JP2016015162A - Recommendation device, recommendation method, and recommendation program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To reflect the opinion of the whole users to a reservation facility.SOLUTION: A recommendation device 10 includes: a collection unit 15a that collects logs of messages from a room in which messages posted by users are placed; an extraction unit 15b that extracts keywords from the messages collected in the predetermined room out of the logs collected by the collection unit 15a; and a search unit 15c that searches for a facility with a heavier weight added to keywords extracted from a newly posted message out of the keywords extracted by the extraction unit 15b.

Description

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

  Reservation sites for reserving restaurants and facilities such as lodging are known. For example, when reserving a restaurant, the user searches for a restaurant corresponding to the search condition by specifying a search condition such as an area where the restaurant is to be searched, a date and time when the user wants to use the restaurant, or a genre of the restaurant. be able to.

JP 2012-14659 A

  However, in the above-described conventional technology, the search condition is determined according to the hobbies and preferences of some users who make reservations, and thus the opinions of the entire user cannot be reflected in the reservation facility.

  The present application has been made in view of the above, and an object thereof is to provide a recommendation device, a recommendation method, and a recommendation program that can reflect the opinions of the entire user in a reservation facility.

  A recommendation device according to an aspect includes a collection unit that collects a message log from a room in which a message posted by a user is posted, and a message collected in a predetermined room among the logs collected by the collection unit. An extraction unit for extracting a keyword; and a search unit for searching for a facility by assigning a greater weight to a keyword extracted from a newly posted message among the keywords extracted by the extraction unit.

  According to the aspect of the embodiment, there is an effect that the opinion of the entire user can be reflected in the reservation facility.

FIG. 1 is a diagram illustrating a configuration example of a network system according to the embodiment. FIG. 2 is a diagram illustrating a configuration example of the recommendation device according to the embodiment. FIG. 3 is a diagram illustrating an example of a message log. FIG. 4 is a diagram illustrating an example of a message log. FIG. 5 is a flowchart illustrating a procedure of recommendation processing according to the embodiment. FIG. 6 is a diagram illustrating an example of a recommendation display method. FIG. 7 is a diagram illustrating an example of a recommendation display method. FIG. 8 is a diagram illustrating an example of a recommendation display method. FIG. 9 is a diagram illustrating an example of a remind screen. FIG. 10 is a hardware configuration diagram illustrating an example of a computer that implements the function of the recommendation device.

  Hereinafter, modes for implementing a recommendation device, a recommendation method, and a recommendation program according to the present application (hereinafter referred to as “embodiments”) will be described in detail with reference to the drawings. In addition, the recommendation apparatus, the recommendation method, and the recommendation program which concern on this application are not limited by this embodiment. In the following embodiments, the same portions are denoted by the same reference numerals, and redundant description is omitted.

[1. System configuration]
First, the configuration of the network system according to the embodiment will be described with reference to FIG. FIG. 1 is a diagram illustrating a configuration example of a network system according to the embodiment. As shown in FIG. 1, the network system 1 accommodates a recommendation device 10, an SNS (Social Networking Service) server 20, a reservation server 30, and terminal devices 50A to 50N. Hereinafter, the terminal devices 50A, 50B,..., 50N will be collectively referred to as “terminal device 50”.

  The recommendation device 10, the SNS server 20, the reservation server 30, and the terminal device 50 are connected to be communicable with each other via a network N (not shown). As such a network N, any type of communication network such as the Internet (Internet), a LAN (Local Area Network), a VPN (Virtual Private Network) or the like can be adopted regardless of wired or wireless.

  Among these, the SNS server 20 is a server device operated by a service provider that provides a social networking service. As one aspect, the SNS server 20 provides various services to the terminal device 50 in which the client-side communication application that receives the provision of the social networking service is installed. Hereinafter, the communication application may be abbreviated as “communication app”. For example, the SNS server 20 starts a chat-type message exchange service, a My Page function for publishing one's profile and avatar, an address book function for registering friends and acquaintances who interact on the SNS, and a My Page It is possible to provide a visit history reference function that allows the user's history to be referred to, and a diary function that can easily update entries like a blog.

  The reservation server 30 is a server device operated by a service provider that provides a facility reservation service. As one aspect, the reservation server 30 collects the vacancy status of the facility from the computer installed in the facility according to the day and the time zone, and manages the vacancy status according to the day and the time zone for each facility. Under the management of the vacancy status, the reservation server 30 accepts search conditions for a reservation site established by the reservation server. For example, as an example of the search condition, the reservation server 30 receives an area where a facility is desired to be searched, a date and time of use, a genre of the facility, and the like. The reservation server 30 provides a search service that outputs a facility that satisfies the search condition as a search result, and a reservation service that reserves a specified facility from the search result.

  The terminal device 50 is an information processing terminal used by a user. The terminal device 50 includes all portable terminals in its category. For example, as the terminal device 50, a mobile communication terminal capable of connecting to a mobile communication network such as a smartphone, a mobile phone, or a PHS (Personal Handyphone System) can be adopted. The terminal device 50 may be a tablet terminal such as a PDA (Personal Digital Assistants) that does not have the ability to connect to a mobile communication network.

  The recommendation device 10 is a computer that provides the following recommendation service. One embodiment of the recommendation device 10 can be implemented by installing a recommendation program that provides a recommendation service as package software or online software on a desired computer. For example, the recommendation device 10 may be implemented as a Web server that provides a recommendation service, or may be implemented as a cloud that provides a recommendation service by outsourcing.

  Under such a configuration, the recommendation device 10 according to the embodiment realizes a recommendation service that reflects the opinions of the entire user in the reservation facility. In order to realize such a recommendation service, there is a contrivance in using a talk room in which messages are exchanged in a chat format for specifying search conditions. The “talk room” referred to here can be a communication function that allows a plurality of users to post messages in a conversational format and allows each user to check the history of the messages. Here, the “message” may include an image or moving image data acquired / captured by the user, in addition to a text comment.

  This point will be described with reference to the example of FIG. As shown in FIG. 1, the recommendation device 10 collects a log of messages from a talk room in which messages posted by a plurality of users are posted (1). After that, the recommendation device 10 extracts keywords related to the location, category, time, and number of people from the messages included in the log collected in (1) (2).

  For example, when the log collected from the talk room is the log L1 shown in FIG. 1, the recommendation device 10 extracts the underlined word attached to the message shown in FIG. 1 as a keyword. Specifically, “3: 3” is extracted from the message T1 as a keyword related to the number of people, and “Roppongi” is extracted from the message T3 as a keyword related to the place (place name). Further, “Chinese Chinese” is extracted from the message T4 as a keyword related to the category, and “Italian” is extracted from the message T5 as a keyword related to the category. Further, “7 o'clock” is extracted from the message T7 as a keyword related to time. In this example, keywords are not extracted from the remaining messages T2, T6, and T8.

  After that, the recommendation device 10 weights each keyword when there are a plurality of overlapping keywords as search conditions among the keywords related to the location, category, time, and number of people previously extracted when the reservation server 30 is requested to search for the facility. Execute. In the example of FIG. 1, the keywords “Chinese Chinese” and “Italian” related to the facility category are duplicated as search condition items. In this case, the message from which “Chinese” is extracted is T4, the message from which “Italian” is extracted is T5, and the message T5 is newer. For this reason, “Italian” is given a greater weight than “Chinese”. As a result, “Italian” is used as a keyword for narrowing down the category of the facility.

  Thus, after recommending the keyword, the recommendation device 10 requests the reservation server 30 to search for the facility using the location, category, time, and number of people as search conditions (3). In the example of FIG. 1, an Italian restaurant “Italian restaurant A” having 6 vacant seats in the time zone from 19:00 in the Roppongi neighborhood is searched. Then, the recommendation apparatus 10 outputs the recommendation of the facility responded as a search result from the reservation server 30 to the terminal apparatus 50 possessed by the user participating in the talk room (4). In response to the output, the terminal device 50 displays the facility recommendation output by the recommendation device 10 as a banner advertisement in the talk room (5). For example, referring to the example of FIG. 1, the recommendation R 1 of the facility “Italian restaurant A” is displayed at the bottom of the display area P 1 of the terminal device 50. In this example, since the messages T1 to T2 in the log L1 do not fit in the display area P1, the messages T1 to T2 are not displayed.

  Here, when an operation, for example, a tap operation is performed on the recommendation R1 displayed on the terminal device 50, the recommendation device 10 performs the tap operation on the reservation confirmation screen of the facility corresponding to the recommendation R1. (6). In response to the output, the terminal device 50 displays the reservation confirmation screen Y output by the recommendation device 10 (7). At this time, it is assumed that the OK button B1 on the reservation confirmation screen Y displayed on the terminal device 50 is tapped. In this case, the recommendation device 10 is not limited to the recommendation, but the time and the number of people previously specified as the search condition, as well as information registered in the SNS account, for example, the user ID (identifier) and the contact information, etc. As an application to the facility, a request is made to a computer or the like operated by the facility to reserve the “Italian restaurant A”. When the cancel button B2 on the reservation confirmation screen Y displayed on the terminal device 50 is tapped, the terminal device 50 returns to the talk room display.

  As described above, the recommendation device 10 according to the present embodiment uses a talk room in which messages are exchanged in a chat format for specifying a search condition. For this reason, unlike the reservation using the reservation site, the search condition is not determined by the arbitrary decision of the user who makes the reservation, and the keyword as the search condition is extracted from the discussion among all users who use the facility. Therefore, according to the recommendation device 10 according to the present embodiment, the opinion of the entire user can be reflected in the reservation facility. In addition, the keyword is extracted from the newly posted message, and the facility is searched with a greater weight. For this reason, the keyword extracted from the message recently discussed between the users is easily used for the facility search. Therefore, even if there is a change in the story line between users, it is possible to search for facilities along the flow of conversation.

[2. Configuration of recommendation device]
Then, the structure of the recommendation apparatus 10 which concerns on embodiment is demonstrated using FIG. FIG. 2 is a diagram illustrating a configuration example of the recommendation device 10 according to the embodiment. As illustrated in FIG. 2, the recommendation device 10 includes a communication I / F (Interface) unit 11, a storage unit 13, and a control unit 15. The recommendation device 10 may include various functional units included in a known computer other than the functional units illustrated in FIG. 2, for example, functional units such as various input devices and audio output devices.

(About the communication I / F unit 11)
The communication I / F unit 11 is an interface that performs communication control with other devices such as the SNS server 20, the reservation server 30, and the terminal device 50. As an aspect of the communication I / F unit 11, a network interface card such as a LAN card can be employed.

  For example, the communication I / F unit 11 receives a log of a message posted from the SNS server 20 to the talk room, and transmits a facility recommendation, a reservation confirmation screen, and the like to the SNS server 20. Further, the communication I / F unit 11 transmits a facility search request including a search condition to the reservation server 30 and receives a search result from the reservation server 30.

(About the storage unit 13)
The storage unit 13 is a storage device that stores various programs such as an OS (Operating System) executed by the control unit 15 and a recommendation program that provides the recommendation service. As one aspect of the storage unit 13, a storage device such as a semiconductor memory element such as a flash memory, a hard disk, or an optical disk is given. The storage unit 13 is not limited to the type of storage device described above, and may be a RAM (Random Access Memory) or a ROM (Read Only Memory).

  The storage unit 13 stores log data 13 a as an example of data used for a program executed by the control unit 15. In addition to the log data 13a, other electronic data such as dictionary data used when analyzing morphemes from a character string included in text can also be stored.

(About log data 13a)
The log data 13a is data relating to a log of messages included in the talk room. As an aspect of the log data 13a, data in which the user ID of the SNS of the user who posted the message is associated with each message arranged in time series can be adopted. As an example, the log data 13a is registered in the storage unit 13 when collected from the SNS server 20 by the collection unit 15a described later. As another example, the log data 13a is referred to by an extraction unit 15b described later in order to extract a keyword for searching for a facility.

(About the control unit 15)
The control unit 15 has an internal memory for storing programs defining various processing procedures and control data, and executes various processes using these. As shown in FIG. 2, the control unit 15 includes a collection unit 15a, an extraction unit 15b, a search unit 15c, a display unit 15d, and a reservation unit 15e.

  The control unit 15 can be realized by causing a CPU (Central Processing Unit), an MPU (Micro Processing Unit), or the like to execute a privilege grant program. Each functional unit described above can also be realized by a hard wired logic such as an application specific integrated circuit (ASIC) or a field programmable gate array (FPGA).

(About the collection unit 15a)
The collection unit 15a is a processing unit that collects a message log from the talk room. As one aspect, the collection unit 15a collects a log of messages posted from the SNS server 20 to the talk room at the timing illustrated below. For example, each time a new message is posted in the talk room, the collection unit 15a collects a message log from the talk room. Further, the collection unit 15a posts from the SNS server 20 to the talk room every time a certain period, for example, 30 seconds or 1 minute elapses, or when the current time internally managed by the OS becomes a regular time. Collected logs After collecting the log, the collection unit 15a registers the log of the message collected from the SNS server 20 in the storage unit 13 as the log data 13a. Although the case where the log collected from the SNS server 20 is once registered in the storage unit 13 is illustrated here, it may be input to the subsequent extraction unit 15b as it is.

(About the extraction unit 15b)
The extraction unit 15b is a processing unit that extracts a keyword from a message included in the log data 13a. As one aspect, the extraction unit 15b extracts keywords related to location, category, time, and number of people from each message included in the log data 13a. For example, the extraction unit 15b performs morphological analysis on the text string of each message included in the log. Then, the extraction unit 15b extracts, as keywords, morphemes of nouns related to place, category, time, or number of people from among morphemes whose part of speech obtained as an analysis result is “noun” or “proprietary noun”.

(About the search unit 15c)
The search unit 15c is a processing unit that searches for a facility by assigning a greater weight to a keyword extracted from a newly posted message among the keywords extracted by the extraction unit 15b. As one aspect, the search unit 15c has a plurality of duplicate keywords as search condition items among keywords related to the place, category, time, and number of people extracted by the extraction unit 15c when requesting the reservation server 30 to search for a facility. It is determined whether or not to do. At this time, when there are a plurality of overlapping keywords as search condition items, the search unit 15c performs weighting on the plurality of keywords.

  First, the search unit 15c assigns a greater weight to a keyword extracted from a newly posted message among a plurality of existing keywords. Furthermore, the search part 15c performs the following weighting to the keyword which appears just before an affirmative word and a negative word among a plurality of existing keywords. For example, the search unit 15c gives a keyword having an affirmative word until the next keyword appears to a weight greater than the weight of a keyword having no affirmative word before the next keyword appears. . That is, the search unit 15c gives a large weight to a keyword that appears immediately before an affirmative word. In addition, the search unit 15c gives a keyword having a negative word before the next keyword appears to a weight smaller than the weight of a keyword having no negative word before the next keyword appears. . That is, the search unit 15c gives a small weight to the keyword that appears immediately before the negative word. In addition, the search unit 15c gives a smaller weight to a keyword extracted from a message posted by a user who has a large number of utterances among a plurality of existing keywords.

  After executing these weightings, the search unit 15c employs the keyword having the largest weight among the plurality of existing keywords as the search condition. In addition, when all of the four keywords of place, category, time, and number of people are available, the search unit 15c uses the keywords as search conditions and requests the reservation server 30 to search for facilities.

(About display unit 15d)
The display unit 15d is a processing unit that displays the facility recommendation searched by the search unit 15c in the talk room. As an aspect, the display unit 15d outputs the facility recommendation, which is responded as a search result from the reservation server 30, to the terminal device 50 held by the user participating in the talk room. In response to the output, the terminal device 50 displays the facility recommendation output by the display unit 15d as a banner advertisement in the talk room. After that, when an operation is performed on the recommendation displayed on the terminal device 50, the display unit 15d outputs a reservation confirmation screen of the facility where the recommendation has been performed to the terminal device 50 on which the tap operation has been performed. In response to the output, the terminal device 50 displays the reservation confirmation screen output by the display unit 15d. In addition, although the case where a banner advertisement is displayed is illustrated here, it may be a text advertisement.

(About reservation part 15e)
The reservation unit 15e is a processing unit that executes a reservation process for a facility that has made a recommendation. As one aspect, when the reservation execution operation is executed on the reservation confirmation screen displayed on the terminal device 50, the reservation unit 15e can determine the time and the number of people previously specified as the search condition, Information registered in the account, such as a user ID and contact information, is used as an application item for the facility, and a reservation for the facility is requested to a computer operated by the facility. Here, the case where the recommendation device 10 requests the facility computer to make a reservation is illustrated, but the facility reservation may be made via the reservation server 30.

[3-1. Specific example of weighting 1]
Next, a specific example 1 of keyword weighting will be described with reference to FIG. FIG. 3 is a diagram illustrating an example of a message log. Here, as an example, it is assumed that the weight of the initial state of each keyword is “1”, and the following coefficient is given by each weight. For example, “2” is assigned as a weighting factor to keywords extracted from the latest message among keywords that overlap as search condition items. A keyword appearing immediately before the affirmative word is given “1.5” as a weighting coefficient. Further, “0.5” is assigned as a weighting factor to a keyword that appears immediately before a negative word. In addition, each keyword that is duplicated as a search condition item is given a reciprocal number of the number of utterances of the user who posted the message from which the keyword was extracted.

  As shown in FIG. 3, “3: 3” is extracted as a keyword related to the number of people from the message T10 included in the log L2, and “Roppongi” is extracted as a keyword related to the location from the message T12. Further, “Chinese Chinese” is extracted from the message T13 as a keyword related to the category, and “Italian” is extracted from the message T14 as a keyword related to the category. Furthermore, “German food” is extracted as a keyword related to the category from the message T15, “7 o'clock” is extracted as a keyword related to the time from the message T16, and “Thai food” is extracted as a keyword related to the category from the message T17. In this example, keywords are not extracted from the remaining messages T11 and T18.

  In the example of FIG. 3, there are four keywords related to the category, “Chinese”, “Italian”, “German food”, and “Thai food”, which are duplicated as search condition items. Among these, “Thai food” is a keyword from which the latest message is extracted among the keywords related to the category, and therefore “2” is assigned as a weighting coefficient. Subsequently, since there is no keyword related to the category before the affirmative word or the negative word, weighting for the affirmative word and the negative word is not executed. Finally, a weight coefficient “1 / number of utterances” is assigned to each keyword. Of these four keywords, “Chinese”, “Italian”, and “Thai food” are the remarks of the user “XX”, and “German food” is the remarks of the user “Maruko”. Thus, “Chinese”, “Italian” and “Thai food” are given a weighting factor “1/3”, while “German food” is given a weighting factor “1/1”. become.

  As a result, the weight of “Chinese” and “Italian” is “0.333... (= 1/3)”, and the weight of “Thai food” is “0.666... (= 2/3)” The weight of “German cuisine” is “1 (= 1/1)”. Therefore, the keyword “German cuisine” is used as the search condition for the category. As a result, the recommendation R2 of the facility “German Restaurant D” is displayed at the bottom of the display area P2 of the terminal device 50.

  In this way, by assigning a smaller weight to a keyword extracted from a message posted by a user who has a large number of utterances, it is not drowned out by the utterance of the user “XX” who makes utterances about the category three times. It is possible to respect the remarks of the user “Maruko” who has remarked only once.

[3-2. Specific example 2 of weighting]
Next, specific example 2 of keyword weighting will be described with reference to FIG. FIG. 4 is a diagram illustrating an example of a message log. Here, the following description will be given on the assumption that the same weighting coefficient as in the first specific example is given.

  As shown in FIG. 4, “3: 3” is extracted as a keyword related to the number of people from the message T21 included in the log L3, and “Roppongi” is extracted as a keyword related to the location from the message T23. Further, “Chinese Chinese” is extracted from the message T24 as a keyword related to the category, and “German cuisine” is extracted from the message T25 as a keyword related to the category. Furthermore, “xx cooking” is extracted from the message T27 as a keyword related to the category, and “7 o'clock” is extracted from the message T32 as a keyword related to the time. Although not a keyword used for facility search, “good” is extracted as a positive word from message T26, “yada” is extracted as a negative word from message T28, and “bad” is negative from message T29. Extracted as

  In the example of FIG. 4, there are three categories-related keywords “Chinese”, “German cuisine”, and “XX cuisine”, which are duplicated as search condition items. Among these, “xx cooking” is a keyword from which the latest message is extracted among the keywords related to the category, and thus “2” is assigned as a weighting coefficient. However, the keyword that appears immediately before the negative word “I am weak” is “XX dish”, and the keyword that appears immediately before the negative word “Yada” is also “XX dish”. For this reason, “0.5” and “0.5” are given as weighting factors. On the other hand, since the keyword that appears immediately before the affirmative word “good” is “German cuisine”, “1.5” is assigned as a weighting coefficient. Finally, a weight coefficient “1 / number of utterances” is assigned to each keyword. Of these three keywords, “Chinese” and “German cuisine” are remarks of the user “XX”, and “XX cuisine” is remarks of the user “that person”. Therefore, a weighting factor “1/2” is assigned to “Chinese food” and “German cuisine”, while a weighting factor “1/1” is assigned to “XX cuisine”.

  As a result, the weight of “XX cuisine” is “0.5 (= 1 × 2 × 0.5 × 0.5 × 1)”, and the weight of “Chinese” is “0.5 (= 1/1 / 2) ”, and the weight of“ German cuisine ”is“ 0.75 (= 1 × 1.5 / 2) ”. Therefore, the keyword “German cuisine” is used as the search condition for the category. As a result, the recommendation R3 of the facility “German Restaurant D” is displayed at the bottom of the display area P3 of the terminal device 50.

  In this way, while assigning a large weight to a keyword that appears immediately before an affirmative word, a small weight is assigned to a keyword that appears immediately before a negative word, thereby reflecting the evaluation of keywords other than the speaker. As a result of performing the weighting, the opinions of the entire user can be reflected more accurately in the reservation facility.

[4. Processing procedure using recommendation device]
Next, a processing procedure by the recommendation device 10 according to the embodiment will be described. FIG. 5 is a flowchart illustrating a procedure of recommendation processing according to the embodiment. In the flowchart illustrated in FIG. 5, as an example, processing is executed in parallel for each talk room.

  As shown in FIG. 5, when a new message is written in the talk room (step S101 Yes), the collection unit 15a collects a message log from the talk room (step S102). Subsequently, the extraction unit 15b extracts keywords related to place, category, time, and number of people from each message included in the log collected in step S102 (step S103).

  Then, the search unit 15c determines whether or not the keyword extracted in step S103 is lacking in any of the items of place, category, time, or number of people (step S104). At this time, if there is a shortage of keywords (No in step S104), the facility cannot be narrowed down even if the facility search is executed, and the process returns to step S101.

  On the other hand, when there is no shortage of keywords (Yes in step S104), the search unit 15c gives a greater weight to a keyword extracted from a newly posted message among a plurality of existing keywords. At the same time, the search unit 15c gives a large weight to a keyword that appears immediately before an affirmative word among a plurality of keywords, while assigning a small weight to a keyword that appears just before a negative word (step S105). In addition, the search unit 15c gives a smaller weight to a keyword extracted from a message posted by a user who has a large number of utterances among a plurality of existing keywords (step S106).

  After that, the search unit 15c adopts the keyword having the highest weight among a plurality of overlapping keywords as the search condition items, and searches for facilities using the keywords of the four items of place, category, time, and number of people as the search conditions. The reservation server 30 is requested (step S107).

  And the display part 15d outputs the recommendation of the facility responded as a search result from the reservation server 30 to the terminal device 50 which the user who participates in a talk room has (step S108). Thus, the terminal device 50 displays the facility recommendation as a banner advertisement in the talk room.

  Subsequently, when an operation is performed on the recommendation displayed on the terminal device 50 (step S109 Yes), the display unit 15d outputs the reservation confirmation screen of the facility that performed the recommendation to the terminal device 50 on which the tap operation has been performed. (Step S110). In response to the output, the terminal device 50 displays the reservation confirmation screen output by the display unit 15d. If no operation is performed on the recommendation (No in step S109), the process returns to step S101.

  After that, when a reservation execution operation is executed on the reservation confirmation screen displayed on the terminal device 50 (Yes in step S111), the reservation unit 15e determines the time and number of people previously specified as the search condition, Information registered in the SNS account, such as a user ID or contact information, is used as an application item for the facility, and a reservation is made to the computer operated by the facility (step S112), and the process is terminated. If the reservation execution operation is not executed on the reservation confirmation screen (step S111: No), the process returns to step S101.

[5. Modifications and application examples]
The recommendation device 10 according to the above embodiment may be implemented in various different forms other than the above embodiment. Therefore, in the following, another embodiment of the recommendation device 10 will be described.

[5-1. Multiple recommendation display]
In the above embodiment, the case where the recommendation device 10 displays one facility recommendation at the bottom of the display area of the terminal device 50 is exemplified, but a plurality of facility recommendations can also be displayed. When a plurality of facility recommendations are displayed, the recommendation device 10 according to the present embodiment can display facility recommendations in association with a message including a keyword used for facility search.

  FIG. 6 is a diagram illustrating an example of a recommendation display method. As shown in FIG. 6, “3: 3” is extracted from the message T41 included in the log L4 as a keyword related to the number of people, and “Roppongi” is extracted from the message T43 as a keyword related to the place. “Chinese Chinese” is extracted from the message T44 as a keyword related to the category, and “Italian” is extracted from the message T45 as a keyword related to the category. Further, “German food” is extracted as a keyword related to the category from the message T46, and “7 o'clock” is extracted as a keyword related to the time from the message T48.

  In the example of FIG. 6, there are three keywords related to the category “Chinese”, “Italian”, and “German cuisine”, which are duplicated as search condition items. In this case, the recommendation device 10 displays the recommendation R10 of the Chinese restaurant C in association with the message T44 including the keyword “Chinese”, and recommends the recommendation R11 of the Italian restaurant A in association with the message T45 including the keyword “Italian”. And the recommendation R12 of the German restaurant D is displayed in association with the message T46 including the keyword “German cuisine”. Thereby, the user who participates in the talk room can think whether it deserves a reservation, comparing between several facilities by a banner advertisement. Moreover, since the recommendation of the facility is displayed like a stamp, the design can be improved.

  As described above, when a plurality of facility recommendations are displayed on the terminal device 50, the voting buttons B10 to B12 may be further displayed on each of the three facility recommendations R10 to R12. For example, if the voting button B10 of the recommendation R10 of the Chinese restaurant C displayed on the terminal device 50 is tapped, is the recommendation device 10 agreeing to go to the Chinese restaurant C or vice versa? Is displayed on the terminal device 50. As shown in FIG. 6, the vote screen is provided with an approval button B20 and an opposite button B21. Each member who participates in the talk room can indicate whether he / she agrees to go to the Chinese restaurant C by tapping either the approval button B20 or the opposite button B21.

[5-2. Recommendation stage display]
In addition, as shown in FIG. 6, the recommendation device 10 can also display a plurality of facility recommendations by changing the size of a banner and the level of detail of an advertisement between facilities. For example, the larger the number of votes voted for a facility, the larger the banner advertisement of the facility or the more detailed the content of the advertisement. On the contrary, as the number of votes voted against the facility increases, the banner advertisement of the facility can be made smaller or the contents of the advertisement can be made rougher.

  FIG. 7 is a diagram illustrating an example of a recommendation display method. The message log L4 shown in FIG. 7 is the same as the message log L4 shown in FIG. As shown in FIG. 7, among the recommendation R10 of the Chinese restaurant C, the recommendation R11 of the Italian restaurant A, and the recommendation R120 of the German restaurant D, the vote rate of the vote for the German restaurant D is a predetermined threshold, for example, more than half. . For this reason, the recommendation R120 of the German restaurant D is larger than the recommendation R10 of the Chinese restaurant C and the recommendation R11 of the Italian restaurant A, and the contents of the advertisement are detailed. By executing the recommendation stage display, it is possible to intuitively understand which facilities other users are highly interested in from the size and level of detail of the display. Furthermore, not only the vote button B12 but also the reservation button B120 are displayed together on the recommendation R120 of the facilities exceeding half. When the reservation button B120 is tapped, the reservation process is executed without displaying the reservation confirmation screen. As a result, it is also possible to guide a facility widely accepted by users participating in the talk room to a reservation with one touch.

[5-3. [Recommend split display]
In addition, as shown in FIG. 6, the recommendation device 10 can also divide the display area of the terminal device 50 into a message display area and a recommendation display area when displaying a plurality of facility recommendations.

  FIG. 8 is a diagram illustrating an example of a recommendation display method. As shown in FIG. 8, “3: 3” is extracted from the message T51 included in the log L5 as a keyword related to the number of people, and “Roppongi” is extracted from the message T53 as a keyword related to the place. “Chinese Chinese” is extracted from the message T54 as a keyword related to the category, and “Italian” is extracted from the message T55 as a keyword related to the category. Further, “German food” is extracted from the message T57 as a keyword related to the category, and “7 o'clock” is extracted from the message T61 as a keyword related to the time.

  As shown in FIG. 8, when a predetermined division operation is performed on the terminal device 50, for example, a button tap is performed, a boundary line Z is set as a boundary in the display area P6, and a message is displayed on the right side. On the left side, facility recommendations R30 to R32 and their voting buttons B30 to B32 are displayed. As described above, when the display is performed by dividing the message display area and the recommendation display area, the recommendations can be aligned along the timeline without being associated with the message. For this reason, when a recommendation is displayed in association with a message, the recommendation associated with the old message cannot be displayed on the screen, but at the time of division, the recommendation associated with the old message is displayed. Can do.

[5-4. Remind]
As an application example of the above-described embodiment, the recommendation device 10 according to the present embodiment can also call attention when the set time set before the reservation time is reached.

  For example, the recommendation device 10 determines whether or not the current time internally managed by the OS or the like is a time before the reservation time and set in advance, for example, one hour before or on the current day of the reservation time. To do. Then, the recommendation device 10 displays the remind screen shown in FIG. 9 on the terminal device 50A when the current time reaches the set time. FIG. 9 is a diagram illustrating an example of a remind screen. FIG. 9 illustrates a remind screen that is notified to the terminal device 50A when the date of the successful combination is reached. The display of the remind screen shown in FIG. 9 allows the user to remember that there was a joint party schedule at German restaurant D in Roppongi from 19:00 on August 28th of today. Necessary cash can be prepared in advance by displaying the budget on the remind screen in conjunction with the reminder of the joint party. Furthermore, when the “view map” button B50 displayed on the remind screen shown in FIG. 9 is pressed, a map from the office to the German restaurant D in Roppongi, which is the meeting place, can be displayed. As a result, the user can arrive at the site without delaying the reservation time and without hesitation.

[5-5. Performance-based billing]
As an application example of the above-described embodiment, the recommendation device 10 according to the present embodiment can also charge the facility for a customer's consideration when a facility reservation is established.

[5-6. Distributed and integrated]
In addition, each component of each illustrated apparatus does not necessarily need to be physically configured as illustrated. In other words, the specific form of distribution / integration of each device is not limited to that shown in the figure, and all or a part thereof may be functionally or physically distributed or arbitrarily distributed in arbitrary units according to various loads or usage conditions. Can be integrated and configured. For example, the collection unit 15a, the extraction unit 15b, the search unit 15c, the display unit 15d, or the reservation unit 15e may be connected as an external device of the recommendation device 10 via a network. In addition, the collection unit 15a, the extraction unit 15b, the search unit 15c, the display unit 15d, or the reservation unit 15e each have a separate device, and the functions of the recommendation device 10 are realized by being connected to a network and cooperating. You may do it. That is, the recommendation device 10 may be realized by a plurality of server computers, and depending on the function, the configuration may be flexibly realized by calling an external platform or the like with an API (Application Programming Interface) or network computing. Can change.

[5-7. Recommendation Program]
The recommendation device 10 in the above embodiment is realized by a computer 100 having a configuration as shown in FIG. 10, for example. FIG. 10 is a hardware configuration diagram illustrating an example of a computer 100 that implements the function of the recommendation device 10. The computer 100 includes a CPU 110, a RAM 120, a ROM 130, an HDD 140, a communication interface (I / F) 150, an input / output interface (I / F) 160, and a media interface (I / F) 170.

  The CPU 110 operates based on a program stored in the ROM 130 or the HDD 140 and controls each unit. The ROM 130 stores a boot program executed by the CPU 110 when the computer 100 is started up, a program depending on the hardware of the computer 100, and the like.

  The HDD 140 stores a program executed by the CPU 110, data used by the program, and the like. The communication interface 150 receives data from other devices via the network N and sends the data to the CPU 110, and transmits data generated by the CPU 110 to other devices via the network N.

  The CPU 110 controls an output device such as a display and a printer and an input device such as a keyboard and a mouse via the input / output interface 160. The CPU 110 acquires data from the input device via the input / output interface 160. Further, the CPU 110 outputs the generated data to the output device via the input / output interface 160.

  The media interface 170 reads a program or data stored in the recording medium 180 and provides it to the CPU 110 via the RAM 120. The CPU 110 loads the program from the recording medium 180 onto the RAM 120 via the media interface 170, and executes the loaded program. The recording medium 180 is, for example, an optical recording medium such as a DVD (Digital Versatile Disc) or PD (Phase change rewritable disk), a magneto-optical recording medium such as an MO (Magneto-Optical disk), a tape medium, a magnetic recording medium, or a semiconductor memory. Etc.

  When the computer 100 functions as the recommendation device 10 in the above-described embodiment, the CPU 110 of the computer 100 executes a program loaded on the RAM 120, thereby collecting the collection unit 15a, the extraction unit 15b, the search unit 15c, and the display unit 15d. And each function of the reservation part 15e is implement | achieved. The HDD 140 stores data in the storage unit 13.

  The CPU 110 of the computer 100 reads these programs from the recording medium 180 and executes them, but as another example, these programs may be acquired from other devices via the network N.

DESCRIPTION OF SYMBOLS 10 Recommendation apparatus 11 Communication I / F part 13 Storage part 13a Log data 15 Control part 15a Collection part 15b Extraction part 15c Detection part 15d Display part 15e Reservation part 20 SNS server 30 Reservation server 50A, 50B, 50N Terminal apparatus

Claims (10)

A collection unit that collects message logs from the room where messages posted by users are posted;
Of the logs collected by the collection unit, an extraction unit that extracts keywords from messages collected in the predetermined room;
A recommendation device comprising: a search unit that searches for a facility by assigning a greater weight to a keyword extracted from a newly posted message among the keywords extracted by the extraction unit. The extraction unit includes:
Extract at least one keyword for location, category, time and number of people
The search unit
The recommendation apparatus according to claim 1, wherein a facility that satisfies a condition specified by at least one keyword related to the place, category, time, and number of people extracted by the extraction unit is searched. The search unit
The recommendation apparatus according to claim 1 or 2, wherein a facility is searched by assigning a weight determined by the number of postings of a user who posted a message including the keyword to the keyword extracted by the extraction unit. The extraction unit includes:
Further extracting an affirmative word or a negative word contained in a message posted after the message from which the keyword is extracted;
The search unit
The recommendation device according to claim 1, wherein a facility is searched by assigning a weight to a keyword that appears immediately before an affirmative word and a negative word extracted by the extraction unit. A display unit for displaying the recommendations of the facility searched by the search unit in the room;
The recommendation device according to any one of claims 1 to 4, wherein the display unit displays a recommendation of an available facility corresponding to the keyword among the facilities searched by the search unit. . The display unit
The facility recommendation searched by the search unit is displayed in association with a message in which a keyword used for searching the facility is extracted from messages displayed on the timeline of the room. The recommendation device according to claim 5. The display unit
The recommendation device according to claim 5 or 6, wherein a display mode of the recommendation of the facility is changed according to a voting result for the recommendation of the facility displayed in association with the message. The display unit
The display area of the terminal device in which the room is displayed is divided into a message display area and a facility recommendation display area to display the message and the facility recommendation. The recommendation apparatus as described in any one. Computer
A collection process that collects a log of messages from the room where messages posted by users are posted;
An extraction step of extracting keywords from messages collected in the predetermined room among the logs collected by the collection step;
And a search step of searching for a facility by assigning a greater weight to a keyword extracted from a newly posted message among the keywords extracted in the extraction step. On the computer,
A collection procedure that collects a log of messages from the room where messages posted by users are posted;
An extraction procedure for extracting keywords from messages collected in the predetermined room out of the logs collected by the collection procedure;
A recommendation program, wherein a keyword extracted from a newly posted message among the keywords extracted by the extraction procedure is assigned a greater weight to search for a facility is executed.

Images