, and accepts the registration of the diagnosis result. The collecting unit 41 may estimate the personality diagnosis result of each user based on the usage history of the web service on the fortune-telling site or the like. Then, the collecting unit 41 registers the personality diagnosis result in the user database 31.

The learning unit 42 learns the learning model 33. For example, the learning unit 42 classifies each user into categories according to the personality diagnosis result, and extracts characteristic words from the past logs of the users classified into each category. Then, the learning unit 42 learns the learning model 33 by using a technique such as backpropagation so as to output a character string or a numerical value indicating the user's personality diagnosis result when the extracted feature word is input. conduct. That is, the learning unit 42 learns the co-occurrence between the information input by the user and the personality of the user who has input the information and is pre-diagnosed by a predetermined personality diagnosis method. Generate.

The reception unit 43 accepts registration of new questions and answers. For example, when the reception unit 43 receives a sentence to be a question from the terminal device T used by the questioner, the reception unit 43 registers the received sentence as a question in the posting database 32. Further, when the reception unit 43 receives the text to be the answer from the terminal device T used by the respondent, the reception unit 43 registers the received text in the posting database 32 in association with the posting ID of the corresponding question.

The estimation unit 44 estimates the personality of the user based on the information input by the user. For example, the estimation unit 44 refers to the user database 31 and selects a user whose personality diagnosis result is not registered. Subsequently, the estimation unit 44 estimates the character of the user based on the characteristics of the character string input by the selected user. More specifically, the estimation unit 44 uses a learning model that learns the relationship between the characteristics of the input information and the personality of the user who input the information, and the information input by the user. From, the personality of the user is estimated.

For example, the estimation unit 44 refers to the user database 31 and the posting database 32, and collects sentences of questions and answers posted by the user whose personality is to be estimated. In addition, the estimation unit 44 collects the text of the answer selected by the user as the best answer. Then, the estimation unit 44 inputs the collected sentences into the learning model 33, and estimates the personality of the user according to the output of the learning model 33.

By such processing, the estimation unit 44 can estimate the character of the user as a respondent, for example, based on the sentence input by the user as the respondent. Further, the estimation unit 44 is, for example, from the text input by the user as a questioner or the text of the answer selected by the user as the questioner and selected as the best answer, by the user as the questioner. You can estimate your personality. That is, the estimation unit 44 can also estimate the character of the provider based on the history of the questions posted by the provider, and among the answers to the questions posted by the provider, the answer selected by the provider. It is also possible to estimate the personality of the respondents preferred by the provider based on the history of.

Then, the estimation unit 44 registers the estimation result in the user database 31. The estimation unit 44 may execute the estimation of the personality of the user at an arbitrary timing. For example, the estimation unit 44 may update the user database 31 at predetermined time intervals. Further, the estimation unit 44 may estimate a new personality not only for a user whose personality has not been estimated, but also for a user whose personality has been estimated for a predetermined period of time, for example. .. In addition, the estimation unit 44 may exclude users whose personality has been diagnosed by various personality diagnosis methods from the estimation target.

The selection unit 45 selects the provision target provided to the user. For example, the selection unit 45 selects a provision target provided by a provider having a personality corresponding to the estimated user personality as a provision target provided to the user.

For example, when the selection unit 45 receives the delivery request of the posted question from the terminal device T, the selection unit 45 identifies the user who is the delivery request source, that is, the respondent who is the information providing destination, and the identified respondent. The personality diagnosed by the personality diagnosis method or the estimated personality is read from the user database 31. Subsequently, the selection unit 45 extracts various questions from the posting database 32, and identifies the personality of the user who posted each of the extracted questions, that is, the questioner who is the information provider from the user database 31. .. For example, the selection unit 45 identifies the personality diagnosed by various personality diagnosis methods or the personality estimated by the estimation unit 44 as the personality of the questioner.

Then, the selection unit 45 compares the personality of each questioner with the personality of the respondent, and identifies the questioner having the personality corresponding to the personality of the respondent. Then, the selection unit 45 selects the question posted by the specified questioner as the distribution target. For example, the selection unit 45 selects a question posted by a questioner having a personality similar to that of the respondent or a questioner having a personality previously associated with the personality of the respondent as a distribution target.

The providing unit 46 provides the respondent with information about the question selected by the selection unit 45. For example, the providing unit 46 generates content indicating a thumbnail or the like of the question selected by the selection unit 45, and distributes the generated content to the terminal device T.

[3. Processing procedure]
Next, the procedure of the learning process by the information providing apparatus 10 according to the embodiment will be described with reference to FIG. FIG. 5 is a flowchart showing an example of the flow of the learning process according to the first embodiment. For example, the information providing device 10 collects the personality diagnosis result of the user (step S101). Then, the information providing device 10 extracts a feature word from the past log of each user for each personality category (step S102), and makes the model learn the relationship between the feature word and the personality category (step S103). , End the process.

Next, an example of the flow of the selection process executed by the information providing device 10 will be described with reference to FIG. FIG. 6 is a flowchart showing an example of the flow of the selection process according to the first embodiment. As shown in FIG. 6, the information providing device 10 estimates the personality of the questioner from the history of the answers selected by the questioner using the learning model (step S201). Subsequently, the information providing device 10 estimates the personality of the respondent from the response history of the respondent (step S202). Then, the information providing device 10 selects a questioner having a personality corresponding to the character of the predetermined respondent from the questioners (step S203), and provides the question of the selected questioner to the respondent (step S203). Step S204), the process is terminated. The information providing device 10 may change the order in which the steps S201 and S202 are executed. It may be executed at any time.

[4. Modification example]
The information providing device 10 described above may be implemented in various different forms other than the above embodiment. Therefore, another embodiment of the information providing device 10 will be described below.

[4-1. Execution timing]
In the above description, the information providing device 10 executes the selection process when the question delivery request is received, and delivers the question by the questioner having a personality that matches the personality of the respondent. However, the embodiments are not limited to this. The information providing device 10 may execute the above-mentioned selection process at an arbitrary execution timing.

For example, the information providing device 10 selects a question posted by a questioner having a personality corresponding to each user's personality at a predetermined time interval, and provides each user with content that proposes an answer to the selected question. May be provided. Further, when a new question is posted, the information providing device 10 selects a user corresponding to the character of the questioner as a respondent, and answers the new question to the selected respondent. Content that suggests doing so may be delivered.

That is, if the personality of the questioner and the personality of the respondent can be matched, the information providing device 10 executes the above-mentioned selection process at an arbitrary timing, and the question or answer selected as a result of the execution. Should be provided to each user.

[4-2. About the target of selection processing]
Here, in addition to the above-mentioned processing, the information providing device 10 corresponds to the character of the information application destination (for example, the respondent who provides the answer) among the information providing sources (for example, the questioner who provides the question). If the information provided by the personality information provider is provided to the information provider, the above-mentioned selection process may be executed in any process. Further, if the information providing device 10 provides the information provided by the information providing source to the information providing destination when the character of the information providing destination and the character of the information providing source correspond to each other, the information providing device 10 is the information providing source. The information provider may be selected using the personality as a key, or the information provider may be selected using the personality of the information provider as a key.

Here, regarding what kind of personality is to be matched, what is the type and content of the information provided from the information provider to the information provider, the content of the service provided by the information providing device 10, and the learning model. It is considered that it will change as appropriate depending on whether the information is to be learned or what kind of information is input by the information provider or the information provider to estimate the personality. Therefore, the information providing device 10 may execute the above-mentioned selection process based on the types and contents of the provided information, the contents of the service, and the like, based on the modes of the preset personalities.

For example, in SNS and the like, other second users may give various comments to posts posted by the first user. In such an SNS, the first user can be regarded as the same information provider as the questioner on the Q & A site, and the second user can be regarded as the same information provider as the respondent on the Q & A site. Can be considered. Therefore, the information providing device 10 may select the posting of the first user to be provided to the second user based on the matching result of the personality.

For example, considering the post on the SNS and the comment on the post, it is considered that the personality of the first user and the personality of the second user should be similar. Therefore, the information providing device 10 estimates the personality of the first user based on the posting history and the comment history of the first user, and the second user is based on the posting history and the comment history of the second user. 2 Estimate the personality of the user. Then, the information providing device 10 may notify the second user of the posting of the first user when the personality of the first user and the personality of the second user are similar.

In addition, in SNS, it may be possible to evaluate posts, comments, and the like as "likes". Posts and comments that have been evaluated as "likes" are not sentences posted by each user, but can be said to be comments that indicate the user's preference. Therefore, the information providing device 10 estimates the personality of the first user from the posts and comments that the first user has given a favorable evaluation such as "like", that is, the posts and comments selected by the first user. At the same time, the personality of the second user is estimated from the posts and comments selected by the second user. Then, the information providing device 10 may provide the post of the first user to the second user when the estimated personality matches.

In addition, on the Q & A site, a summary of the question or a thumbnail containing a part of the question is provided to the respondent, and when the respondent selects one of the thumbnails, the selected thumbnail and the corresponding question are displayed. There are cases. Further, on various portal sites and news sites, thumbnails of contents are presented to users, and contents corresponding to the thumbnails selected by the users are provided.

In this way, when thumbnails are provided prior to the content, the character of the information provider is estimated based on the thumbnail, not the content itself, and matching is performed using the estimated character to determine the thumbnail selection frequency. It is also considered that the frequency of browsing the content by the user to whom the information is provided can be improved.

Therefore, the information providing device 10 is provided by a provider who estimates the character of the user based on the summary selected by the user among the summaries of the contents and has the property corresponding to the estimated property of the user. Content summaries may be selected for delivery to users. For example, the information providing device 10 identifies a thumbnail selected by a user who is an information providing destination among thumbnails including a summary of contents such as various questions and news and some sentences. Then, the information providing device 10 estimates the personality of the user who is the information providing destination based on the text included in the specified thumbnail, and the thumbnail of the content posted by the information provider having the estimated personality and the corresponding personality. Should be selected as the delivery target.

Here, the user who is the information provider, that is, the provider does not necessarily have to be a specific person. For example, news texts and the like may reflect the preferences of a plurality of people. Therefore, the information providing device 10 estimates the property estimated from a certain content or the summary of the content as the character of the virtual information provider that provides the content or the summary of the content, and has a character that matches the estimated character. Content, summaries, etc. may be provided to the information provider.

[Second Embodiment]
[5-1. About advertisement delivery]
The information providing device 10 may apply the above-mentioned selection process to so-called advertisement distribution. For example, the information providing device 10a estimates the character of the creator who created the advertisement and the virtual information provider who provides the advertisement from the text or the like included in the advertisement to be delivered. Then, the information providing device 10a may select an advertisement whose personality presumed to be the personality of the user to be delivered is matched as the delivery target.

For example, FIG. 7 is a diagram showing an example of a selection process executed by the information providing device according to the second embodiment when delivering an advertisement. In the example shown in FIG. 7, an example of a selection process for selecting an advertisement to be provided to the user U1 from the advertisements # A1 to A3 provided by the plurality of providers P1 to P3 has been described. Here, the providers P1 to P3 are personas of the designer who created the advertisements # A1 to A3 and the copywriter who created the advertisement wording. For example, the provider P1 is a designer who created a sentence, a catch phrase, etc. included in an advertisement.

When delivering such advertisements # A1 to A3 to the user U1, the information providing device 10a does not match the advertisements # A1 to A3 themselves with the attributes of the user U1 and the like, but the advertisements # A1 to The advertisement to be delivered is selected based on the matching result between the characteristics of the providers P1 to P3 appearing in A3 and the characteristics of the user U1. For example, the information providing device 10a receives advertisements # A1 to A3 created by each provider P1 to P3 from the terminal devices PT1 to PT3 used by each provider (step S1). In such a case, the information providing device 10a estimates the character of each provider P1 to P3 from the advertisement (step S2).

For example, the information providing device 10a inputs the sentences included in the advertisements # A1 to A3 into the learning model learned by the learning process described above, and estimates the character of the advertisement providers P1 to P3. The information providing device 10a is a learning model in which learning is performed so as to output the personality diagnosis result of the user U, for example, when the user U inputs a still image or a moving image of an advertisement selected in the past. May be used to estimate the character of the providers P1 to P3 from the advertisements # A1 to A3.

Subsequently, the information providing device 10a collects the past log # U1 of the user U1 (step S3). For example, the information providing device 10a collects the history of advertisements selected by the user U1, the history of posted information posted by the user U1, search queries, and the like. Then, the information providing device 10a estimates the character of the user U1 from the past log # U1 of the user U1 using the learning model (step S4).

Then, the information providing device 10a selects an advertisement provided by a provider having a character corresponding to the character of the user U1 as a distribution target (step S5). For example, when the personality of the user U1 and the personality of the provider P2 are similar, the information providing device 10a selects the advertisement # A2 provided by the provider P2 as a target to be delivered to the user U1. do. Then, the information providing device 10a provides the advertisement # A2 to the user U1 (step S6).

As described above, when the information providing device 10a distributes the advertisement as information to the information providing destination, the information providing device 10a estimates the character of the information providing source from the advertisement to be provided. Further, the information providing device 10a estimates the character of what kind of advertisement the information providing destination prefers from the history of the advertisement selected by the information providing destination. Then, the information providing device 10a selects an advertisement to be delivered to the information providing destination according to the matching result of the estimated personality. As a result of such processing, the information providing device 10a can select an advertisement that is easier for the user to like as a distribution target.

[5-2. Configuration of information providing device]
Hereinafter, an example of the functional configuration of the above-mentioned information providing device 10a will be described. In the following description, an example of the functional configuration of the information providing device 10a that executes the selection process at the time of advertisement distribution will be shown. FIG. 8 is a diagram showing a configuration example of the information providing device according to the second embodiment. As shown in FIG. 8, the information providing device 10a includes a communication unit 20, a storage unit 30a, and a control unit 40a.

Similar to the storage unit 30, the storage unit 30a is realized by a semiconductor memory element such as a RAM or a flash memory, or a storage device such as a hard disk or an optical disk. Further, the storage unit 30a stores the user database 31a, the advertisement database 32a, and the learning model 33a. Hereinafter, an example of information registered in the user database 31a and the advertisement database 32a will be described with reference to FIGS. 9 and 10.

Information of the user who is the advertisement delivery destination is registered in the user database 31a. For example, FIG. 9 is a diagram showing an example of information registered in the user database according to the second embodiment. As shown in FIG. 9, information having items such as "user ID", "selection history", and "personality information" is registered in the user database 31a.

Here, the "selection history" is information indicating the history of the advertisement selected by the user. For example, in the example shown in FIG. 9, information such as the user ID "U1", the selection history "selection history # U1", and the personality information "personality # U1" is registered. For such information, the advertisement indicated by the selection history "selection history # U1" is selected by the user indicated by the user ID "U1", and the user has the personality indicated by the personality information "personality # U1". Show that.

In the example shown in FIG. 9, a conceptual value such as selection history # U1 is described, but in reality, an identifier or the like of an advertisement selected by the user in the past is registered. Further, in addition to the information shown in FIG. 9, arbitrary information about the user may be registered in the user database 31a.

Information about the advertisement to be delivered is registered in the advertisement database 32a. For example, FIG. 10 is a diagram showing an example of information registered in the advertisement database according to the second embodiment. As shown in FIG. 10, information having items such as "advertisement ID", "title", and "provider personality information" is registered in the advertisement database 32a.

Here, the "advertisement ID" is an identifier for identifying an advertisement. Further, the "title" is information indicating various character strings included in the advertisement. Further, the "provider personality information" is information indicating the personality estimated as the personality of the provider who provided the advertisement. For example, in the example shown in FIG. 10, the advertisement ID "A1", the title "Title # A1", and the provider personality information "Personality # A1" are registered in association with each other. Such information includes the text indicated by the title "Title # A1" in the advertisement indicated by the advertisement ID "A1", and the personality presumed as the personality of the provider of this advertisement is the provider personality information "personality". It indicates that it is "# A1".

In the example shown in FIG. 10, conceptual values such as "Title # A1" and "Personality # A1" are described, but in reality, character strings and values indicating the character strings and characters appearing in the advertisement are registered. The Rukoto. Further, in addition to the information shown in FIG. 10, it is assumed that contents such as still images and moving images constituting the advertisement to be delivered are registered in the advertisement database 32a. Further, various information related to the advertisement may be registered in the advertisement database 32a, such as targeting information indicating the attribute of the user who is the delivery destination.

The explanation will be continued by returning to FIG. The control unit 40a is, for example, a controller, and is realized by executing various programs stored in the storage device inside the information providing device 10a using the RAM as a work area by a CPU, an MPU, or the like. Further, the control unit 40a is a controller, and is realized by, for example, an integrated circuit such as an ASIC or FPGA.

As shown in FIG. 8, the control unit 40a includes a collection unit 41a, a learning unit 42a, a reception unit 43a, an estimation unit 44a, a selection unit 45a, and a providing unit 46a.

Similar to the collection unit 41, the collection unit 41a collects learning data for learning the learning model 33a. For example, the collection unit 41a collects past logs showing advertisements selected by each user in the past, and also collects personality diagnosis results of each user.

The learning unit 42a learns the learning model 33a. For example, the learning unit 42a classifies each user into categories according to the personality diagnosis result, and features from the title of the advertisement (that is, the text included in the advertisement) selected in the past by the users classified into each category. Extract words. Then, the learning unit 42a learns the learning model 33a by using a technique such as backpropagation so as to output a character string or a numerical value indicating the user's personality diagnosis result when the extracted feature word is input. conduct. The learning unit 42a may learn the learning model 33a so as to output the personality diagnosis result of the user when the text included in the advertisement or the image itself of the advertisement is input.

The reception unit 43a accepts the registration of a new advertisement. For example, when the reception unit 43a receives various information of the advertisement to be delivered from the advertiser, the advertisement agency, or the like, the reception unit 43a registers the received information in the advertisement database 32a.

The estimation unit 44a estimates the character of the user to whom the advertisement is provided and the character of the provider of the advertisement. For example, the estimation unit 44a refers to the user database 31a and selects a user whose personality diagnosis result is not registered. Subsequently, the estimation unit 44a estimates the personality of the user based on the history of the advertisement selected by the selected user in the past. More specifically, the estimation unit 44a estimates the personality of the user by inputting the text included in the advertisement selected by the user into the learning model 33a. Further, the estimation unit 44a estimates the character of the advertisement provider by inputting the text included in the advertisement to be delivered into the learning model 33a.

The selection unit 45a selects an advertisement to be provided to the user. For example, when the selection unit 45a receives the advertisement distribution request from the terminal device T, the selection unit 45a refers to the user database 31a and specifies the character of the user who uses the distribution request source terminal device T. Subsequently, the selection unit 45a refers to the advertisement database 32a and selects an advertisement provided by a provider having a personality corresponding to the personality of the user as a distribution target among the advertisements.

The providing unit 46a provides the advertisement selected by the selection unit 45a. For example, the providing unit 46a distributes various data of the advertisement selected by the selecting unit 45a to the terminal device T.

As described above, the information providing device 10a selects the content created by the provider having the property corresponding to the estimated user property among the contents related to the advertisement as the content to be provided to the user. do. As a result, since the information providing device 10a can deliver an advertisement that matches the personality of the user, it is possible to increase the chances of selecting an advertisement without damaging the impression of the user.

[Third Embodiment]
[6-1. About inquiries]
The information providing device 10 may apply the above-mentioned selection process to the selection of an operator who responds to the user when receiving an inquiry from the user by telephone or the like. For example, the information providing device 10b estimates the nature of the user based on the information input by the user when receiving the provision of the predetermined service, and among the providers providing the predetermined service, the estimated use. You may select the offering target provided by the provider who has the property corresponding to the nature of the person. That is, the information providing device 10b may select a provider whose personality matches the personality of the user and provide the information provided by the selected provider to the user.

For example, FIG. 11 is a diagram showing an example of a selection process executed when the information providing device according to the third embodiment receives an inquiry. In the example shown in FIG. 11, an example of a process of selecting which of a plurality of respondents (for example, operators) R1 to R3 responds to the questioner Q1 who makes various inquiries by telephone will be described. ..

For example, the information providing device 10b receives an initial utterance from the questioner Q1 (step S1). In such a case, the information providing device 10b estimates the character of the questioner from the initial utterance # Q1 of the questioner Q1 (step S2). That is, the information providing device 10b estimates the nature of the user based on the utterance of the user made before receiving the provision of the predetermined service. For example, the information providing device 10b converts the initial utterance # Q1 of the questioner Q1 into a character string, and estimates the character of the questioner Q1 using the learning model learned by the learning process described above.

Then, the information providing device 10b selects a respondent having a personality that matches the personality of the questioner (step S3). For example, the information providing device 10b selects the respondent R3 when the personality of the questioner Q1 and the personality of the respondent R3 correspond to each other. Then, the information providing device 10b transfers the call from the questioner Q1 to the respondent whose personality matches (step S4). As a result of such processing, the information providing device 10b can provide the inquirer Q1 with a response by the respondent R3 whose personality corresponds to the personality of the questioner Q1.

Here, any response can be adopted as to what kind of respondent is matched with what kind of questioner. For example, when the information providing device 10b has a high seriousness score in Big5 as the character of the questioner, the respondent may similarly select a respondent having a high seriousness score in Big5. Further, when the information providing device 10b has a high score of mental instability in Big5 as the personality of the questioner, even if a respondent with a high degree of kindness in the egogram or a veteran respondent with a long working period is selected. good. Further, the information providing device 10b may select a respondent who has a high degree of cooperation in Big5 when the character of the questioner has a high degree of negativity to others in the OK gram. Further, the information providing device 10b may select a respondent having a high degree of affirmation of others in the OK gram when the questioner has a high degree of freedom in the egogram. Further, the information providing device 10 may select a respondent having an arbitrary personality or a new respondent for the questioner who has a high affirmation of others in the OK gram as the personality of the questioner.

[6-2. About personality estimation]
Here, the information providing device 10b may estimate the character of the questioner by using the learning model in which the learning is performed by the learning process described above. However, in order to estimate the personality of the questioner more accurately, the information providing device 10b has characteristics of the voice generated by the user, such as the personality diagnosis result of the user and the suppression and speed of the voice of the user. The personality of the questioner may be estimated using a learning model that has learned the relationship with.

Further, the information providing device 10b may use a learning model in which the characteristics of the user's voice are learned and a learning model in which the characteristics of the text input by the user are learned in combination. For example, the information providing device 10b estimates the character of the questioner as well as the probability that the questioner has that character by using a learning model that learns the characteristics of the text input by the user. Then, when the estimated accuracy is lower than a predetermined threshold value, the information providing device 10b may estimate the personality of the questioner by using a learning model that learns the characteristics of the user's voice. Further, the information providing device 10b may estimate the personality of the questioner by using the personality and the accuracy estimated by each learning model in an integrated manner.

[6-3. Configuration of information providing device]
Hereinafter, an example of the functional configuration of the above-mentioned information providing device 10b will be described. In the following description, an example of the functional configuration of the information providing device 10b that executes the selection process when an inquiry is received will be shown. FIG. 12 is a diagram showing a configuration example of the information providing device according to the third embodiment. As shown in FIG. 12, the information providing device 10b has a communication unit 20, a storage unit 30b, and a control unit 40b.

Similar to the storage unit 30, the storage unit 30b is realized by a semiconductor memory element such as a RAM or a flash memory, or a storage device such as a hard disk or an optical disk. Further, the storage unit 30b stores the operator database 31b and the learning model 33b. Hereinafter, an example of the information registered in the operator database 31b will be described with reference to FIG. 13.

Information about respondents is registered in the operator database 31b. For example, FIG. 13 is a diagram showing an example of information registered in the operator database according to the third embodiment. As shown in FIG. 13, information having items such as "operator ID" and "personality information" is registered in the operator database 31b.

Here, the "operator ID" is an identifier for identifying the respondent corresponding to the question from the questioner, that is, the operator. For example, in the example shown in FIG. 13, the operator ID “O1” and the personality information “personality # 01” are registered in association with each other. Such information indicates that the personality of the operator indicated by the operator ID "O1" is "personality # O1".

In the example shown in FIG. 13, a conceptual value such as personality # O1 is described, but in reality, various character strings and numerical values indicating the personality of the operator are registered. In addition to the information shown in FIG. 13, arbitrary information about the operator, such as the gender and years of service of the operator, may be registered in the operator database 31b.

The explanation will be continued by returning to FIG. The control unit 40b is, for example, a controller, and is realized by executing various programs stored in the storage device inside the information providing device 10b using the RAM as a work area by a CPU, an MPU, or the like. Further, the control unit 40b is a controller, and is realized by, for example, an integrated circuit such as an ASIC or FPGA.

As shown in FIG. 12, the control unit 40b includes a collection unit 41b, a learning unit 42b, a reception unit 43b, an estimation unit 44b, a selection unit 45b, and a providing unit 46b.

Similar to the collection unit 41, the collection unit 41b collects learning data for learning the learning model 33b. For example, the collection unit 41b collects the utterance history of each user and also collects the personality diagnosis result of the user.

The learning unit 42b learns the learning model 33b. For example, the learning unit 42b classifies each user into categories according to the personality diagnosis result, and extracts characteristic words included in the utterances of the users classified into each category. Then, the learning unit 42b learns the learning model 33b by using a technique such as backpropagation so as to output a character string or a numerical value indicating the user's personality diagnosis result when the extracted feature word is input. conduct. The learning unit 42b may learn the learning model 33b so as to output the personality diagnosis result of the user when the voice of the user's utterance is input.

The reception unit 43b accepts the registration of the operator. For example, when the reception unit 43b accepts the registration of a new operator, the reception unit 43b registers the results of various personality diagnoses by the operator in the operator database 31b together with the operator ID.

The estimation unit 44b estimates the character of the user to whom the service is provided. For example, when a call is received from the questioner, the estimation unit 44b urges the questioner to speak the content of the question by voice or the like by automatic voice or the like. Then, the estimation unit 44b estimates the character of the questioner from the utterance of the questioner using the learning model 33b.

The selection unit 45b selects an operator who provides a service to the questioner. For example, the selection unit 45b refers to the operator database 31b and selects an operator having a personality corresponding to the personality of the questioner estimated by the estimation unit 44b.

The providing unit 46b provides the questioner with information provided by the operator selected by the selection unit 45b. For example, the providing unit 46b connects the telephone from the questioner to the operator selected by the selection unit 45b.

In this way, the information providing device 10b estimates the nature of the user based on the information input by the user when receiving the provision of the predetermined service, and is estimated among the providers who provide the predetermined service. Since the provision target provided by the provider having the property corresponding to the property of the user is selected, it is possible to realize, for example, a response in an appropriate manner to the questioner who asks a question by telephone.

[Other embodiments]
[7-1. Information to be provided]
In the above-mentioned example, the information providing devices 10, 10a and 10b (hereinafter, collectively referred to as "information providing device 10") are a question provided to the respondent, an advertisement provided to the user, or a question. The operator who provides the service to the person was selected. However, the embodiments are not limited to this. When providing arbitrary information to an information provider, the information providing device 10 may select information provided by an information provider having a personality corresponding to the personality of the information provider.

For example, when a plurality of answers are posted to a certain question, the information providing device 10 estimates the personality of the respondent from each answer and is posted by the respondent having the personality corresponding to the personality of the questioner. Only the answer may be provided to the questioner. Further, the information providing device 10 may estimate the personality of the questioner from the text input by the questioner on the inquiry bulletin board or the like, and change the content of the response according to the estimated personality. For example, the information providing device 10 may provide a reply based on a reply sentence policy (template, guideline) according to the character of the questioner.

Further, the information providing device 10 selects and selects an information providing source in consideration of various attributes (demographic attribute and psychographic attribute) of the user in addition to the character of the user who is the information providing destination. Information provided by the information provider may be provided. As described above, if the information providing device 10 at least estimates the personality of the user to whom the information is provided and selects the information provided by the information provider having the personality corresponding to the estimated personality as the provision target. , Arbitrary processing may be executed.

The information providing device 10 may disperse the matching results. For example, the information providing device 10 estimates the personality of the respondent on the Q & A site, and the ratio of the respondents to whom a certain question is provided is the ratio of the respondents who match the personality of the questioner to the respondents who do not match. It may be selected so as to have a predetermined ratio. Further, in the information providing device 10, the ratio of the question provided to the respondent by the questioner having a personality that matches the personality of the respondent and the question by the questioner having a personality that does not match the personality of the respondent. You may select the question so that is a predetermined ratio.

Further, the information providing device 10 may update the learning model based on the evaluation of each question or answer. For example, the information providing device 10 may estimate the character of the questioner by using a learning model that learns the characteristics of the answer whose evaluation exceeds a predetermined threshold among the answers selected by the questioner as the best answer. Further, the information providing device 10 may estimate the personality of the respondent by using a learning model in which the respondent has learned the characteristics of the question and the answer that have been positively evaluated. Further, the information providing device 10 may estimate the personality of the user by using a learning model that learns the characteristics of the advertisement whose click rate exceeds a predetermined threshold among the advertisements selected by the user.

[7-2. others〕
Further, among the processes described in the above-described embodiment, all or a part of the processes described as being automatically performed can be manually performed, and conversely, the processes described as being manually performed. It is also possible to automatically perform all or part of the above by a known method. In addition, information including processing procedures, specific names, various data and parameters shown in the above documents and drawings can be arbitrarily changed unless otherwise specified. For example, the various information shown in each figure is not limited to the information shown in the figure.

Further, each component of each of the illustrated devices is a functional concept, and does not necessarily have to be physically configured as shown in the figure. That is, the specific form of distribution / integration of each device is not limited to the one shown in the figure, and all or part of them may be functionally or physically distributed / physically in any unit according to various loads and usage conditions. Can be integrated and configured.

In addition, the above-described embodiments can be appropriately combined as long as the processing contents do not contradict each other.

[7-3. program〕
Further, the information providing device 10 according to the above-described embodiment is realized by, for example, a computer 1000 having a configuration as shown in FIG. FIG. 14 is a diagram showing an example of a hardware configuration. The computer 1000 is connected to the output device 1010 and the input device 1020, and the arithmetic unit 1030, the primary storage device 1040, the secondary storage device 1050, the output IF (Interface) 1060, the input IF 1070, and the network IF 1080 are connected by the bus 1090. Have.

The arithmetic unit 1030 operates based on a program stored in the primary storage device 1040 or the secondary storage device 1050, a program read from the input device 1020, or the like, and executes various processes. The primary storage device 1040 is a memory device that temporarily stores data used by the arithmetic unit 1030 for various operations such as RAM. Further, the secondary storage device 1050 is a storage device in which data used by the calculation device 1030 for various calculations and various databases are registered, such as a ROM (Read Only Memory), an HDD (Hard Disk Drive), and a flash memory. Is realized by.

The output IF 1060 is an interface for transmitting information to be output to an output device 1010 that outputs various information such as a monitor and a printer. For example, USB (Universal Serial Bus), DVI (Digital Visual Interface), and the like. It is realized by a connector of a standard such as HDMI (registered trademark) (High Definition Multimedia Interface). Further, the input IF 1070 is an interface for receiving information from various input devices 1020 such as a mouse, a keyboard, a scanner, and the like, and is realized by, for example, USB.

The input device 1020 is, for example, an optical recording medium such as a CD (Compact Disc), a DVD (Digital Versatile Disc), a PD (Phase change rewritable Disk), a magneto-optical recording medium such as an MO (Magneto-Optical disk), or a tape. It may be a device that reads information from a medium, a magnetic recording medium, a semiconductor memory, or the like. Further, the input device 1020 may be an external storage medium such as a USB memory.

The network IF 1080 receives data from another device via the network N and sends it to the arithmetic unit 1030, and also transmits the data generated by the arithmetic unit 1030 to the other device via the network N.

The arithmetic unit 1030 controls the output device 1010 and the input device 1020 via the output IF 1060 and the input IF 1070. For example, the arithmetic unit 1030 loads a program from the input device 1020 or the secondary storage device 1050 onto the primary storage device 1040, and executes the loaded program.

For example, when the computer 1000 functions as the information providing device 10, the arithmetic unit 1030 of the computer 1000 realizes the function of the control unit 40 by executing the program loaded on the primary storage device 1040.

[8. effect〕
As described above, the information providing device 10 is provided by a provider who estimates the characteristics of the user based on the information input by the user and has properties corresponding to the estimated characteristics of the user. Select the provision target as the provision target provided to the user. As a result of such processing, the information providing device 10 can provide information that matches the nature of the user, so that appropriate information can be provided to the user.

Further, the information providing device 10 estimates the property of the provider based on the information input by the provider, and selects the provision target provided by the provider whose estimated property corresponds to the property of the user. .. As a result of such processing, the information providing device 10 can provide the user with the information provided by the provider whose property matches the property of the user.

Further, the information providing device 10 estimates the characteristics of the user based on the characteristics of the character string input by the user. Further, the information providing device 10 uses the information input by the user by using a learning model that learns the relationship between the characteristics of the input information and the nature of the user who input the information. Estimate the nature of the person. Further, as a learning model, the information providing device 10 provides co-occurrence between the characteristics of the input information and the characteristics of the user who input the information, which are pre-diagnosed by a predetermined property diagnosis method. Estimate the nature of the user using the learned learning model. As a result of such processing, the information providing device 10 can accurately estimate the nature of the user.

Further, the information providing device 10 estimates the nature of the user based on the history of the answers posted by the user as the answer to the question posted by another user, and is estimated among the posted questions. Select the question posted by the provider corresponding to the nature of the user as the target to be provided to the user. For example, the information providing device 10 estimates the nature of the provider based on the history of the questions posted by the provider, and among the posted questions, the estimated nature corresponds to the nature of the user. Select the question posted by the user as the target to be provided to the user. Further, the information providing device 10 estimates the nature of the respondent preferred by the provider based on the history of the answers selected by the provider among the answers to the questions posted by the provider, and the posted question. Among them, the question posted by the provider whose estimated property corresponds to the property of the user is selected as the target to be provided to the user. As a result of such processing, the information providing device 10 can promote the use of the so-called Q & A site as a result of prompting the questioner to post an answer presumed to be liked.

Further, the information providing device 10 estimates the characteristics of the user based on the summary selected by the user among the summaries of the predetermined contents, and is provided by the provider having the characteristics corresponding to the estimated characteristics of the user. Select the summary of the content to be provided as the target to be provided to the user. As a result of such processing, the information providing device 10 can provide a summary of contents that the user is presumed to like.

Further, the information providing device 10 selects, among the contents related to the advertisement, the content created by the provider having the property corresponding to the estimated user property as the provision target to be provided to the user. Therefore, the information providing device 10 can provide an advertisement that is presumed to be liked by the user.

Further, the information providing device 10 estimates the nature of the user based on the information input by the user when receiving the provision of the predetermined service, and is the estimated use among the providers who provide the predetermined service. Select the offering target provided by the provider who has the property corresponding to the nature of the person. For example, the information providing device 10 estimates the nature of the user based on the utterance of the user made before receiving the provision of the predetermined service. Further, the information providing device 10 estimates the nature of the user based on the characteristics of the voice spoken by the user. Therefore, the information providing device 10 can realize, for example, in various inquiry services and the like, the provision of services by a provider whose personality matches the user.

The embodiments of the present application have been described in detail with reference to the drawings, but this is an example, and various modifications and improvements are made based on the knowledge of those skilled in the art, including the embodiments described in the disclosure column of the invention. It is possible to carry out the present invention in other forms described above.

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

10, 10a, 10b Information providing device 20 Communication unit 30, 30a, 30b Storage unit 31, 31a User database 31b Operator database 32 Posting database 32a Advertising database 33, 33a, 33b Learning model 40, 40a, 40b Control unit 41, 41a , 41b Collection unit 42, 42a, 42b Learning unit 43, 43a, 43b Reception unit 44, 44a, 44b Estimating unit 45, 45a, 45b Selection unit 46, 46a, 46b Providing unit QT1, RT1 to RT3, PT1 to PT3, UT1 , T terminal device

Claims (20)

An estimation unit that estimates the nature of the user based on the information entered by the user,
It has a selection unit that selects a provision target provided by a provider having properties corresponding to the characteristics of the user estimated by the estimation unit as a provision target provided to the user.
The estimation unit uses a learning model that learns the relationship between the characteristics of the input information and the nature of the user who input the information, and uses the information input by the user. Estimate the nature of a person
A selection device characterized by that. The estimation unit estimates the nature of the provider based on the information input by the provider.
The selection device according to claim 1, wherein the selection unit selects a provision target provided by a provider whose properties estimated by the estimation unit correspond to the properties of the user. The selection device according to claim 1 or 2, wherein the estimation unit estimates the characteristics of the user based on the characteristics of the character string input by the user. As the learning model, the estimation unit learns the co-occurrence between the characteristics of the input information and the properties of the user who input the information, which are pre-diagnosed by a predetermined property diagnosis method. The selection device according to any one of claims 1 to 3 , wherein the learning model is used to estimate the nature of the user. An estimation unit that estimates the nature of the user based on the information entered by the user,
A selection unit that selects a provision target provided by a provider having properties corresponding to the characteristics of the user estimated by the estimation unit as a provision target provided to the user.
Have,
The estimation unit estimates the nature of the user based on the history of the answers posted by the user as the answer to the question posted by the other user.
Among the posted questions, the selection unit selects the question posted by the provider corresponding to the nature of the user estimated by the estimation unit as the provision target to be provided to the user. A selection device characterized by. The estimation unit estimates the nature of the provider based on the history of questions posted by the provider.
The selection unit serves as a provision target for providing the question posted by the provider whose property estimated by the estimation unit corresponds to the property of the user among the posted questions to the user. The selection device according to claim 5 , wherein selection is made. The estimation unit estimates the nature of the respondent preferred by the provider based on the history of the answers selected by the provider among the answers to the questions posted by the provider.
The selection unit serves as a provision target for providing the question posted by the provider whose property estimated by the estimation unit corresponds to the property of the user among the posted questions to the user. The selection device according to claim 5 or 6 , characterized in that selection is made. An estimation unit that estimates the nature of the user based on the information entered by the user,
A selection unit that selects a provision target provided by a provider having properties corresponding to the characteristics of the user estimated by the estimation unit as a provision target provided to the user.
Have,
The estimation unit estimates the nature of the user based on the summary selected by the user among the summaries of the predetermined contents.
The selection unit is characterized in that a summary of content provided by a provider having properties corresponding to the user properties estimated by the estimation unit is selected as a provision target provided to the user. Selection device. An estimation unit that estimates the nature of the user based on the information entered by the user,
A selection unit that selects a provision target provided by a provider having properties corresponding to the characteristics of the user estimated by the estimation unit as a provision target provided to the user.
Have,
The selection unit selects, among the contents related to the advertisement, the content created by the provider having the property corresponding to the property of the user estimated by the estimation unit as the content to be provided to the user. A selection device characterized by having a content. The estimation unit estimates the nature of the user based on the information input by the user when receiving the provision of the predetermined service.
The selection unit is characterized in that, among the providers who provide the predetermined service, the provision target provided by the provider having the property corresponding to the property of the user estimated by the estimation unit is selected. The selection device according to any one of claims 1 to 4 . The selection device according to claim 10 , wherein the estimation unit estimates the nature of the user based on the utterance of the user before receiving the provision of the predetermined service. The selection device according to claim 11 , wherein the estimation unit estimates the characteristics of the user based on the characteristics of the voice spoken by the user. It is a selection method performed by the selection device.
An estimation process that estimates the nature of the user based on the information entered by the user, and
It has a selection step of selecting a provision target provided by a provider having a property corresponding to the user property estimated by the estimation step as a provision target provided to the user.
The estimation step uses a learning model that learns the relationship between the characteristics of the input information and the nature of the user who input the information, and uses the information input by the user. Estimate the nature of a person
A selection method characterized by that. An estimation procedure that estimates the nature of the user based on the information entered by the user, and
It has a selection procedure for selecting a provision target provided by a provider having properties corresponding to the characteristics of the user estimated by the estimation procedure as a provision target provided to the user.
The estimation procedure uses a learning model that learns the relationship between the characteristics of the input information and the nature of the user who input the information, and uses the information input by the user. Estimate the nature of a person
A selection program that lets your computer do that. It is a selection method performed by the selection device.
An estimation process that estimates the nature of the user based on the information entered by the user, and
A selection step of selecting a provision target provided by a provider having properties corresponding to the user properties estimated by the estimation step as a provision target provided to the user.
Have,
The estimation process estimates the nature of the user based on the history of the answers posted by the user as answers to the questions posted by other users.
The selection process selects, among the posted questions, the question posted by the provider corresponding to the nature of the user estimated by the estimation process as the provision target to be provided to the user.
A selection method characterized by that . An estimation procedure that estimates the nature of the user based on the information entered by the user, and
A selection procedure for selecting a provision target provided by a provider having properties corresponding to the characteristics of the user estimated by the estimation procedure as a provision target provided to the user.
Have,
The estimation procedure estimates the nature of the user based on the history of the answers posted by the user as answers to the questions posted by other users.
The selection procedure selects, among the posted questions, the question posted by the provider corresponding to the nature of the user estimated by the estimation procedure as the provision target to be provided to the user.
A selection program that lets your computer do that . It is a selection method performed by the selection device.
An estimation process that estimates the nature of the user based on the information entered by the user, and
A selection step of selecting a provision target provided by a provider having properties corresponding to the user properties estimated by the estimation step as a provision target provided to the user.
Have,
The estimation step estimates the nature of the user based on the summary selected by the user among the summaries of the predetermined contents.
The selection step selects a summary of content provided by a provider having properties corresponding to the user's properties estimated by the estimation step as a provision target provided to the user.
A selection method characterized by that. An estimation procedure that estimates the nature of the user based on the information entered by the user, and
A selection procedure for selecting a provision target provided by a provider having properties corresponding to the characteristics of the user estimated by the estimation procedure as a provision target provided to the user.
Have,
The estimation procedure estimates the nature of the user based on the summary selected by the user among the summaries of the predetermined contents.
The selection procedure selects a summary of content provided by a provider having properties corresponding to the user properties estimated by the estimation procedure as a provision target provided to the user.
A selection program that lets your computer do that . It is a selection method performed by the selection device.
An estimation process that estimates the nature of the user based on the information entered by the user, and
A selection step of selecting a provision target provided by a provider having properties corresponding to the user properties estimated by the estimation step as a provision target provided to the user.
Have,
In the selection process, among the contents related to the advertisement, the content created by the provider having the property corresponding to the property of the user estimated by the estimation process is selected as the content to be provided to the user. do
A selection method characterized by that. An estimation procedure that estimates the nature of the user based on the information entered by the user, and
A selection procedure for selecting a provision target provided by a provider having properties corresponding to the characteristics of the user estimated by the estimation procedure as a provision target provided to the user.
Have,
In the selection procedure, among the contents related to the advertisement, the content created by the provider having the property corresponding to the property of the user estimated by the estimation procedure is selected as the content to be provided to the user. do
A selection program that lets your computer do that .

Images