Suggestion system, suggestion method, and program

Abstract

To display word-of-mouth information that will be helpful for each user in considering visiting spots. [Solution] One aspect of the present invention is a suggestion system comprising: a memory unit that stores spot information indicating a spot, word-of-mouth information about the spot, and scene and situation information indicating the scene or situation of the poster of the word-of-mouth information; an acquisition unit that acquires scene and situation information indicating the user's scene or situation in response to receiving a search request from a user specifying a spot; an extraction unit that compares the user's scene and situation information acquired by the acquisition unit with the scene and situation information of the poster who posted the word-of-mouth information about the spot specified by the search request stored in the memory unit, and extracts word-of-mouth information to be suggested to the user from the word-of-mouth information stored in the memory unit based on the comparison result; and a suggestion unit that suggests the word-of-mouth information extracted by the extraction unit.

Description

[Technical field]

[0001] The present invention relates to a suggestion system, a suggestion method, and a program. [Background technology]

[0002] Conventionally, there is known a technology for providing content including word-of-mouth in response to a user's search request. This type of technology is described, for example, in Patent Document 1. The information processing device described in Patent Document 1 includes a control unit that estimates a potential request according to a current user situation, searches for word-of-mouth information corresponding to the request, and controls to present the searched word-of-mouth information to the user, and the control unit searches word-of-mouth information posted by other users in the user's vicinity. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Republished WO2018 / 225429 Summary of the Invention [Problem to be solved by the invention]

[0004] In technology for displaying word-of-mouth information, even if the word-of-mouth information is displayed in order of newness or high rating, the word-of-mouth information that is useful to everyone will be displayed at the top, but there is a problem in that the word-of-mouth information that is more useful to each individual user will not be displayed at the top.

[0005] The present invention has been made in consideration of the above-mentioned problems, and aims to provide a suggestion system, a suggestion method, and a program that can display word-of-mouth information that will be useful for each user when considering spots to visit. [Means for solving the problem]

[0006] (1) One aspect of the present invention is spot information indicating a spot;and A storage unit that stores word-of-mouth information about the spot, and a search request that specifies the spot from a user is received. reception Department and Extracting reviews based on the positivity of the reviews and a fulfillment estimation unit that estimates information fulfillment indicating that a plurality of types of elements that a user considers important are included in the word-of-mouth information of the poster. The suggestion unit extracts word-of-mouth information narrowed down based on the positivity of the word-of-mouth, and The above information completeness The suggestion system includes review information ranked based on the above in list content including a list of spots, and includes review information ranked based on a comprehensive suggestion score obtained by adding a score based on statistical information to the information richness in detail content including details of the spots.

[0011] ( 2 In one aspect of the present invention, an information processing device is and a step of storing word-of-mouth information regarding the spot; and a step of receiving a search request specifying the spot from a user by the information processing device; Extracting reviews based on the positivity of the reviews and estimating information enrichment indicating that a plurality of types of elements that the user considers important are included in the word-of-mouth information of the poster, and the step of suggesting the word-of-mouth information includes extracting word-of-mouth information narrowed down based on the positivity of the word-of-mouth, and The above information completeness The suggestion method includes review information ranked based on the above in list content including a list of spots, and includes review information ranked based on a comprehensive suggestion score obtained by adding a score based on statistical information to the information richness in detail in detailed content including details of the spots.

[0012] ( 3 In one aspect of the present invention, a computer of an information processing device stores spot information indicating spots, anda step of storing word-of-mouth information about the spot; and a step of receiving a search request specifying the spot from a user. Extracting reviews based on the positivity of the reviews and estimating information enrichment indicating that a plurality of types of elements that the user considers important are included in the word-of-mouth information of the poster, and the step of suggesting the word-of-mouth information includes extracting word-of-mouth information narrowed down based on the positivity of the word-of-mouth, and The above information completeness The program includes review information ranked based on the above in list content including a list of spots, and includes review information ranked based on a comprehensive suggested score obtained by adding a score based on statistical information to the information richness in detail content including details of the spots. Effect of the Invention

[0013] According to one aspect of the present invention, it is possible to display word-of-mouth information that will be helpful for each user in considering spots to visit. [Brief description of the drawings]

[0014] [Figure 1] 1 is a block diagram showing an example of a configuration of a suggestion system 1 according to an embodiment. [Diagram 2] FIG. 2 is a diagram illustrating an example of a concept of suggesting content and word-of-mouth based on a scene / situation in an embodiment. [Diagram 3] FIG. 13 is a diagram showing the relationship between posters, word-of-mouth reviews, scenes / situations, and viewers in the embodiment. [Figure 4] FIG. 2 is a diagram showing an example of content that is browsed by a user in the embodiment. [Diagram 5] 1A and 1B are diagrams illustrating an example of a spot list screen and a spot details screen according to an embodiment. [Figure 6]FIG. 2 is a diagram showing an example of processing in a referrer terminal device 110, a suggestion device 200, and a database unit 270 in the suggestion system 1 according to the embodiment. [Figure 7] FIG. 11 is a diagram for explaining the contents of process 1-1 in an embodiment, in which (a) is an example of word-of-mouth data, (b) is a diagram showing an example of data on a viewer, and (c) is a diagram showing an example of the similarity of a scene / situation for each poster to the viewer. [Figure 8] FIG. 11 is a diagram explaining the process of calculating the similarity of scenes and situations, where (a) is a diagram showing an example of a judgment word, (b) is a diagram showing the process of determining the similarity from the review text, and (c) is a diagram showing the process of determining the similarity from the attributes of the poster's scene or situation and the attributes of the viewer's scene or situation. [Figure 9] FIG. 11 is a diagram showing another example of the process of calculating a similarity score of a scene / situation in the embodiment. [Figure 10] 11 is a diagram for explaining a calculation process of an information richness level in an embodiment. FIG. [Figure 11] 2 is a diagram showing the relationship between content categories, information elements, and words representing the information elements in an embodiment. FIG. [Figure 12] FIG. 11 is a diagram for explaining an example of the calculation process for information richness for word-of-mouth information in an embodiment, where (a) is a diagram showing the word-of-mouth content and information richness of content in the dining category, and (b) is a diagram showing the word-of-mouth content and information richness of content in the accommodation category. [Figure 13] 1A and 1B are diagrams for explaining the process of calculating a suggestion score in an embodiment, and are diagrams showing the correspondence between statistical information and suggestion screens ((a) is a list screen, and (b) is a details screen). [Figure 14] FIG. 11 is a diagram illustrating an example of a formula for calculating a suggestion score in the embodiment. [Figure 15] 11 is a diagram showing an example of items for calculating a suggestion score, examples of values ​​for calculating a score, a score, and an example of weights in an embodiment. FIG. [Figure 16]13 is a diagram for explaining a process of creating a ranking of content on a spot list screen in an embodiment. FIG. [Figure 17] 13 is a diagram for explaining a process of creating a ranking of content on a spot details screen in an embodiment. FIG. [Figure 18] 11A and 11B are diagrams illustrating an example of a spot list screen and a spot details screen according to an embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0015] A suggestion system, a suggestion method, and a program to which the present invention is applied will be described below with reference to the drawings. In response to receiving a search request from a user (referencer) specifying a spot, the suggestion system, the suggestion method, and the program to which the present invention is applied compare the scene and situation information of the user with the scene and situation information of the poster who posted the word-of-mouth information about the spot specified by the search request, extract word-of-mouth information to be suggested to the user from the word-of-mouth information based on the comparison result, and suggest the extracted word-of-mouth information. Suggestion is a process of providing a user with data including a list of contents and word-of-mouth or details of contents and word-of-mouth. In this way, the suggestion system realizes an information providing service that is highly convenient for users. In the following embodiments, the application field of the suggestion system is tourism and walking around town, but is not limited thereto, and can be applied to all fields such as shopping. In the following embodiments, a person who refers to word-of-mouth information or content will be described as a reference person or a user, and a person who posts word-of-mouth information will be described as a poster.

[0016] <Suggestion system configuration> First, the overall configuration and processing contents of a suggestion system 1 according to the embodiment will be described. FIG. 1 is a block diagram showing an example of a configuration of a suggestion system 1 according to an embodiment. The suggestion system 1 includes, for example, a contributor terminal device 100, a referrer terminal device 110, a suggestion device 200, a content providing device 300, and an environmental information providing device 400. The contributor terminal device 100, the referrer terminal device 110, the suggestion device 200, the content providing device 300, and the environmental information providing device 400 are connected to, for example, a communication network NW. Each device connected to the communication network NW includes a communication interface such as a network interface card (NIC) or a wireless communication module (not shown in FIG. 1). The communication network includes, for example, the Internet, a wide area network (WAN), a local area network (LAN), a cellular network, and the like.

[0017] The contributor terminal device 100 and the referrer terminal device 110 are, for example, portable terminal devices such as smartphones and tablet terminals. The contributor terminal device 100 and the referrer terminal device 110 run a UA (User Agent) such as a browser or an application program. The UA is, for example, an application for receiving an information providing service provided by the suggestion device 200. The contributor terminal device 100 and the referrer terminal device 110 generate operation information based on a user's operation, location information of the referrer terminal device 110 using a GPS (Global Positioning System, Global Positioning Satellite), and various requests, and transmit them to the suggestion device 200 or the content providing device 300.

[0018] The contributor terminal device 100 transmits various information such as a posting request specifying a spot, word-of-mouth information, and image information, and stores the transmitted various information in the suggestion device 200. The reference terminal device 110 transmits a search request specifying a spot, and performs display processing and operation reception processing using the content provided by the suggestion device 200 and the content received from the content providing device 300.

[0019] The suggestion device 200 is, for example, an information processing device that performs processing for an information providing service. The suggestion device 200 includes, for example, a reception unit 210, a scene / situation acquisition unit 220, a word-of-mouth information extraction unit 230, a scene / situation estimation unit 240, a fulfillment level estimation unit 250, a suggestion unit 260, and a database unit 270. The functional units such as the reception unit 210, the scene / situation acquisition unit 220, the word-of-mouth information extraction unit 230, the scene / situation estimation unit 240, the fulfillment level estimation unit 250, and the suggestion unit 260 are realized by a processor such as a CPU (Central Processing Unit) executing a program stored in a program memory. In addition, some or all of these functional units may be realized by hardware such as an LSI (Large Scale Integration), an ASIC (Application Specific Integrated Circuit), or an FPGA (Field-Programmable Gate Array), or may be realized by cooperation between software and hardware.

[0020] The receiving unit 210 receives requests such as a posting request transmitted by the contributor terminal device 100 and a search request transmitted by the referrer terminal device 110 .

[0021] The scene / situation acquisition unit 220 acquires scene / situation information indicating the user's scene or situation in response to receiving a search request from the user specifying a spot. The scene or situation is information such as accompanying persons, time period, weather, and location. The scene / situation acquisition unit 220 may acquire environmental information such as the weather of the spot from the environmental information providing device 400 based on the search request. The scene / situation of the user (referee) acquired by the scene / situation acquisition unit 220 is stored in the database unit 270 as word-of-mouth referee attribute data 274.

[0022] The word-of-mouth information extraction unit 230 compares the scene and situation information of the viewer acquired by the scene and situation acquisition unit 220 with the scene and situation information of the poster who posted the word-of-mouth information about the spot specified by the search request stored in the database unit 270, and extracts word-of-mouth information to be suggested to the user from the word-of-mouth information stored in the database unit 270 based on the comparison result.

[0023] The scene / situation estimation unit 240 estimates the poster's scene or situation based on the content of the poster's word-of-mouth information. The poster's scene or situation estimated by the scene / situation estimation unit 240 is stored in the database unit 270 as word-of-mouth poster attribute data 272.

[0024] The fulfillment estimation unit 250 estimates an information fulfillment indicating that a plurality of elements that a user considers important are included in the word-of-mouth information of the poster. The information fulfillment estimated by the fulfillment estimation unit 250 is stored in the database unit 270 as the review poster attribute data 272.

[0025] The suggestion unit 260 suggests the word-of-mouth information extracted by the word-of-mouth information extraction unit 230. Specifically, the suggestion unit 260 transmits to the reference terminal device 110 display data for displaying a spot list screen or a spot detail screen including the word-of-mouth of the contributor who visited the spot together with the content related to the spot requested by the user. Based on the comparison result of the word-of-mouth information extraction unit 230, the suggestion unit 260 may transmit scene / situation information of the contributor as the reason for suggesting the word-of-mouth information to the reference terminal device 110. The suggestion unit 260 may suggest the word-of-mouth information based on the fulfillment level estimated by the fulfillment level estimation unit 250.

[0026] The word-of-mouth information extraction unit 230 may extract word-of-mouth information based on the scene and situation information of the viewer acquired by the scene and situation acquisition unit 220, and the suggestion unit 260 may include word-of-mouth information narrowed down based on the freshness, recommendation level, text length, and positivity of the word-of-mouth information in a list content including a list of spots, which is word-of-mouth information ranked based on the scene and situation, information richness, and a suggestion score calculated from statistical information, and may include all word-of-mouth information not narrowed down based on the suggestion score in detailed content including details of the spots.

[0027] The database unit 270 is realized by, for example, a hard disk drive (HDD), a flash memory, an electrically erasable programmable read only memory (EEPROM), a read only memory (ROM), or a random access memory (RAM), or a hybrid storage device using a plurality of these. The database unit 270 may be realized by an external storage device accessible via various networks. An example of an external storage device is a network attached storage (NAS) device. The database unit 270 manages information required for processing of each unit in the suggestion device 200. The database unit 270 is a storage unit that stores spot information indicating spots, word-of-mouth information regarding spots, and scene / situation information indicating the scene or situation of a person who posted the word-of-mouth information.

[0028] The content providing device 300 is, for example, a server device that provides various contents in response to a search request from the reference user terminal device 110. The content providing device 300 provides, for example, contents related to various spots at travel destinations, such as restaurants, shops, public facilities (such as parks), transportation facilities, and commercial facilities. The content providing device 300 stores, for example, various content data in the content storage unit 300A. Content data is added with content attribute information indicating attributes such as the content category (dining category, lodging category, shopping category, sightseeing category). The content attribute is information indicating the nature of the content, for example, information corresponding to the scene or situation of the content.

[0029] The environmental information providing device 400 is a server device that distributes ever-changing environmental information such as weather information, traffic information, and disaster information. The environmental information providing device 400 transmits environmental information in response to a request from, for example, the suggestion device 200. The environmental information may include current location information and spot information indicating a spot at the current location (for example, a store name or a tourist spot name).

[0030] <Overview of Suggestion System 1> FIG. 2 is a diagram illustrating an example of a concept of suggesting content and word-of-mouth based on a scene and situation in the embodiment. First, the user terminal device 110 transmits a request specifying a travel record, spots, etc., to the suggestion device 200 before the user travels (before the trip). The suggestion device 200 suggests content including word-of-mouth information and photos shared as travel records, and content including photos and word-of-mouth shared in relation to the spots. In this way, the suggestion system 1 allows the user to view the travel record, shared photos, and word-of-mouth posted by the poster, encouraging the user to think of having a similar experience to the poster.

[0031] The user terminal device 110 transmits a request to view a travel record specifying a spot during the user's trip (during the trip) to the suggestion device 200. The suggestion device 200 suggests content such as photos and reviews related to the spot. In this way, the suggestion system 1 allows the user to view the photos and reviews posted by the poster, thereby enabling the user to discover that he or she would like to have the same experience as the poster.

[0032] After a trip, the contributor terminal device 100 posts a record of the trip, and photos and reviews of the spots. This allows the suggestion device 200 to accumulate reviews of the spots.

[0033] Fig. 3 is a diagram showing the relationship between a poster, a word-of-mouth, a scene / situation, and a viewer in the embodiment. Fig. 4 is a diagram showing an example of content that is browsed by a viewer in the embodiment. When the poster terminal device 100 posts word-of-mouth information, the suggestion device 200 accumulates the word-of-mouth information (word-of-mouth text data 276) in the database unit 270 in association with the spot and the scene / situation (word-of-mouth poster attribute data 272). In response to a request from the reference terminal device 110, the suggestion device 200 transmits word-of-mouth information of the spot corresponding to the request and the scene / situation corresponding to the reference person's scene / situation to the reference terminal device 110. For example, when the reference person's scene / situation is "sunset" and "with children", the suggestion device 200 displays, on the reference terminal device 110, a content list including, for example, content related to the spot to which word-of-mouth information related to the sunset has been posted, and the word-of-mouth information, and content related to the spot to which word-of-mouth information related to children has been posted, and the word-of-mouth information. In this way, the suggestion system 1 can add word-of-mouth including the poster's real experience and the scene / situation to the content list in order to encourage a way of enjoying the content according to the scene / situation.

[0034] FIG. 5 is a diagram showing an example of a spot list screen and a spot details screen according to the embodiment. The spot list screen includes a content image that lists the spot, a review image of that content, and other content images that list the spot. The review information on the spot list screen is positive, conveying the appeal of the spot. The review image includes a poster icon, a review title image, a review text image, a recommendation level icon, and a scene / situation icon. The scene / situation icon indicates the poster's scene / situation information as the reason for suggesting the review information. In the example of Figure 5, it shows that the scenes / situations "Family" and "Rain" corresponding to the content displayed at the top of the spot list screen match the scene / situation related to the viewer.

[0035] The spot details screen is a screen showing details of content selected by the viewer from among the contents included in the spot list screen. The spot details screen displays reviews with similar scenes and situations at the top of the list for spots that the viewer found attractive among the spots displayed on the spot list screen. The suggestion system 1 provides information that contributes to the user's decision-making by displaying the spot details screen on the viewer terminal device 110.

[0036] The word-of-mouth information extraction unit 230 extracts word-of-mouth information based on the scene and situation information of the referrer acquired by the scene and situation acquisition unit 220. The suggestion unit 260 includes word-of-mouth information narrowed down based on the freshness, recommendation level, text length, and positivity of the word-of-mouth information, ranked based on the scene and situation, information richness, and suggestion score calculated from statistical information, in content including a list of spots (spot list screen), and includes word-of-mouth information ranked based on the suggestion score of all word-of-mouth information not narrowed down in content including details of spots (spot detail screen).

[0037] [Overall processing of suggestion system 1] 6 is a diagram showing an example of processing in the referrer terminal device 110, the suggestion device 200, and the database unit 270 in the suggestion system 1 in the embodiment. The database unit 270 is assumed to store in advance review poster attribute data 272 including scene / situation information indicating the scene or situation of the reviewer of the review information, review referrer attribute data 274 including scene / situation information indicating the scene or situation of the reviewer of the review information, review body data 276 as review information related to spots, and review statistical data 278. The content providing device 300 also stores spot information indicating spots to be suggested to the referrer.

[0038] First, the user terminal device 110 transmits a spot list request or a spot details request to the suggestion device 200 (step S200). The suggestion device 200 performs a calculation process of the similarity of the scene / situation in process 1-1 and process 1-2 (steps S100, S102), a calculation process of the information richness in process 2 (step S104), a calculation process of the suggestion score in process 3 (step S106), and creates a suggestion screen in process 4 (step S108). The suggestion device 200 transmits display data for displaying the suggestion screen to the user terminal device 110, and the user terminal device 110 displays a spot list screen or a spot details screen as a suggestion screen based on the display data (step S202). Each process will be described in detail below.

[0039] [Process 1-1] The suggestion device 200 performs process 1-1 in response to receiving a spot list request or a spot detail request by the receiving unit 210 (step S100). Process 1-1 is a scene / situation similarity calculation process. The scene / situation similarity calculation process is a process for calculating the similarity between the scene or situation of the reviewer and the scene or situation of the reviewer based on the review poster attribute data 272, the review referrer attribute data 274, and the review text data 276. Process 1-1 calculates the similarity of the accompanying person and the weather among the scenes / situations, but is not limited thereto.

[0040] FIG. 7 is a diagram for explaining the contents of process 1-1 in an embodiment, in which (a) is an example of word-of-mouth data, (b) is a diagram showing an example of data on a viewer, and (c) is a diagram showing an example of the similarity of a scene / situation for each poster with respect to the viewer. The review data in Fig. 7(a) includes, for example, a review sequence number, a spot ID, a review poster, a review title, review text data 276, and accompanying persons at the time of visit and weather at the time of visit included in the review poster attribute data 272. The review referrer data in Fig. 7(b) is data corresponding to the review referrer attribute data 274, and includes, for example, the user name of the review referrer, accompanying persons of the referrer, and weather of the referrer.

[0041] The suggestion device 200 calculates a match score between the scene or situation (companion, weather) of the referrer and the scene or situation (companion, weather) of the poster, for each referrer, based on the review referrer attribute data 274, the review poster attribute data 272, and the review text data 276. As a result, the suggestion device 200 calculates a companion similarity score and a weather match score for, for example, user X, as shown in FIG. 7(c). For example, in the review 001, if the companion at the time of the review poster's visit is "family" and the weather at the time of the visit is "sunny," and the companion at the time of the visit of user X is "family" and the weather at the time of the visit is "rainy," the companion match score is "1" and the weather match score is "0." For example, in the review 002, if "son" is described in the review text data 276 and the companion at the time of the visit of user X is "family," the companion match score is "1.5." For example, in the review 003, if "rain" is written in the review body data 276 and the weather at the time of user X's visit is "rain," the weather match score will be "1.5."

[0042] FIG. 8 is a diagram explaining the process of calculating the similarity of scenes and situations, where (a) is a diagram showing an example of a judgment word, (b) is a diagram showing the process of determining the similarity from the review text, and (c) is a diagram showing the process of determining the similarity from the attributes of the poster's scene or situation and the attributes of the viewer's scene or situation. The suggestion device 200 performs the following process to calculate a matching score (similarity) of the scene / situation based on the word-of-mouth text data 276. First, as a preliminary step, the determination words are defined. In this embodiment, as shown in FIG. 8(a), a group of words for determining a specific attribute for each of the companion and the weather is defined. First, actual word-of-mouth text data 276 is collected, topic modeling (a natural language processing method) using LDA is calculated, and words are extracted from the word-of-mouth text data 276.

[0043] The suggestion device 200 performs the following determination process (1) and determination process (2) for the accompanying person and the weather, respectively. The judgment process (1) searches whether the words defined in the preliminary work corresponding to the scene / situation attributes (e.g., family, rain) of the review referrer are included in the review body data 276. As shown in Fig. 8(b), if the words defined in the preliminary work (e.g., son) are included in the review body data 276, a companion matching score of "1.5" is assigned, and the judgment process ends. If the words defined in the preliminary work are not included in the review text data 276, in the judgment process (2), the scene / situation attributes of the review viewer (e.g., family, rain) are compared with the scene or situation attributes of the review poster, and if there is a match between "family members" or between "partner" and "family" as shown in Figure 8(c), a companion match score of "1.0" is assigned, and if there is no match, a companion match score of "0" is assigned.

[0044] When the review text contains words related to the scene or situation attributes of the reviewer, there is a high possibility that the review information contains useful information for the reviewer, but it is difficult to extract the scene or situation from the review text, and there is a high possibility that the scene or situation will be missed or incorrectly extracted. On the other hand, the scene or situation attributes entered by the reviewer themselves can be easily obtained, and the reliability of the scene or situation is high, but the reviewer's review text does not necessarily match the scene or situation. Therefore, the suggestion device 200 defines the judgment words by limiting them to words with high reproducibility so as to reduce false detection, and in the judgment process (1), a high matching score can be given to a review containing a word corresponding to the judgment word, since it is highly likely to be useful. Also, even if the judgment word is not described in the review text, the suggestion device 200 extracts the review based on the attributes of the scene and situation in the judgment process (2). In the case of the judgment process (2), the suggestion device 200 gives a lower matching score than in the judgment process (1) because it is highly likely that the review text does not describe an experience according to the scene and situation.

[0045] [Process 1-2] The suggestion device 200 performs process 1-2 after or in parallel with process 1-1 (step S102). Process 1-2 is a scene / situation similarity calculation process. The scene / situation similarity calculation process of process 1-2 is a process of calculating a match score (similarity) between the scene or situation of the reviewer and the scene or situation of the reviewer based on the review poster attribute data 272 and the review referrer attribute data 274. Process 1-2 calculates a match score for the date and time period of the scene / situation, but is not limited to this.

[0046] FIG. 9 is a diagram showing another example of the scene / situation match score calculation process in the embodiment. When the scene / situation items are a companion and weather, the suggestion device 200 calculates the matching score by the above-mentioned process 1-1, but when the scene / situation items are a date and time, the suggestion device 200 calculates the matching score in process 1-2.

[0047] When calculating the match score of a date as a scene / situation, the suggestion device 200 uses the poster's visit date as the date. When the difference between the poster's visit date and the date specified by the viewer is N days (e.g., N=20) or less, the suggestion device 200 assigns "1.0" to the match score of the review by the poster. When the difference between the poster's visit date and the date specified by the viewer is greater than N days and less than 2N days, the suggestion device 200 assigns "0.5" to the match score of the review by the poster. Note that the suggestion device 200 does not need to consider the year of the poster's visit date in order to suggest reviews of the same season (date). For example, if the date specified by the viewer is July 23, 2024, the suggestion device 200 assigns "1.0" to a review whose poster's visit date is July 23, 2023.

[0048] When calculating the matching score of the time period as a scene / situation, the suggestion device 200 compares the visiting time period of the contributor with the time period specified by the reference user to calculate the matching score for the time period. The suggestion device 200 assigns "1.0" when the time periods match, and assigns "0" when the time periods do not match. The suggestion device 200 may assign "1.0" when the time period matches "daytime" or "nighttime". The suggestion device 200 may change the time period according to the category of the content. For example, in the case of sightseeing or shopping, the suggestion device 200 may determine that before 5 p.m. is daytime and from 5 p.m. is nighttime, and in the case of eating, before 11 a.m. is morning, from 11 a.m. to before 5 p.m. is daytime, and from 5 p.m. is nighttime.

[0049] [Process 2] The suggestion device 200 performs a process 2 for calculating the richness of information included in the word-of-mouth (step S104). The process 2 estimates the information richness based on the word-of-mouth text data 276. The information richness increases as the number of elements that the reader considers important increases. For example, the information richness increases as multiple elements that the reader considers important are set for each content category, and the number of elements mentioned in the word-of-mouth increases.

[0050] FIG. 10 is a diagram for explaining the calculation process of the information richness level in the embodiment. The content categories are, for example, meals, accommodation, shopping, and sightseeing. For example, five elements are set in the meals category: (1) meal content, (2) customer service, (3) price, (4) atmosphere / view, and (5) crowding / reservations. The suggestion device 200 counts the number of elements mentioned in reviews among elements (1) to (5), and multiplies the count value by 0.2 to calculate the information richness. When the number of elements mentioned in reviews is 0, the information richness is 0.0; when the number of elements mentioned in reviews is 1, the information richness is 0.2; and when the number of elements mentioned in reviews is 5, the information richness is 1.0.

[0051] The selection of elements was carried out by, for example, referring to a research report on the travel industry, extracting the elements that reviewers consider important when considering which spots to visit, and selecting and aggregating the information elements based on the extraction results, thereby selecting five elements. There may be categories in which multiple elements can be selected through element selection, and categories in which multiple elements cannot be selected. For example, five elements could be selected in the dining and accommodation categories, so information richness can be calculated for reviews in the dining and accommodation categories. On the other hand, important elements could not be clearly determined in the shopping and sightseeing categories, so information richness does not need to be calculated.

[0052] 11 is a diagram showing the relationship between content categories, information elements, and words expressing the information elements in the embodiment. The words expressing the information elements are the result of calculating topic modeling (a natural language processing method) using LDA for review body data 276, extracting representative words expressing each information element, and adding, deleting, and modifying the extracted words to provide content and reviews related to spots.

[0053] The fulfillment estimation unit 250 judges whether or not the word corresponding to the content category and the information element is included in the word-of-mouth body data 276. The fulfillment estimation unit 250 extracts word-of-mouth information that includes one or more words corresponding to the content category and the information element. The fulfillment estimation unit 250 may divide the word-of-mouth body into words, perform morphological analysis, and convert adjectives, verbs, and the like into their original forms. For example, the fulfillment estimation unit 250 converts "It was delicious" to "It's delicious."

[0054] FIG. 12 is a diagram for explaining an example of the calculation process of information richness for word-of-mouth information in an embodiment, where (a) is a diagram showing the word-of-mouth content and information richness of content in the dining category, and (b) is a diagram showing the word-of-mouth content and information richness of content in the accommodation category. The review text shown in the upper part of Fig. 12(a) contains words that describe meals in the meal category, words that describe service in the meal category, words that describe price, and words that describe atmosphere and scenery, so the fulfillment estimation unit 250 calculates an information fulfillment of 0.8. The review text shown in the lower part of Fig. 12(a) contains three words that describe meals in the meal category, but does not contain any words that describe other elements, so the fulfillment estimation unit 250 calculates an information fulfillment of 0.2. The review text shown in the upper part of Fig. 12(b) contains words describing the food in the accommodation category, words describing the customer service in the accommodation category, words describing the room, and words describing cleanliness, so the fulfillment estimation unit 250 calculates an information fulfillment of 0.8. The review text shown in the lower part of Fig. 12(a) contains three words describing the food in the accommodation category, but does not contain words describing other elements, so the fulfillment estimation unit 250 calculates an information fulfillment of 0.2.

[0055] [Process 3] The suggestion device 200 performs a process 3 of calculating a suggestion score based on the similar scenes / situations calculated in the processes 1-1 and 1-2, the information richness calculated in the process 2, and statistical information including the word-of-mouth statistical data 278 (step S106). The suggestion unit 260 calculates an overall suggestion score by adding the score based on the statistical information to the scene / situation matching score calculated in the process 1 and the information richness score calculated in the process 2.

[0056] FIG. 13 is a diagram for explaining the process of calculating the suggestion score in the embodiment, showing the correspondence between the statistical information and the suggestion screen ((a) is a list screen, and (b) is a details screen). The suggestion unit 260 calculates (1) a photo score, (2) a title score, (3) a freshness score, (4) a usefulness score, (5) a recommendation score, (6) a text length score, and (7) a positivity score.

[0057] (1) The photo score is higher because the more photos attached to a review, the more likely it is to be a useful review. For example, the suggestion unit 260 multiplies the number of photos attached by 0.25, and calculates the value to be 1.0 when the number of photos is 5 or more. (2) The title score is useful because reviews that include the title of the review allow users to check the information concisely. If a review has a title, the score is 1.0, and if there is no title, the score is 0. (3) The freshness score is a value according to the number of days that have passed since the visit date, because reviews written a short time after the visit date are useful. For example, the suggestion unit 260 sets the freshness score to 1.0 if the visit date is within six months, 0.5 if the visit date is between six months and one year, 0.25 if the visit date is between one year and two years, and 0 if the visit date is more than two years. (4) The more helpfulness scores there are from reviewers, the higher the helpfulness score, since reviews with a large number of helpfulness scores from reviewers are of high quality. For example, the suggestion unit 260 adds a value obtained by taking the logarithm of the number of helpfulness scores (logarithm base 10) as the helpfulness score. This is to prevent reviews with an extremely large number of helpfulness scores from being frequently suggested, since the number of helpfulness scores does not have a set range of values. For this reason, a logarithmic transformation may be performed as a mechanism to make it difficult to increase the score when the number of helpfulness scores becomes relatively large. (5) The recommendation score is calculated based on the recommendation level of the review (a system in which the review assigns stars on a 5-point scale) since the review with a high recommendation level from the poster describes the attraction of the spot. The suggestion unit 260 calculates the recommendation score by multiplying the recommendation level by 0.2, for example. (6) The text length score is increased according to the length of the review text, because the longer the text length of a review, the more information it contains. The suggestion unit 260 calculates the text length score by, for example, multiplying the number of characters by 0.0025. However, if the number of characters is 401 or more, the text length score becomes 1. This is to prevent reviews with extremely long text lengths from being frequently suggested when the numerical range of the text length is not determined. (7) The positivity score is calculated based on the positivity of the review text, because reviews with a high positivity score describe the attractions of the spot and are helpful to readers. The positivity score is calculated, for example, by using known natural language processing, so that the more positive words the review text contains, the higher the score will be.

[0058] FIG. 14 is a diagram illustrating an example of a formula for calculating the suggestion score according to the embodiment. The suggestion unit 260 multiplies each of the companion match score, time zone match score, season (date) match score, weather match score, information richness score, photo score, title score, freshness score, usefulness score, recommendation score, text length score, and positivity score by a weight based on the similarity of the scene / situation, and calculates the suggestion score of the content by adding up the multiplied values.

[0059] FIG. 15 is a diagram showing an example of items for calculating a suggestion score, examples of values ​​for calculating a score, scores, and weights according to the embodiment. For example, the suggestion unit 260 sets the weights of process 1-1, process 1-2, and information richness to "2.0" to give them a higher weight than other items. This allows the suggestion unit 260 to calculate a suggestion score that prioritizes the scene / situation and information richness. Also, by setting the upper limit values ​​of the companion match score and time period match score in process 1-1 to 1.5, which is higher than other similarity scores, it is possible to calculate a suggestion score that prioritizes the similarity score calculated in process 1-1.

[0060] [Process 4] The suggestion device 200 performs process 4 after process 3 (step S108). Process 4 is a process of creating a spot list screen or a spot details screen based on the suggestion score calculated in process 3. The suggestion device 200 transmits information for displaying the created spot list screen or spot details screen to the reference user terminal device 110. The reference user terminal device 110 displays the spot list screen or the spot details screen based on the received information (step S202).

[0061] FIG. 16 is a diagram for explaining a process of creating a ranking of contents on the spot list screen in the embodiment. The suggestion unit 260 creates a ranking of the content to determine the content and reviews to be displayed on the spot list screen based on the calculated suggestion score. The spot list screen picks up one positive review that conveys the appeal of the tourist spot, and displays the picked review together with the spot. For this reason, the suggestion unit 260 performs a two-stage process: first, it narrows down the reviews, and then it ranks the reviews.

[0062] The suggestion unit 260 narrows down the reviews to those that satisfy all four conditions for freshness, recommendation level, text length, and positivity. The reviews that are narrowed down are the ones shown in the upper and middle rows of the reviews shown in FIG. (1) Freshness: The reviewer's visit date must be within the past three years. (2) Conditions for recommendation level: The reviewer must give the product a recommendation level of 3 stars or higher. (3) Length: The length of the review text must be 100 characters or more. (4) Condition for positivity: The positivity level of the review text must be 0.8 or higher.

[0063] The suggestion unit 260 creates a ranking for the narrowed down reviews based on the suggestion scores calculated in process 3. The suggestion unit 260 creates display data for displaying a spot list screen that displays the review ranked first at the top. The review shown in the top row of the reviews shown in Fig. 16 is ranked first and is the review displayed first on the spot list screen.

[0064] 17 is a diagram for explaining the process of creating a ranking of contents on the spot details screen in the embodiment. The spot details screen displays raw opinions of users with similar scenes and situations for the spot that the user found attractive on the spot list screen at the top, providing information that contributes to the user's decision-making. The suggestion unit 260 displays the reviews displayed on the spot list screen at the top (leading position) of the spot details screen. The suggestion unit 260 creates a ranking for the second and subsequent reviews on the spot details screen based on the suggestion scores calculated in process 3. The suggestion unit 260 does not narrow down the reviews as it does for the reviews displayed on the spot list screen. As a result, old reviews, reviews with low recommendation ratings, reviews with short sentence length, and reviews with low positivity ratings can also be used as a reference for the viewer's decision-making, so rankings can be created without narrowing down the results, unlike the spot list.

[0065] FIG. 18 is a diagram showing an example of a spot list screen and a spot details screen according to the embodiment. The suggestion unit 260 first displays the first ranked review on the spot list screen by filtering reviews and ranking them based on the suggestion score (FIG. 18(a)). When a review displayed on the spot list screen is selected, the suggestion unit 260 ranks the review displayed on the spot list screen as first, creates a ranking of the other reviews, and displays a spot details screen (FIG. 18(b)).

[0066] (Effects of the embodiment) As described above, according to the suggestion system 1 of the embodiment, a suggestion device 200 can be realized, which includes a database unit 270 that stores spot information indicating a spot, word-of-mouth information regarding the spot, and scene and situation information indicating a scene or situation of a person who posted the word-of-mouth information, a scene and situation acquisition unit 220 that acquires scene and situation information indicating a scene or situation of a user in response to receiving a search request specifying a spot from a person who refers to the user, a word-of-mouth information extraction unit 230 that compares the scene and situation information of the user acquired by the scene and situation acquisition unit 220 with the scene and situation information of a person who posted word-of-mouth information regarding the spot specified by the search request stored in the database unit 270, and extracts word-of-mouth information to be suggested to the user from the word-of-mouth information stored in the database unit 270 based on the comparison result, and a suggestion unit 260 that suggests the word-of-mouth information extracted by the word-of-mouth information extraction unit 230. According to this suggestion system 1, word-of-mouth information that is useful for considering a visiting spot for each user can be displayed.

[0067] According to the suggestion system 1, the suggestion unit 260 can transmit scene and situation information of the poster as the reason for suggesting the word-of-mouth information based on the comparison result. Even if a word-of-mouth corresponding to a request according to the current user situation is searched for from word-of-mouth information posted by other users around the user, the reason for displaying the word-of-mouth may not be conveyed to the user, and the user may not be convinced of the displayed word-of-mouth. In contrast, according to the suggestion system 1, the scene and situation information of the poster is transmitted as the reason for suggesting, so that the user can be convinced of the word-of-mouth.

[0068] Furthermore, reviews that are highly rated and short, or reviews that repeatedly write the same content, tend to be displayed at the top of the results, which can deprive users of the opportunity to access meaningful information. Furthermore, because reviews that are highly ranked in both the spot list content and the spot detail content provided in response to a user's request are displayed at the top, it is not possible to display reviews that are appropriate for each of the spot list content and the spot detail content, and it is not possible to display reviews that will attract users in each content.

[0069] According to the suggestion system 1, the scene / situation estimation unit 240 can estimate the scene or situation of the poster based on the content of the poster's word-of-mouth information. Word-of-mouth reviews written by posters often do not properly describe the scene or situation related to the review due to mistakes or omissions in the review, and even if suggestions are made using the review, the accuracy of the suggestions may decrease. In contrast, according to the suggestion system 1, the scene or situation is estimated based on the actual content of the review, so the accuracy of suggesting reviews that match the scene / situation of the viewer can be improved.

[0070] According to the suggestion system 1, it is possible to estimate information richness indicating that multiple types of elements that a user considers important are included in the word-of-mouth information of a poster, and to suggest word-of-mouth information based on the information richness by the suggestion unit 260. According to the suggestion system 1, it is possible to prevent highly rated, short word-of-mouth reviews and reviews that repeatedly write the same content from being displayed at the top, and it is possible to increase the opportunities for users to access meaningful information.

[0071] According to the suggestion system 1, the word-of-mouth information extraction unit 230 extracts word-of-mouth information narrowed down based on the positivity of the word-of-mouth, the narrowed down word-of-mouth information is ranked based on the scene / situation and information richness, and the word-of-mouth information ranked based on the suggestion score is included in the list content (spot list screen) including a list of spots, and the detailed content (spot detail screen) including the details of the spot can include the word-of-mouth information ranked based on the suggestion score. According to the suggestion system 1, the spot list screen can display reviews with high positivity and high ranking, and the spot detail screen can display the reviews with high rankings at the top. As a result, the suggestion system 1 can attract users with the positive word-of-mouth displayed on the spot list screen, and provide information that contributes to the user's decision-making by the word-of-mouth including the poster's honest opinion displayed on the spot detail screen.

[0072] Although each embodiment and each variant has been described, these are merely examples and are not intended to be limiting. For example, any of the embodiments or variants, or a part of each embodiment or a part of each variant, may be combined with one or more other embodiments or one or more other variants to realize one aspect of the present invention.

[0073] In addition, programs for executing each process of the contributor terminal device 100, the referrer terminal device 110, the suggestion device 200, the content providing device 300, and the environmental information providing device 400 in this embodiment may be recorded on a computer-readable recording medium, and the programs recorded on the recording medium may be read into a computer system and executed to perform the various processes described above related to the contributor terminal device 100, the referrer terminal device 110, the suggestion device 200, the content providing device 300, and the environmental information providing device 400.

[0074] Note that the "computer system" referred to here may include hardware such as the OS and peripheral devices. Furthermore, if a WWW system is used, the "computer system" also includes the homepage provision environment (or display environment). Furthermore, "computer-readable recording medium" refers to storage devices such as flexible disks, magneto-optical disks, ROMs, writable non-volatile memories such as flash memory, portable media such as CD-ROMs, and hard disks built into computer systems.

[0075] Furthermore, the term "computer-readable recording medium" includes a storage medium that holds a program for a certain period of time, such as a volatile memory (e.g., DRAM (Dynamic Random Access Memory)) inside a computer system that serves as a server or client when the program is transmitted via a network such as the Internet or a communication line such as a telephone line. The program may also be transmitted from a computer system that stores the program in a storage device or the like to another computer system via a transmission medium or by a transmission wave in the transmission medium.

[0076] Here, the "transmission medium" that transmits the program refers to a medium that has the function of transmitting information, such as a network (communication network) such as the Internet or a communication line (communication line) such as a telephone line. The above program may also be one that realizes part of the above-mentioned functions. Furthermore, it may be one that can realize the above-mentioned functions in combination with a program already recorded in the computer system, a so-called difference file (difference program).

[0077] Although the embodiment of the present invention has been described in detail above with reference to the drawings, the specific configuration is not limited to this embodiment, and the present invention also includes designs within the scope of the gist of the present invention. [Explanation of symbols]

[0078] 1. Suggestion System 100 Contributor terminal device 110 User terminal device 200 Suggestion Device 210 Reception 220 Scene and situation acquisition unit 230 Word-of-mouth information extraction unit 240 Scene and Situation Estimation Unit 250 Estimation of Comprehensiveness 260 Suggestion Department 270 Database Department 272 Reviewer attribute data 274 Reviewer Attribute Data 276 Review text data 278 Reviews Statistics 300 Content providing device 300A Content storage unit 400 Environmental information provision equipment

Claims

1. A storage unit that stores spot information indicating a spot and word-of-mouth information regarding the spot; a reception unit that receives a search request specifying a spot from a user; A suggestion section that extracts word-of-mouth information narrowed down based on the positivity of the word-of-mouth; a fulfillment level estimation unit that estimates an information fulfillment level indicating that a plurality of types of elements that a user considers important are included in the word-of-mouth information of the poster, The suggestion unit is Extracts reviews based on the positivity of the reviews, The word-of-mouth information narrowed down based on the positivity of the word-of-mouth is ranked based on the information richness, and the word-of-mouth information is included in a list content including a list of spots, and the word-of-mouth information ranked based on a comprehensive suggestion score obtained by adding a score based on statistical information to the information richness is included in a detailed content including details of the spots. Suggestion system.

2. an information processing device storing spot information indicating a spot and word-of-mouth information regarding the spot; a step of receiving a search request specifying a spot from a user by the information processing device; A step of extracting word-of-mouth information narrowed down based on the positivity of the word-of-mouth; and estimating an information richness indicating that a plurality of types of elements that a user considers important are included in the word-of-mouth information of the poster, The step of suggesting word-of-mouth information includes: Extracts reviews based on the positivity of the reviews, The word-of-mouth information narrowed down based on the positivity of the word-of-mouth is ranked based on the information richness, and the word-of-mouth information is included in a list content including a list of spots, and the word-of-mouth information ranked based on a comprehensive suggestion score obtained by adding a score based on statistical information to the information richness is included in a detailed content including details of the spots. Suggested methods.

3. The computer of the information processing device storing spot information indicating a spot and word-of-mouth information regarding the spot; receiving a search request specifying a spot from a user; A step of extracting word-of-mouth information narrowed down based on the positivity of the word-of-mouth; and estimating an information richness indicating that a plurality of types of elements that a user considers important are included in the word-of-mouth information of the poster, The step of suggesting word-of-mouth information includes: Extracts reviews based on the positivity of the reviews, The word-of-mouth information narrowed down based on the positivity of the word-of-mouth is ranked based on the information richness, and the word-of-mouth information is included in a list content including a list of spots, and the word-of-mouth information ranked based on a comprehensive suggestion score obtained by adding a score based on statistical information to the information richness is included in a detailed content including details of the spots. program.

Images