Conversation device and communication system

Abstract

A conversation device 20 functioning as a conversation bot for conversing with a user is provided with a first determination unit 230a, a first selection unit 230g, and a transmission control unit 230h. Upon receipt of an inquiry from the user, the first determination unit 230a determines an answer to the inquiry. The first selection unit 230g selects one piece of conversation information on the basis of a degree of relevance with a proposal to be presented to the user, from among a plurality of pieces of conversation information corresponding one-to-one to a plurality of conversations carried out in the past between the user and a conversation partner of the user. The transmission control unit 230h transmits, to the user, a response including the answer determined by the first determination unit 230a, a summary of the conversation represented by the conversation information selected by the first selection unit 230g, and the proposal to be presented to the user.

Description

Conversation device and communication system

[0001] The present disclosure relates to a conversation device and a communication system.

[0002] In recent years, there are cases where a help desk or a call center that responds to inquiries from users is realized by a robot that converses with users (hereinafter referred to as a conversation bot). The conversation by the conversation bot is preferably a natural conversation like a conversation between people. Patent Document 1 discloses an invention of a conversation bot that can naturally conduct a personal conversation according to a user without the user having to pre-register their personal information.

[0003] Japanese Unexamined Patent Application Publication No. 2022 - 168015

[0004] In a help desk or a call center provided by a company, it is desirable to be able to provide detailed customer support tailored to each customer. However, the technology disclosed in Patent Document 1 has problems such as being unable to provide detailed customer support tailored to each customer.

[0005] An object of the present disclosure is to provide a technology that enables the realization of detailed customer support tailored to each customer.

[0006] A conversation device according to an aspect of the present disclosure includes a first determination unit that determines an answer to a question when the question is received from a user's terminal via a communication device, a first selection unit that selects first conversation information from among a plurality of conversation information that corresponds one-to-one to a plurality of past conversations conducted between the user and the user's conversation partner, based on the degree of relevance to a proposal presented to the user, and a transmission control unit that transmits a response including the answer, a summary of the conversation represented by the first conversation information, and the proposal to the terminal via the communication device.

[0007] Furthermore, a communication system according to one aspect of the present disclosure comprises a user's terminal and a conversation device that communicates with the terminal via a communication network, wherein the conversation device, upon receiving a question from the terminal via the communication network, performs the following: determining an answer to the question; selecting a first conversation piece from among a plurality of conversation pieces that correspond one-to-one to a plurality of past conversations between the user and the user's conversation partner, based on the degree of relevance to a proposal to be presented to the user; and transmitting a response including the answer, a summary of the conversation represented by the first conversation piece, and the proposal to the terminal via the communication network; and the terminal performs the following: transmitting a question input by the user to the conversation device via the communication network; and outputting a response received from the conversation device via the communication network.

[0008] According to one aspect of this disclosure, it is possible to provide detailed customer support tailored to each individual customer.

[0009] This is a block diagram showing the configuration of a communication system 1 including a conversation device 20 according to one embodiment of the present disclosure. This is a diagram showing an example of a conversation between a conventional conversation bot and a user. This is a block diagram showing the configuration of the conversation device 20. This is a diagram showing an example of the configuration of the first management table TBL1. This is a diagram showing an example of the configuration of the second management table TBL2 and an example of conversation information stored in the second management table TBL2. This is a diagram for explaining the processing performed by the estimation unit 230b. This is a diagram for explaining the processing performed by the second determination unit 230d. This is a diagram showing an example of a conversation in this embodiment. This is a flowchart showing the processing flow in a conversation method executed by the processing device 230 according to the program PR1.

[0010] A. Figure 1 of the embodiment shows the overall configuration of the communication system 1 according to the embodiment. As shown in Figure 1, the communication system 1 comprises terminals 10[1], terminals 10[2], ... terminals 10[m], ... terminals 10[n] and a conversation device 20. n is an integer of 1 or more. m is an integer of 1 or more and n or less. In this embodiment, terminals 10[1] to 10[n] have the same configuration. However, terminals with different configurations may be included. In the following description, terminal 10[m] may be described as a representative example of terminals 10[1] to 10[n].

[0011] In the communication system 1, terminals 10[1] to 10[n] and the conversation device 20 are connected to each other via a communication network NW so that they can communicate with one another. In Figure 1, user U[m] uses terminal 10[m]. In this embodiment, terminal 10[m] is a smartphone, but it may also be a tablet terminal. Also, terminal 10[1] may be a stationary or portable personal computer.

[0012] The conversational device 20 is a device operated and managed by carrier X, which provides communication services (hereinafter referred to as Service A) via a communication network NW. The conversational device 20 functions as a conversational bot that converses with user U[m] via the communication network NW. In this embodiment, the conversational device 20 realizes a call center function related to Service A provided by carrier X. The call center function is a function that provides answers to inquiries from user U[m].

[0013] An example of a conversation between user U[m] and the conversation device 20 is a combination of an inquiry regarding service A provided by business operator X and a response to that inquiry. In this embodiment, the conversation between user U[m] and the conversation device 20 also includes small talk between user U[m] and the conversation device 20. An example of small talk between user U[m] and the conversation device 20 is a combination of user U's statements regarding their hobbies or preferences, i.e., statements regarding user U[m]'s interests, and responses from the conversation device 20, such as nods of agreement, to those statements.

[0014] In conventional conversational bots, conversations with user U[m] are conducted as follows. For example, suppose user U[m] is already subscribed to service A and wishes to change their pricing plan for service A. In this case, as shown in Figure 2, user U[m] inputs text US1, which represents an inquiry about changing the pricing plan for service A, such as "I would like to change the pricing plan for service A," into terminal 10[m].

[0015] Terminal 10[m] transmits text data representing the content of the input inquiry (hereinafter referred to as inquiry data) to a conversational bot functioning as a call center via a communication network NW. This conversational bot sends back to Terminal 10[m] a text response MS1 to the inquiry whose content is represented by the inquiry data received via the communication network NW, for example, text data representing "We are taking your inquiry at the counter" (hereinafter referred to as response data). As shown in Figure 2, Terminal 10[m] displays the response text MS1 represented by the response data received via the communication network NW. By user U[m] viewing the response text MS1 displayed on Terminal 10[m], one round-trip conversation between user U[m] and the chatbot is realized. Hereafter, regardless of whether it is a question to the call center or small talk, user U[m]'s statements to the conversational device 20 will be referred to as inquiries, and the statements of the conversational device 20 to user U[m] will be referred to as answers.

[0016] Figure 3 shows an example of the configuration of the conversation device 20. As shown in Figure 3, the conversation device 20 comprises a communication device 210, a storage device 220, a processing device 230, and a bus 240 that connects these devices to each other.

[0017] The communication device 210 includes a communication circuit connected to a communication network NW by wireless or wired means. The communication device 210 communicates with other devices connected to the communication network NW. Examples of these other devices include terminals 10[1] to 10[n]. The communication device 210 provides the processing unit 230 with query data received from terminal 10[m] via the communication network NW. The communication device 210 also transmits response data received from the processing unit 230 to terminal 10[m] via the communication network NW.

[0018] The storage device 220 is a recording medium that can be read by the processing device 230. The storage device 220 includes, for example, non-volatile memory and volatile memory. Non-volatile memory includes, for example, ROM (Read Only Memory), EPROM (Erasable Programmable Read Only Memory), and EEPROM (Electrically Erasable Programmable Read Only Memory). Volatile memory includes, for example, RAM (Random Access Memory). The storage device 220 stores the first management table TBL1, the second management table TBL2, and the program PR1.

[0019] Figure 4 shows an example of the contents stored in the first management table TBL1. The first management table TBL1 stores multiple pieces of proposal candidate information, which represent candidate proposals (hereinafter referred to as "proposal candidates") to be presented to a user who has made a question to the call center, along with the answer to that question. As shown in Figure 4, the multiple proposal candidates in this embodiment are three types of video distribution services: "animation distribution service," "sports video distribution service," and "pet video distribution service," and all three of these services are services associated with Service A. Service A is an example of the first service in this disclosure, and the proposal candidate services indicated by the proposal candidate information are an example of the second service in this disclosure. The business operator X has determined the priority order for sales expansion for these multiple proposal candidates. In Figure 4, the priority information stored in the first management table in association with the proposal candidate information represents the priority order for sales expansion for the proposal candidate indicated by the proposal candidate information. In this embodiment, an integer of 1 or more is set for the priority information, and a smaller value of the priority information means a higher priority for sales expansion. As shown in Figure 4, the priority of the "animation streaming service" is "1", the priority of the "sports video streaming service" is "2", and the priority of the "pet video streaming service" is "3". Therefore, in this embodiment, the "animation streaming service" has the highest priority, followed by the "sports video streaming service", and the "pet video streaming service" has the lowest priority.

[0020] Figure 5 shows an example of the contents stored in the second management table TBL2. The second management table TBL2 stores date and time information indicating the date and time a casual conversation took place between the user and the conversation device 20, and conversation information representing the content of the conversation, associated with user identification information that uniquely identifies the user who has registered to use service A. The conversation information is text data representing one or more exchanges of text related to casual conversation between the user and the conversation device 20. As shown in Figure 5, in this embodiment, multiple conversation information is stored that corresponds one-to-one with multiple conversations that took place between the user identified by the user identification information and the conversation device 20, associated with the user identification information. Since casual conversation often reflects the user's interests, in this embodiment, the conversation device 20 determines that a conversation that began with the receipt of inquiry data is a casual conversation if the text represented by the inquiry data contains words belonging to a group of words assumed to be included in conversations related to the user's interests. The group of words assumed to be included in conversations related to the user's interests is predetermined by the service provider X. This group of words includes multiple types of words, such as words indicating the genre of interest (e.g., "mountain climbing," "sports," "alcohol," "anime," etc.) and words indicating more detailed classifications such as sports and types of alcohol. If the text represented by the inquiry data contains words belonging to the above group of words, the conversation device 20 stores conversation information representing the full text of the conversation and date and time information indicating the date and time of the conversation in the second management table TBL2, associated with the user identification information of the user who sent the inquiry data.

[0021] In Figure 5, "UID[m]" is the user identification information of user U[m]. Figure 5 shows that on August 1, 2024, casual conversation took place between user U[m] and the conversation device 20 regarding "climbing Mt. Fuji with a friend". Also in Figure 5, casual conversation took place between user U[m] and the conversation device 20 regarding "drinking beer every night" on August 2, 2024. Furthermore, in Figure 5, casual conversation took place between user U[m] and the conversation device 20 regarding "recently starting to play tennis" on August 3, 2024. As a concrete example of conversation information, Figure 5 also shows the specific content of the casual conversation regarding "recently starting to play tennis". In Figure 5, "U:" means a statement by user U, and "M:" means a statement by the conversation device 20.

[0022] The processing unit 230 includes one or more CPUs (Central Processing Units). One or more CPUs are examples of one or more processors. Each of the processors and CPUs is an example of a computer. The processing unit 230 reads program PR1 from the storage device 220, for example, when the conversation device 20 is powered on. By executing the read program PR1, the processing unit 230 functions as the first decision unit 230a, estimation unit 230b, second selection unit 230c, second decision unit 230d, third decision unit 230e, fourth decision unit 230f, first selection unit 230g, and transmission control unit 230h shown in Figure 3.

[0023] The first decision unit 230a determines an answer to a question when it receives a question from user U[m], that is, when it receives inquiry data representing a question from the communication device 210 to the call center from terminal 10[m] via the communication network NW. If the text of the inquiry data received from the communication device 210 contains the word "Service A", the first decision unit 230a determines that the inquiry data represents a question from the user to the call center and determines an answer to that question. Existing technologies may be used as appropriate for the specific method of determining the answer to the question.

[0024] The estimation unit 230b estimates the interests of user U[m] by analyzing the past behavior of user U[m], the source of the inquiry data. Figure 6 is a diagram illustrating the process performed by the estimation unit 230b. First, the estimation unit 230b obtains history information D1 representing the history of user U[m]'s past actions. History information D1 is a list in which action information indicating user U[m]'s actions and date and time information indicating the date and time when the action was performed are arranged in ascending order of the date and time indicated by the date and time information. For example, if history information D1 shows the history of user U[m]'s access to a web page using service A, the action information is the identification information of the web page accessed by user U[m]. This history information D1 is recorded on the server that provides service A. The estimation unit 230b obtains history information D1 related to user U[m] by communicating with the server. Next, the estimation unit 230b accesses the web page indicated by the behavioral information included in the history information D1 and identifies the genre to which the article published in the web news belongs. Then, the estimation unit 230b aggregates the number of web page accesses by user U[m] for each article genre, estimates the genre with the most views to be the object of user U[m]'s interest, and outputs result information D2 representing the estimation result. For example, if the web page with the most views was an article about "sports," the estimation unit 230b outputs result information D2 representing "sports." Similarly, if the web page with the most views was an article about "animation," the estimation unit 230b outputs result information D2 representing "animation." Note that if the web page with the most views is an article about "sports" and the web page with the most views is an article about "animation," and the difference between the two is less than a threshold (for example, 10), the estimation unit 230b may output two types of result information D2: one representing "sports" and another representing "animation."

[0025] Another example of history information D1 is information representing the purchase history of user U[m] in online shopping using service A. A specific example of behavioral information in this case is information representing the items of products purchased by user U[m] through online shopping. When using this history information D1, the estimation unit 230b obtains the history information D1 by communicating with the server that implements the online shopping. The estimation unit 230b aggregates the number of products purchased by user U[m] through online shopping for each item and estimates the genre corresponding to the item with the most purchases as the object of interest for user U[m]. For example, if user U[m] purchased the most sports equipment, including tennis rackets, the estimation unit 230b outputs result information D2 representing "sports". Also, if the most purchased alcoholic beverages such as beer were purchased, the estimation unit 230b outputs result information D2 representing "alcohol". Furthermore, if the number of sports goods purchased is the highest, followed by the number of different types purchased, and the difference between the two is less than a threshold (for example, 10), the estimation unit 230b may output result information D2 representing "sports" and result information D2 representing "alcohol."

[0026] Another specific example of history information D1 is the second management table TBL2, in which case conversation information corresponds to behavioral information. As mentioned above, the conversation information stored in the second management table TBL2 represents small talk conversations between user U[m] and the conversation device 20, and small talk often reflects user U[m]'s interests. When the second management table TBL2 is used as history information D1, the estimation unit 230b identifies the genre of conversation (genre of topic) represented by each conversation piece of information and aggregates the number of conversations for each conversation genre. Then, it estimates the conversation genre with the most conversations as the object of user U[m]'s interest and outputs result information D2 representing the estimation result.

[0027] The second selection unit 230c selects a first candidate proposal information from among the multiple candidate proposal information stored in the first management table TBL1, based on the result information D2 (i.e., the estimation result regarding the user U[m]'s interests by the estimation unit 230b). Specifically, the second selection unit 230c identifies the genre corresponding to each candidate proposal information for the multiple candidate proposal information stored in the first management table TBL1, and selects the candidate proposal information corresponding to the genre represented by the result information D2 as the first candidate proposal information. If multiple result information D2s are output from the estimation unit 230b, the second selection unit 230c selects the candidate proposal information with the highest priority from among the candidate proposal information corresponding to each of the multiple result information D2s as the first candidate proposal information. The following describes the case where "sports video streaming service" is selected as the proposal to be presented to user U[m].

[0028] The second determination unit 230d determines a summary vector for each of the multiple conversation information stored in the second management table TBL2, which is associated with user identification information indicating the user who sent the question. The summary vector in this embodiment is a vector whose elements are values ​​corresponding to the frequency of occurrence of words belonging to the aforementioned word group, and is a vector that has been normalized to have a length of 1. For example, suppose the above word group contains J words (W[1] to W[J]) (where J is an integer of 2 or more), and for conversation information D, the frequency of occurrence of word W[1] is a[1], the frequency of occurrence of word W[2] is a[2], ... the frequency of occurrence of word W[J] is a[J]. In this case, the second determination unit 230d determines the summary vector corresponding to conversation information D as "a[1] / S, a[2] / S, ... a[J] / S". S is the sum of the squares of the occurrence frequencies of each word belonging to the above word group in the conversation information D, i.e., a[1]² + a[2]² + ... + a[J]². Since the elements of the summary vector are values ​​corresponding to the occurrence frequency of words, they are 0 or positive values. The summary vector determined based on the conversation information is an example of the first vector in this disclosure.

[0029] More specifically, in this embodiment, the second decision unit 230d first applies existing technology to the conversation information to generate a summary of the entire conversation represented by the conversation information. To reduce processing load, the second decision unit 230d may generate summaries for up to K (K is an integer greater than or equal to 2; in this embodiment, 5) rounds of conversation from the start of the conversation. Next, the second decision unit 230d performs morphological analysis on the generated summary to extract multiple words contained in the summary. Then, based on the extracted multiple words and the aforementioned word group, the second decision unit 230d identifies the frequency of occurrence of each word belonging to the aforementioned word group, and determines a summary vector based on this identification result. For example, since "Mount Fuji" is a word related to "climbing," for conversation information concerning "climbing Mount Fuji," the second decision unit 230d determines "0.5,0,0…" as the summary vector, as shown in Figure 7. Furthermore, since "tennis" is a word related to "sports," the second decision unit 230d determines "0,0.5,0,..." as the summary vector for the conversation information regarding "having started playing tennis," as shown in Figure 7. Similarly, since "beer" is a word related to "alcohol," the second decision unit 230d determines "0,0,0.5,..." as the summary vector for the conversation information regarding "drinking beer every night," as shown in Figure 7.

[0030] The third decision unit 230e determines the aforementioned summary vector based on the proposal selected by the second selection unit 230c. For example, if the second selection unit 230c selects "sports video streaming service" as a proposal to present to user U, the third decision unit 230e determines "0,1,0,..." as the summary vector corresponding to that proposal. The summary vector determined based on the proposal selected by the second selection unit 230c is an example of the second vector in this disclosure.

[0031] The fourth determination unit 230f determines a similarity score for each of the multiple first vectors, representing the degree of similarity with the second vector. In other words, in this embodiment, multiple similarity scores corresponding one-to-one to the multiple first vectors are determined by the fourth determination unit 230f. More specifically, the fourth determination unit 230f calculates the dot product (scalar product) of the first vector and the second vector as an index representing the similarity between the first vector and the second vector. As mentioned above, in this embodiment, the values ​​of each component constituting the first vector are 0 or positive, and the values ​​of each component constituting the second vector are also 0 or positive. Since the first vector and the second vector are normalized to have a length of 1, the dot product of the first vector and the second vector will be a value in the range of 0 or greater and 1 or less. The closer the value of the dot product of the first vector and the second vector is to 0, the lower the similarity between the first vector and the second vector, and the closer it is to 1, the higher the similarity between the first vector and the second vector.

[0032] The first selection unit 230g selects conversation information corresponding to the first vector with the highest similarity to the second vector, i.e., the first vector with the largest dot product value with the second vector, based on multiple similarity values ​​that correspond one-to-one to multiple first vectors. For example, suppose "0.5,0,0…" is determined as the first vector for "climbing Mt. Fuji", "0,0.5,0,…" is determined as the first vector for "starting to play tennis", and "0,0,0.5,…" is determined as the first vector for "drinking beer every night". On the other hand, suppose "0,1,0,…" is determined as the second vector for the proposal selected by the second selection unit 230c. In this case, the similarity to "starting to play tennis" is the highest, so the first selection unit 230g selects conversation information related to "starting to play tennis". If there are multiple conversation pieces with the highest similarity, the first selection unit 230g selects the conversation piece that has date and time information indicating a date and time close to the current date and time from among the multiple conversation pieces with the highest similarity.

[0033] The transmission control unit 230h transmits response data to the terminal 10, which represents a response MS2 including the answer MS21 determined by the first determination unit 230a, a summary MS22 of the conversation representing the conversation information selected by the first selection unit 230g, and a proposal MS23 selected by the second selection unit 230c. Upon receiving the response data, the terminal 10 displays the response MS2 including the answer MS21, summary MS22, and proposal MS23, as shown in Figure 9. Since the proposal MS23 is tailored to the user U's interests, even if the summary MS22 is not included and only the answer MS21 and proposal MS23 are sent back to the user U, it is likely that the user U will subscribe to the "sports video streaming service" based on the proposal MS23 to some extent. According to this embodiment, in addition to the answer MS21 and proposal MS23, the summary MS22 is also presented to the user U. The summary MS22 plays the role of a sales pitch that naturally invites the user U to subscribe to the "sports video streaming service" based on the proposal MS23. Therefore, proposals regarding "sports video streaming services" are more appealing to user U, and an improvement in the probability of securing a subscription contract (conversion rate) is expected.

[0034] Furthermore, the processing unit 230, which operates according to program PR1, executes a conversation method that prominently features the characteristics of this disclosure upon receiving inquiry data related to a question via the communication network NW. Figure 9 is a flowchart showing the processing flow in this conversation method. As shown in Figure 9, this conversation method includes the processing of steps SA110 to SA140.

[0035] In step SA110, the processing unit 230 functions as a first determination unit 230a. In step SA110, the processing unit 230 determines the answer to the question represented by the received inquiry data.

[0036] In step SA120, which follows step SA110, the processing unit 230 functions as an estimation unit 230b and a second selection unit 230c. In step SA120, the processing unit 230 selects, from among a plurality of suggested candidate information stored in the first management table TBL1, a suggested candidate information representing one suggested candidate to be presented to the user who asked the question, based on the results obtained by analyzing the past behavior of user U. As described above, in this embodiment, "sports video streaming service" is selected as the suggested candidate to be presented to user U.

[0037] In step SA130, which follows step SA120, the processing unit 230 functions as a second determination unit 230d, a third determination unit 230e, a fourth determination unit 230f, and a first selection unit 230g. In step SA130, the processing unit 230 determines the aforementioned summary vector (first vector) for each of the multiple conversation information that has taken place in the past between the user who sent the inquiry data and the conversation device 20. Next, the processing unit 230 determines the aforementioned summary vector (second vector) based on the proposal selected in step SA120. Then, the processing unit 230 calculates the dot product of each of the multiple first vectors with the second vector and selects the conversation information corresponding to the first vector for which the largest dot product was calculated. As mentioned above, suppose the summary vector for "climbing Mt. Fuji" is "0.5,0,0...", the summary vector for "starting to play tennis" is "0,0.5,0,...", and the summary vector for "drinking beer every night" is "0,0,0.5,...". On the other hand, suppose the summary vector for the proposal selected in step SA120 is "0,1,0,...". In this case, the similarity to "starting to play tennis" is maximized, so in step SA130, conversational information related to "starting to play tennis" is selected.

[0038] In step SA140, which follows step SA130, the processing unit 230 functions as a transmission control unit 230h. In step SA140, the processing unit 230 transmits response data to the terminal 10, which represents a response MS2 including the answer MS21 determined in step SA110, a summary MS22 of the conversation representing the conversation information selected in step SA130, and a proposal MS23 selected in step SA120. Upon receiving the response data, the terminal 10 displays the response MS2 including the answer MS21, summary MS22, and proposal MS23, as shown in Figure 9. According to this embodiment, a more appealing proposal can be made to the user U, and the conversion rate can be improved.

[0039] B. Modifications The above embodiment can be modified as follows. Furthermore, the embodiment and each modification may be combined as appropriate. (1) The conversation information stored in the second management table TBL2 represents casual conversations that have taken place between user U and the conversation device 20 in the past, but the conversation information stored in the second management table TBL2 may represent casual conversations that have taken place between user U[m] and user U[k] in the past. Note that k is an integer in the range of 1 to n and is a different integer from m. In other words, the conversation partner of user U"[m]" regarding casual conversation is not limited to the conversation device 20, but may be another user. This is because even casual conversations with other users often reflect the hobbies or preferences of user U[m].

[0040] (2) The second management table TBL2 may store conversation information that represents conversations other than small talk. Specific examples of conversations other than small talk include discussions between user U[m] and the conversation partner, conversations in question-and-answer sessions between user U[m] and the conversation partner, and conversations in greetings exchanged between user U[m] and the conversation partner. When the conversation device 20 detects a conversation between user U[m] and the conversation partner, it may store conversation information representing the content of the conversation in the second management table TBL2, regardless of the type of conversation.

[0041] (3) Instead of the second management table TBL2, a table may be used in which summary information representing a summary of a conversation between the user indicated by the user identification information and the conversation partner is recorded in association with the user identification information, a summary vector generated based on the summary, and date and time information representing the date and time of the conversation. In this case, the second determination unit 230d may be omitted. Also, a summary vector generated based on the proposed candidate represented by the proposed candidate information may be stored in the first management table TBL1 in association with the proposed candidate information. In this case, the third determination unit 230e may be omitted.

[0042] (4) The first selection unit 230g may select the first conversation information based on the degree of relevance between each of the plurality of conversation information and the answer and the degree of relevance between each of the plurality of conversation information and the proposal. Specifically, the conversation device 20 may determine the above-described summary vector (third vector) for the answer presented to the user U[m], and select, as the first conversation information, the conversation information corresponding to the first vector having the maximum similarity to the third vector and also having the maximum similarity to the second vector.

[0043] (5) The estimation unit 230b and the second selection unit 230c are not necessarily essential. For example, one service associated with Service A may be predetermined as an expansion target by the business operator X, and an invitation to join the one service may be defined as a proposal presented to the questioner. Also, the proposal presented to the questioner may be a proposal that conforms to the hobby or preference of the questioner, and may be a proposal unrelated to Service A, such as an invitation to a service unrelated to Service A.

[0044] (4) The program PR1 may be manufactured or sold alone. Specific modes of providing the program PR1 when selling it include a mode of writing the program PR1 on a computer-readable recording medium such as a flash ROM and distributing it, or a mode of distributing the program PR1 by downloading via a telecommunication line such as the Internet.

[0045] C: Others (1) In the above-described embodiments, ROM, RAM, etc. were exemplified as the storage device 220. However, the storage device 220 may be a flexible disk, a magneto-optical disk (e.g., a compact disk, a digital versatile disk, a Blu-ray (registered trademark) disk), a smart card, a flash memory device (e.g., a card, a stick, a key drive), a CD-ROM (Compact Disk-ROM), a register, a removable disk, a hard disk, a floppy (registered trademark) disk, a magnetic strip, a database, a server, or other suitable storage media.

[0046] (2) In the above-described embodiments, the information, signals, etc. described may be represented using any of a variety of different technologies. For example, data, instructions, commands, information, signals, bits, symbols, chips, etc. that may be referred to throughout the above description may be represented by voltage, current, electromagnetic waves, magnetic fields or magnetic particles, optical fields or photons, or any combination thereof.

[0047] (3) In the above-described embodiments, the input and output information, etc. may be stored in a specific location (e.g., memory) or may be managed using a management table. The input and output information, etc. may be overwritten, updated, or appended. The output information, etc. may be deleted. The input information, etc. may be transmitted to other devices.

[0048] (4) In the above-described embodiments, the determination may be made based on a value represented by 1 bit (0 or 1), a truth value (Boolean: true or false), or a numerical comparison (e.g., comparison with a predetermined value).

[0049] (5) The processing procedures, sequences, flowcharts, etc. exemplified in the above-described embodiments may be reordered as long as there is no contradiction. For example, for the methods described in the present disclosure, the elements of various steps are presented using an exemplary order and are not limited to the specific order presented.

[0050] (6) Each function illustrated in Figure 3 is implemented by any combination of at least one of hardware and software. Furthermore, the method of implementing each function block is not particularly limited. That is, each function block may be implemented using one device that is physically or logically coupled, or it may be implemented using two or more physically or logically separated devices that are directly or indirectly connected (for example, using wired, wireless, etc.). A function block may also be implemented by combining the one or more devices with software.

[0051] (7) The programs illustrated in the embodiments described above should be broadly interpreted to mean instructions, instruction sets, code, code segments, program code, programs, subprograms, software modules, applications, software applications, software packages, routines, subroutines, objects, executable files, execution threads, procedures, functions, etc., whether they are called software, firmware, middleware, microcode, hardware description languages ​​or by any other name.

[0052] Furthermore, software, instructions, information, etc., may be transmitted and received via a transmission medium. For example, if software is transmitted from a website, server, or other remote source using at least one of wired technology (such as coaxial cable, fiber optic cable, twisted pair, or digital subscriber line (DSL)) and wireless technology (such as infrared or microwave), then at least one of these wired and wireless technologies is included in the definition of a transmission medium.

[0053] (8) In each of the above-mentioned forms, the terms “system” and “network” shall be used interchangeably.

[0054] (9) The information, parameters, etc. described in this disclosure may be expressed using absolute values, relative values ​​from a given value, or other corresponding information.

[0055] (10) In the embodiments described above, the portable device may be a Mobile Station (MS). A Mobile Station may also be referred to by those skilled in the art as a subscriber station, mobile unit, subscriber unit, wireless unit, remote unit, mobile device, wireless device, wireless communication device, remote device, mobile subscriber station, access terminal, mobile terminal, wireless terminal, remote terminal, handset, user agent, mobile client, client, or several other appropriate terms. In this disclosure, terms such as "mobile station," "user terminal," "user equipment (UE)," and "terminal" may be used interchangeably.

[0056] (11) In the embodiments described above, the terms “connected,” “coupled,” or any variation thereof means any direct or indirect connection or coupling between two or more elements, and may include the presence of one or more intermediate elements between two elements that are “connected” or “coupled” with each other. The coupling or connection between elements may be physical, logical, or a combination thereof. For example, “connection” may be read as “access.” As used in the present disclosure, two elements may be considered to be “connected” or “coupled” with each other using at least one of one or more wires, cables, and printed electrical connections, and, in some non-limiting and non-exclusive examples, electromagnetic energy having wavelengths in the radio frequency domain, microwave domain, and optical (both visible and invisible) domain.

[0057] (12) In the embodiments described above, the phrase "based on" does not mean "based solely on" unless otherwise specified. In other words, the phrase "based on" means both "based solely on" and "based at least on".

[0058] (13) The terms “determinating” and “deciding” as used in this disclosure may encompass a wide variety of actions. “Determinating” and “deciding” may include, for example, judging, calculating, computing, processing, deriving, investigating, looking up, search, inquiry (for example, searching in a table, database or other data structure), and confirming. Furthermore, "judgment" and "decision" may include considering something as a "judgment" or "decision" based on actions such as receiving (e.g., receiving information), transmitting (e.g., sending information), input, output, and access (e.g., accessing data in memory). Additionally, "judgment" and "decision" may include considering something as a "judgment" or "decision" based on actions such as resolving, selecting, choosing, establishing, and comparing. In short, "judgment" and "decision" may include considering something as a "judgment" or "decision" based on some action. Furthermore, "judgment (decision)" may be reinterpreted as "assuming," "expecting," or "considering."

[0059] (14) Where the terms “include,” “including,” and variations thereof are used in the embodiments described above, these terms are intended to be inclusive, as is the term “comprising.” Furthermore, the term “or” as used in this disclosure is not intended to be exclusive OR.

[0060] (15) In the present disclosure, if articles are added by translation, for example, a, an, and the in English, the present disclosure may include the fact that the noun following these articles is plural.

[0061] (16) In this disclosure, the term “A and B are different” may mean “A and B are different from each other.” The term may also mean “A and B are each different from C.” Terms such as “separate” and “combine” may be interpreted in the same way as “different.”

[0062] (17) Each aspect / embodiment described herein may be used individually, in combination, or switched between as needed during implementation. Furthermore, notification of certain information (e.g., notification that "it is X") is not limited to explicit notification, but may also be implicit (e.g., by not providing such notification).

[0063] 1...Communication system, 10...Terminal, 20...Conversation device server, 210...Communication device, 220...Storage device, 230...Processing device, 230a...First decision unit, 230b...Estimation unit, 230c...Second selection unit, 230d...Second decision unit, 230e...Third decision unit, 230f...Fourth decision unit, 230g...First selection unit, 230h...Transmission control unit, PR1...Program, TBL1...First management table, TBL2...Second management table.

Claims

1. A conversation device comprising: a first determination unit that determines an answer to a question when it is received from a user's terminal via a communication device; a first selection unit that selects first conversation information from among a plurality of conversation pieces of information that correspond one-to-one to a plurality of conversations that have taken place in the past between the user and the user's conversation partner, based on the degree of relevance with a proposal to be presented to the user; and a transmission control unit that transmits a response including the answer, a summary of the conversation represented by the first conversation piece of information, and the proposal to the terminal via the communication device.

2. The conversation device according to claim 1, comprising: an estimation unit that estimates the user's interests based on the user's behavior; and a second selection unit that selects a first proposal candidate from among a plurality of proposal candidates as the proposal, based on the user's interests estimated by the estimation unit.

3. The conversation device according to claim 2, wherein the user's actions are a plurality of conversations that have taken place in the past between the user and the user's conversation partner, and the estimation unit estimates the user's interests based on the plurality of conversations.

4. The conversation device according to claim 3, wherein each of the plurality of conversational pieces of information includes a topic relating to the user's interests.

5. The conversation device according to claim 1, wherein the first selection unit selects the first conversation information based on the degree of relevance between each of the plurality of conversations and the answer, and the degree of relevance between each of the plurality of conversations and the proposal.

6. The conversational device according to claim 2, wherein, if the question is a question relating to a first service, the proposal selected by the second selection unit includes an invitation to a second service accompanying the first service.

7. The conversation device according to claim 1, comprising: a second determination unit that determines a first vector for each of the plurality of conversation information, the frequency of occurrence of each of the plurality of words included in the conversation information as elements; a third determination unit that determines a second vector for each of the plurality of words included in the proposal as elements; and a fourth determination unit that determines a similarity score representing the degree of similarity between each of the plurality of first vectors that correspond one-to-one to the plurality of conversation information and the second vector, wherein the first selection unit selects the first conversation information corresponding to the first vector with the highest similarity score based on the plurality of similarity scores that correspond one-to-one to the plurality of first vectors determined by the fourth determination unit.

8. A communication system comprising a user's terminal and a conversation device that communicates with the terminal via a communication network, wherein the conversation device, upon receiving a question from the terminal via the communication network, performs the following: determining an answer to the question; selecting a first conversation piece from among a plurality of conversation pieces that correspond one-to-one to a plurality of past conversations between the user and the user's conversation partner, based on the degree of relevance to a proposal to be presented to the user; and transmitting a response including the answer, a summary of the conversation represented by the first conversation piece, and the proposal to the terminal via the communication network; and the terminal performs the following: transmitting a question input by the user to the conversation device via the communication network; and outputting a response received from the conversation device via the communication network.

Images