Information Acquisition Server, Program, and Information Acquisition System

Abstract

To provide an information acquisition server, a program, and an information acquisition system that acquire information that can be used for operator support and business improvement from utterance content in conversation.SOLUTION: An information acquisition server 1 for acquiring information from conversation between an operator and a customer includes: a candidate word storage unit 32 that stores candidate words in association with one or more classifications; a target word acquisition unit 14 that compares a word included in a text indicating utterance content in the conversation with the candidate words in the candidate word storage unit 32 and acquires a classification target word from the text; a classification acquisition unit 15 that acquires one or more classifications of the candidate word storage unit 32 corresponding to the classification target word acquired by the target word acquisition unit 14; and an associating unit 17 that associates, when the number of the classifications acquired by the acquisition unit 15 is more than one, one classification determined by using a vicinity word located in the vicinity of the classification target word of the text with the classification target word.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to an information acquisition server, a program, and an information acquisition system. [Background technology]

[0002] In call centers, operators must respond to various inquiries from customers while they are talking to them. For example, the operators must organize what the customers say and respond appropriately to their inquiries. For this reason, a system to support the operators is required. In such a situation, for example, a call center support system has been disclosed that includes a recording means for recording audio data of a conversation between a customer and an operator, an audio data conversion means for converting the audio data into text data, a word extraction means for comparing the text data with a word list in which words expressing emotions are pre-registered and extracting matching words, and a notification means for counting the number of extracted words by emotion, thereby converting the words expressing the emotions into numerical values ​​for each emotion, and notifying an administrator terminal that there has been a significant change in the emotion of at least one of the customer or the operator when the numerical values ​​exceed a predetermined threshold (for example, Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2016-092582 A Summary of the Invention [Problem to be solved by the invention]

[0004] The technique described in Patent Document 1 supports an operator by notifying a supervisor when the emotion of the operator or the customer significantly deteriorates. On the other hand, as a mechanism for supporting the operator, a mechanism is needed to support the operator's primary task of sorting out what the customer has said and providing an appropriate response to the inquiry, as described above. This becomes especially necessary when the operator lacks experience or when the inquiry is complicated. It is also important to analyze the content of conversations between operators and customers and make operational improvements based on the analysis results.

[0005] Therefore, an object of the present invention is to provide an information acquisition server, a program, and an information acquisition system that acquire information that can be used to support an operator and improve business operations from the content of utterances in a conversation. [Means for solving the problem]

[0006] The present invention solves the above problems by the following solving means. A first invention is an information acquisition server that acquires information from a conversation between an operator and a customer, the information acquisition server comprising: a candidate word memory unit that stores candidate words in correspondence with one or more classifications; a related word memory unit that stores related words related to the classifications; a target word acquisition means that acquires a classification target word from the text by comparing words included in a text indicating the content of the conversation with the candidate words in the candidate word memory unit; a classification acquisition means that acquires the classification in the candidate word memory unit that corresponds to the classification target word acquired by the target word acquisition means; a classification determination means that determines one of the classifications using the classification target word of the text and the related words in the related word memory unit that are related to the classification acquired by the classification acquisition means; and an association means that associates the one of the classifications determined by the classification determination means with the classification target word. A second invention is an information acquisition server according to the first invention, wherein the classification determination means determines the one classification by using a word having a modification relationship with the classification target word of the text. A third invention is an information acquisition server according to the first invention, wherein the classification determination means determines the one classification by using a neighboring word located in a vicinity of the classification target word in the text. A fourth invention is an information acquisition server according to any one of the first to third inventions, wherein the classification determination means selects the related words according to the frequency of occurrence of words contained in all of the text of the conversation, and determines the one classification. A fifth invention is an information acquisition server according to any one of the first to fourth inventions, which performs processing using the classification determination means when there are multiple classifications acquired by the classification acquisition means. A sixth invention is an information acquisition server according to the first invention, wherein, when there are multiple classifications acquired by the classification acquisition means, the classification determination means refers to the related word storage unit, compares nearby words in the text that are located in the vicinity of the word to be classified with the related words in the related word storage unit that correspond to one of the multiple classifications, acquires one of the related words, and determines the classification in the related word storage unit that corresponds to the one related word to be the one classification. A seventh invention is an information acquisition server according to the sixth invention, further comprising a synonym memory unit that stores a representative word and a synonym of the representative word in correspondence with each other, and when the classification determination means is unable to acquire the one related word, it refers to the synonym memory unit, acquires the one related word based on the synonym that matches the nearby word, and determines the classification of the related word memory unit corresponding to the one related word to be the one classification. An eighth invention is an information acquisition server according to any one of the first to seventh inventions, further comprising a synonym memory unit that stores a representative word and a synonym of the representative word in correspondence with each other, and the target word acquisition means, when it is unable to acquire the target word for classification by comparing a word included in the text with the candidate word in the candidate word memory unit, refers to the synonym memory unit and acquires the target word for classification based on the synonym that matches the word included in the text. A ninth invention is an information acquisition server according to any one of the first to eighth inventions, further comprising a phrase division means for dividing the text into phrases, and the target word acquisition means acquires the classification target word based on the word contained in the phrase. A tenth invention is an information acquisition server which is any of the first to ninth inventions, and which is provided with a utterance data receiving means for receiving utterance data from a speaker, and the target word acquisition means acquires the classification target word from the text corresponding to the utterance data in response to the utterance data being accepted by the utterance data receiving means. An eleventh aspect of the present invention is the information acquisition server according to the tenth aspect of the present invention, wherein the utterance data receiving means receives the text indicating utterance content. A twelfth invention is an information acquisition server according to the tenth invention, wherein the utterance data receiving means receives voice data including utterance content, and the information acquisition server is provided with a text acquisition means for performing voice recognition processing on the voice data received by the utterance data receiving means, and acquiring the text. A thirteenth invention is an information acquisition server, comprising any one of the tenth to twelfth inventions, which is equipped with a speaker associating means for associating the text with speaker identification information that identifies the speaker. A fourteenth aspect of the present invention is a program for causing a computer to function as any one of the information acquisition servers according to the first to thirteenth aspects of the present invention. A fifteenth invention is an information acquisition system comprising an information acquisition server according to any one of the first to thirteenth inventions and a terminal used by the operator, wherein the information acquisition server comprises an output means for outputting classification information including the classification and the classification target word associated by the association means to the terminal. A 16th invention is an information acquisition system according to the 15th invention, wherein the information acquisition server comprises an item storage unit that pre-stores classification items corresponding to the requirements of the customer, and an item update output means that updates the classification items in the item storage unit that correspond to the classification associated by the associating unit, and outputs the classification items to the terminal. A 17th invention is the information acquisition system of the 16th invention, wherein the information acquisition server includes a requirement inference means for inferring requirements of the customer by referring to the item memory unit based on the classification and the classification target word associated by the associating means, and the item update output means outputs the classification item corresponding to the requirement inferred by the requirement inferred means to the terminal. Effect of the Invention

[0007] According to the present invention, it is possible to provide an information acquisition server, a program, and an information acquisition system that acquire information that can be used to support an operator and improve business operations from the contents of statements made in a conversation. [Brief description of the drawings]

[0008] [Figure 1] 1 is an overall configuration diagram of an information acquisition system according to an embodiment of the present invention and a functional block diagram of an information acquisition server. [Diagram 2] FIG. 4 is a diagram illustrating an example of a storage unit of an information acquisition server according to the embodiment. [Diagram 3] FIG. 4 is a diagram illustrating an example of a storage unit of an information acquisition server according to the embodiment. [Figure 4] 10 is a flowchart showing an information acquisition process of the information acquisition server according to the embodiment; [Diagram 5] 10 is a flowchart showing a text classification process of the information acquisition server according to the embodiment. [Figure 6] This is a continuation of Figure 5. [Figure 7] This is a continuation of Figure 6. [Figure 8] FIG. 11 is a diagram illustrating a processing example of an information acquisition server according to the embodiment. [Figure 9] FIG. 11 is a diagram illustrating a processing example of an information acquisition server according to the embodiment. [Figure 10] FIG. 4 is a diagram illustrating an example of a storage unit of an information acquisition server according to the embodiment. [Figure 11] FIG. 11 is a diagram showing an example of a display screen on an operator terminal according to the embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. Note that this is merely an example, and the technical scope of the present invention is not limited to this example. (Embodiment) FIG. 1 is an overall configuration diagram of an information acquisition system 100 according to this embodiment and a functional block diagram of an information acquisition server 1. As shown in FIG. 2 and 3 are diagrams illustrating an example of the storage unit 30 of the information acquisition server 1 according to the present embodiment.

[0010] <Information Acquisition System 100> 1, for example, an information acquisition system 100 receives utterance data in a conversation between a customer and an operator at a call center from a customer terminal 4 and an operator terminal 5 by an information acquisition server 1. The information acquisition system 100 is a system in which the information acquisition server 1 classifies words included in the utterance data, thereby acquiring information that can be used to support the operator and improve business operations from the contents of the utterances in the conversation.

[0011] The information acquisition system 100 includes an information acquisition server 1, a customer terminal 4 of each customer, and an operator terminal 5 (terminal) of each operator. The information acquisition server 1, the customer terminal 4, and the operator terminal 5 are communicatively connected to each other via a communication network N. Although FIG. 1 illustrates one customer terminal 4 and one operator terminal 5, there may be multiple customer terminals 4 and multiple operator terminals 5, each of which is connected to the information acquisition server 1. The communication network N, which will be described later, is a data communication network when text communication is performed between the customer terminal 4 and the operator terminal 5. In addition, when voice communication is performed between the customer terminal 4 and the operator terminal 5, the communication network N includes, for example, a voice communication network in addition to the data communication network.

[0012] In the following embodiment, the information acquisition system 100 will be described taking as an example an operator responding to inquiries from customers about insurance. However, the business in which the information acquisition system 100 can be used is not limited to this.

[0013] <Information Acquisition Server 1> The information acquisition server 1 is a device for acquiring information by classifying words included in utterance data indicating utterances between a customer and an operator. The information acquisition server 1 is owned, for example, by a company that operates a call center, or a company commissioned by the call center. The information acquisition server 1 may be configured, for example, by one server, or may be configured by multiple servers, or may be a cloud.

[0014] The information acquisition server 1 includes a control unit 10, a storage unit 30, and a communication interface unit 39. The control unit 10 is a central processing unit (CPU) that controls the entire information acquisition server 1. The control unit 10 appropriately reads and executes an operating system (OS) and application programs stored in the storage unit 30, thereby cooperating with the above-mentioned hardware and executing various functions. The control unit 10 includes a utterance data receiving unit 11 (utterance data receiving means, text acquisition means), a speaker associating unit 12 (speaker associating means), a phrase delimiting unit 13 (phrase delimiting means), a target word acquiring unit 14 (target word acquiring means), a classification acquiring unit 15 (classification acquiring means), a classification determining unit 16 (classification determining means), an associating unit 17 (associating means), an item updating unit 18 (requirement inference means, item update output means), and an information output unit 19 (output means, item update output means).

[0015] The utterance data receiving unit 11 receives utterance data from the speaker from the customer terminal 4 and the operator terminal 5. The utterance data receiving unit 11 may receive utterance data of text, which is a message, from the customer terminal 4 and the operator terminal 5, using, for example, a Web chat or the like, in which a customer and an operator transmit messages in real time. Furthermore, the utterance data receiving unit 11 may receive utterance data, which is voice data, from the customer terminal 4 and the operator terminal 5, for example, by using a telephone function. In the case of voice data, for example, the voice data of the utterances of both the customer and the operator can be obtained by a call acquisition device (not shown) installed between the operator's telephone (not shown) and the operator's microphone (not shown). The utterance data receiving unit 11 then performs voice recognition processing on the received voice data to obtain text indicating the utterances.

[0016] The speaker associating unit 12 associates identification information (speaker identification information), such as an ID (IDentification) that identifies a speaker, with the text acquired by the utterance data accepting unit 11. When the utterance data accepting unit 11 directly accepts text from the customer terminal 4 and the operator terminal 5, the speaker associating unit 12 may, for example, associate an ID corresponding to the sender with the text. When the utterance data accepting unit 11 accepts voice data, the speaker associating unit 12 may assign different IDs to the text based on the input direction of the voice data to the call acquisition device and associate the text with the IDs. Furthermore, for example, when multiple customers are speaking at the customer terminal 4, different IDs may be assigned to the text and associated with the text based on the characteristics of the voice data for each utterance, the directivity of the voice based on the position of the speaker, the volume, and the like. Then, the speaker associating unit 12 associates the text with the ID and stores it in the conversation storage unit 37.

[0017] The phrase separator 13 separates the text received by the utterance data receiver 11 into phrases and acquires words from the phrases. The words acquired from the phrases are, for example, single words. The target word acquisition unit 14 compares words included in phrases of the text acquired by the phrase delimiter unit 13 with candidate words in the candidate word storage unit 32 to acquire words to be classified from the text. Here, if there is a candidate word in the candidate word storage unit 32 that matches a word included in the text, the target word acquisition unit 14 acquires the candidate word as a word to be classified. If there is no candidate word in the candidate word storage unit 32 that matches a word included in the text, the target word acquisition unit 14 compares the word included in the text with synonyms in the synonym storage unit 34 to check whether there is a matching synonym. Then, if there is a matching synonym, the target word acquisition unit 14 acquires a representative word corresponding to the synonym as a word to be classified. The utterances between the customer and the operator are spoken, especially in the case of voice conversation. Therefore, it is expected that various expressions and words will be used even for words with the same meaning. Therefore, the target word acquisition unit 14 uses the synonym storage unit 34 to deal with the fluctuations in expression. The classification acquisition unit 15 acquires a classification from the candidate word storage unit 32 that corresponds to the classification target word acquired by the target word acquisition unit 14 .

[0018] The category determination unit 16 determines a category (one category) using the category target word acquired by the subject word acquisition unit 14 and the related words in the related word storage unit 33 related to the category in the candidate word storage unit 32 corresponding to the category target word acquired by the category acquisition unit 15. More specifically, the category determination unit 16 may determine a category using, for example, a word having a dependency relationship with the category target word acquired by the subject word acquisition unit 14. The category determination unit 16 may also determine a category using, for example, a neighboring word located in a neighboring position from the category target word acquired by the subject word acquisition unit 14. Here, the neighboring word is, for example, a word that is acquired within a range that is defined in advance as being within a certain number of words from the category target word. Furthermore, the category determination unit 16 may determine one category by selecting a related word in the related word storage unit 33 according to the frequency of appearance of a word included in the text. Furthermore, the category determination unit 16 may perform processing by the category determination unit 16 when, for example, there are multiple categories acquired by the category acquisition unit 15.

[0019] For example, when the classification acquisition unit 15 acquires multiple classifications, the classification determination unit 16 compares nearby words, which are words located in the vicinity of the classification target word in the text, with related words in the related word storage unit 33 that correspond to one of the multiple classifications, acquires one related word, and determines the classification in the related word storage unit 33 that corresponds to the one related word to be the one classification. Furthermore, when there is a related word in the related word storage unit 33 that matches the neighboring word and corresponds to any of the multiple classifications, the classification determination unit 16 acquires the related word as one related word and determines the classification in the related word storage unit 33 that corresponds to the one related word as the one classification. On the other hand, when there is no related word in the related word storage unit 33 that matches the neighboring word and corresponds to any of the multiple classifications, the classification determination unit 16 checks whether there is a matching synonym by comparing the neighboring word with the synonym in the synonym storage unit 34. Then, when there is a matching synonym, if the representative word corresponding to the synonym is a related word that corresponds to any of the multiple classifications, the classification determination unit 16 acquires the representative word as one related word and determines the classification in the related word storage unit 33 that corresponds to the one related word as the one classification.

[0020] The associating unit 17 associates the category acquired by the category determining unit 16 with the category target word. Then, the associating unit 17 stores the category information of the associated category and the category target word in the association storage unit .

[0021] The item updating unit 18 infers the customer's requirements by referring to the item storage unit 35 based on the categories and category target words associated by the associating unit 17. Here, the item updating unit 18 may make the inference based on, for example, the fulfillment rate of the category items in the item storage unit 35 by the categories associated by the associating unit 17. Then, the item updating unit 18 updates the value of the category item in the item storage unit 35 corresponding to the inferred requirement to the category target word corresponding to the category associated by the associating unit 17. The information output unit 19 outputs classification information including the classifications and classification target words associated by the associating unit 17 to the operator terminal 5. In addition, the information output unit 19 outputs the values ​​of the classification items in the item storage unit 35 updated by the item update unit 18 to the operator terminal 5.

[0022] The storage unit 30 is a storage area such as a hard disk, a semiconductor memory device, etc. for storing programs, data, etc. required for the control unit 10 to execute various processes. The storage unit 30 includes a program storage unit 31, a candidate word storage unit 32, a related word storage unit 33, a synonym storage unit , an item storage unit 35, a conversation storage unit 37, and an association storage unit . The program storage unit 31 is a storage area for storing various programs. The program storage unit 31 stores programs for carrying out various functions executed by the control unit 10 of the information acquisition server 1.

[0023] The candidate word storage unit 32 is a storage area that stores candidate words and categories in association with each other. As shown in FIG. 2(A), the candidate word storage unit 32 stores one or more candidate words in association with one or more classifications. Here, a candidate word is a word that succinctly indicates the content of a statement and is a candidate for determining a classification. A classification is a category that indicates the content of a candidate word. For example, the classifications corresponding to candidate words such as "hospitalization insurance" and "surgery insurance" are "subject to claim" and "subject to cancellation". For example, the classification corresponding to candidate words such as "insurance claim" and "document confirmation" is one category, "procedure content". The candidate word storage unit 32 stores candidate words in association with classifications in advance.

[0024] The related word storage unit 33 is a storage area that stores a classification and a related word in association with each other. As shown in Fig. 2B, the related word storage unit 33 stores one or more related words in association with one classification. Here, a related word is a word related to a classification. The related word storage unit 33 stores the classification and the related words in advance in association with each other.

[0025] The synonym storage unit 34 is a storage area that stores a reference word and a synonym in association with each other. As shown in FIG. 3(A), the synonym storage unit 34 stores one synonym in association with one representative word. Here, the representative word refers to a word used as a candidate word or a related word. Note that the synonym storage unit 34 may store a plurality of synonyms in association with one representative word. The synonym storage unit 34 stores the representative word and the synonym in advance in association with each other. Also, the synonym storage unit 34 may be capable of updating by a predetermined registration operation when a new synonym is generated.

[0026] The item storage unit 35 is a storage area that stores requirements and classification items in association with each other. As shown in FIG. 3B, the item storage unit 35 stores requirements in association with a plurality of classification items. Here, requirements are categories that indicate the contents of customer inquiries. Classification items are items that need to be elicited from customers and correspond to classifications. The item storage unit 35 stores requirements in association with classification items in advance.

[0027] The conversation storage unit 37 is a storage area that stores the ID of the speaker and the text of the speech in association with each other. When the control unit 10 receives speech data and identifies the speaker, for example, the ID of the speaker and the text of the speech are registered in the conversation storage unit 37 at any time. In addition, the conversation storage unit 37 stores the ID of the speaker and the text of the speech in association with each other, with the conversation between the customer and the operator from start to finish as one unit.

[0028] The association storage unit 38 is a storage area that stores a classification and an extracted value in association with each other. Here, the extracted value is a classification target word obtained from the text and corresponding to the classification, or a representative word of words obtained from the text and corresponding to the classification. The association storage unit 38 stores a classification and an extracted value in association with each other, with a period from the start to the end of a conversation between a customer and an operator being treated as one unit.

[0029] The above-mentioned storage units of the storage unit 30 are merely examples, and the data may be stored in different ways. For example, the candidate word storage unit 32 and the related word storage unit 33 may be integrated into one unit, and may further include the contents of the synonym storage unit 34. The storage unit 30 is not limited to this, and may have other storage areas, such as a storage area for storing information about the operator. The communication interface unit 39 is an interface for performing data communication and voice communication with other devices via the communication network N.

[0030] <Customer terminal 4> 1 is a terminal used by a customer. The customer uses the customer terminal 4 to make an inquiry to an operator. The customer terminal 4 is, for example, a portable terminal having a computer function, such as a smartphone, etc. The customer terminal 4 may also be a personal computer (PC), a tablet terminal, or a telephone. Although not shown, the customer terminal 4 includes a control unit, a storage unit, an input unit, a display unit, a communication interface unit, etc. The customer terminal 4 may include a touch panel display in which the input unit and the display unit are integrated.

[0031] <Operator Terminal 5> 1 is a terminal used by an operator. The operator uses the operator terminal 5 to answer inquiries from customers. The operator terminal 5 is, for example, a PC. The operator terminal 5 may also include a telephone and a microphone. Although not shown, the operator terminal 5 includes a control unit, a storage unit, an input unit, a display unit, a communication interface unit, and the like.

[0032] Here, a computer refers to an information processing device equipped with a control unit, a memory device, etc., and the information acquisition server 1, the customer terminal 4 and the operator terminal 5 are each information processing devices equipped with a control unit, a memory device, etc., and are included in the concept of a computer.

[0033] <Processing of information acquisition server 1> Next, the process of the information acquisition server 1 will be described. FIG. 4 is a flowchart showing the information acquisition process of the information acquisition server 1 according to this embodiment. 5 to 7 are flowcharts showing the text classification process of the information acquisition server 1 according to this embodiment.

[0034] For example, when the control unit 10 of the information acquisition server 1 detects that the customer terminal 4 has connected to the information acquisition server 1, the information acquisition process shown in FIG. 4 is started. In step S (hereinafter, "step S" will be simply referred to as "S") 11 in FIG. 4, the control unit 10 (utterance data receiving unit 11) receives utterance data. Here, the utterance data may be text or may be voice data. If the utterance data is voice data, the control unit 10 (utterance data receiving unit 11) performs voice recognition processing on the voice data to obtain text indicating the content of the utterance. Furthermore, the utterance data includes both utterances made by customers and utterances made by operators. Furthermore, utterance data refers to an utterance made by any speaker from the start to the end, and may be one sentence or multiple sentences.

[0035] In S12, the control unit 10 (speaker associating unit 12) assigns an ID for identifying the speaker of the received utterance data to the text. In S13, the control unit 10 (speaker associating unit 12) stores the ID and the text in the conversation storage unit 37 in association with each other. In S14, the control unit 10 performs a text classification process. The text classification process is a process for acquiring the classification of words contained in the text. The information acquisition server 1 can organize the comment contents by acquiring the classification of words.

[0036] The text classification process will now be described with reference to FIGS. 5, the control unit 10 (phrase separator 13) separates the text into phrases. Then, the control unit 10 (phrase separator 13) acquires words included in the phrases. Here, when a phrase contains multiple words, the control unit 10 may acquire multiple words. In S22, the control unit 10 (target word acquisition unit 14) compares the acquired words with the candidate words in the candidate word storage unit 32. Here, the comparison with the candidate words may be performed for all the acquired words at once, or may be performed in order starting from the first word of the text. In S23, the control unit 10 (target word acquisition unit 14) judges whether or not matching with a candidate word has been achieved. If matching with a candidate word has been achieved, that is, if there is a candidate word in the candidate word storage unit 32 that matches the acquired word (S23: YES), the control unit 10 shifts the process to S24. On the other hand, if matching with a candidate word has not been achieved (S23: NO), the control unit 10 shifts the process to S25.

[0037] In S24, the control unit 10 (target word acquisition unit 14) acquires the words that have been matched with the candidate words as target words for classification. After that, the control unit 10 shifts the process to S31 in FIG. In S25, the control unit 10 (target word acquisition unit 14) compares the word acquired in the process of S21 with synonyms in the synonym storage unit . In S26, the control unit 10 (target word acquisition unit 14) judges whether matching with synonyms has been successful. If matching with synonyms has been successful, that is, if there is a synonym in the synonym storage unit 34 that matches the acquired word (S26: YES), the control unit 10 shifts the process to S27. On the other hand, if matching with synonyms has not been successful (S26: NO), the control unit 10 shifts the process to S15 in FIG. 4 on the condition that all words included in the phrase have been processed. Note that, if there is an unprocessed word included in the phrase, the control unit 10 shifts the process to S22 and processes the unprocessed words in order. In S27, the control unit 10 (target word acquisition unit 14) acquires, as a classification target word, a representative word corresponding to the synonym in the synonym storage unit 34. After that, the control unit 10 shifts the process to S31 in FIG.

[0038] In S31 of Fig. 6, the control unit 10 (category acquisition unit 15) extracts a category of a candidate word corresponding to the acquired categorization target word by referring to the candidate word storage unit 32, and judges whether or not the extracted category is one. If there is one category (S31: YES), the control unit 10 shifts the process to S32. On the other hand, if there is more than one category (S31: NO), the control unit 10 shifts the process to S33. The case where there is more than one category refers to the case where there are multiple categories. In S32, the control unit 10 (category determination unit 16, associating unit 17) associates the extracted category with the classification target word, and stores the association in the association storage unit 38. Thereafter, the control unit 10 shifts the process to S15 in Fig. 4 on the condition that all words included in the phrase have been processed.

[0039] In S33, the control unit 10 (categorization determination unit 16) compares neighboring words, which are words that are in the vicinity of the word to be categorized, with related words in the related word storage unit 33. In S34, the control unit 10 (category determination unit 16) judges whether or not matching with related words has been achieved. If matching with related words has been achieved, that is, if there is a related word in the related word storage unit 33 that matches the neighboring word (S34: YES), the control unit 10 shifts the process to S35. On the other hand, if matching with related words has not been achieved (S34: NO), the control unit 10 shifts the process to S37 in FIG. In S35, the control unit 10 (category determination unit 16) acquires the related word that has been matched as one related word. In S36, the control unit 10 (category determination unit 16, associating unit 17) associates the category corresponding to one of the acquired related words with the category target word, and stores the association in the association storage unit 38. Thereafter, the control unit 10 shifts the process to S15 in Fig. 4 on the condition that all words included in the phrase have been processed.

[0040] In S37 of FIG. 7, the control unit 10 (category determination unit 16) compares neighboring words that are words that are in the vicinity of the word to be classified with synonyms in the synonym storage unit . In S38, the control unit 10 (category determination unit 16) judges whether matching with synonyms has been successful. If matching with synonyms has been successful, that is, if there is a synonym in the synonym storage unit 34 that matches the neighboring word (S38: YES), the control unit 10 shifts the process to S39. On the other hand, if matching with synonyms has not been successful (S38: NO), the control unit 10 shifts the process to S15 in FIG. 4 on the condition that all words included in the phrase have been processed. In S39, the control unit 10 (category determination unit 16) acquires, as one related word, a representative word from the synonym storage unit 34 that corresponds to the matched synonym. In S40, the control unit 10 (category determination unit 16, associating unit 17) associates the category corresponding to one of the acquired related words with the category target word, and stores the association in the association storage unit 38. Thereafter, the control unit 10 shifts the process to S15 in Fig. 4 on the condition that all words included in the phrase have been processed.

[0041] 4, the control unit 10 judges whether or not the classification has been acquired. If the classification has been acquired (S15: YES), the control unit 10 shifts the process to S16. On the other hand, if the classification has not been acquired (S15: NO), the control unit 10 shifts the process to S18. In S16, the control unit 10 (item update unit 18) performs an item update process. Specifically, the item update unit 18 infers customer requirements by referring to the item storage unit 35 based on the categories and category target words stored in the association storage unit 38. Then, the item update unit 18 updates the value of the category item in the item storage unit 35 corresponding to the inferred requirement to the category target word associated by the association unit 17.

[0042] In S17, the control unit 10 (information output unit 19) performs an output process for the operator terminal 5. Specifically, the information output unit 19 outputs classification information including the classification and the classification target word stored in the association storage unit 38 to the operator terminal 5. In addition, the information output unit 19 outputs the value of the classification item in the item storage unit 35 to the operator terminal 5. In S18, the control unit 10 judges whether the comment has ended. For example, when the control unit 10 of the information acquisition server 1 detects that the customer terminal 4 has disconnected from the information acquisition server 1, the control unit 10 judges that the comment has ended. If the comment has ended (S18: YES), the control unit 10 ends this process. On the other hand, if the comment has not ended (S18: NO), the control unit 10 moves the process to S11.

[0043] <Example> Next, a specific example using the above process will be described. 8 and 9 are diagrams showing a processing example of the information acquisition server 1 according to the present embodiment. FIG. 10 is a diagram illustrating an example of the storage unit of the information acquisition server 1 according to the present embodiment. FIG. 11 is a diagram showing an example of a display screen on the operator terminal 5 according to this embodiment.

[0044] <Example 1> FIG. 8(A) shows an example of the text 41 of the utterance data received from the customer terminal 4. The control unit 10 divides the text 41 into phrases and acquires the words 42 (S21 in FIG. 5). Then, the control unit 10 compares the words 42 with the candidate words in the candidate word storage unit 32 (S22 in FIG. 5). Fig. 8(B) shows the result of the comparison. As shown in Fig. 8(B), the candidate word 51 matches the word 42, so the control unit 10 acquires the word 42 as a word to be classified (S24 in Fig. 5). Next, since there are two classifications of candidate words corresponding to the classification target word, namely, "subject to billing" and "subject to cancellation," as shown in FIG. 8(B) (S31 in FIG. 6 is NO), the control unit 10 obtains nearby words 43 located in the vicinity of word 42, as shown in FIG. 8(C), and compares nearby words 43 with related words in the related word storage unit 33 (S33 in FIG. 6).

[0045] Fig. 8(D) shows the result of the comparison. As shown in Fig. 8(D), there is no related word that matches the neighboring word 43 (NO in S34 in Fig. 6). Therefore, the control unit 10 next compares the neighboring word 43 with the synonyms in the synonym storage unit 34 (S37 in Fig. 7). Fig. 8(E) shows the result of the comparison. As shown in Fig. 8(E), the synonym 52 matches the neighboring word 43, so the control unit 10 acquires the representative word 53 corresponding to the synonym 52 as one related word (S39 in Fig. 7). Then, as shown in FIG. 8(F), the control unit 10 acquires a classification 55 corresponding to one of the related words 54 in the related word storage unit 33, and as shown in FIG. 8(G), associates the classification 55 with the word 42 and stores it in the association storage unit 38 (S40 in FIG. 7). In this way, the information acquisition server 1 can organize the comment contents of the text 41 according to the classification and the extracted value stored in the association storage unit .

[0046] <Example 2> FIG. 9(A) shows an example of text 61 of utterance data received from the customer terminal 4 after the text 41. The control unit 10 divides the text 61 into phrases and acquires the word 62 (S21 in FIG. 5). The control unit 10 acquired the word "claimant" before the word 62 and performed text classification processing (S14 in FIG. 4), but was unable to acquire a classification. The control unit 10 then compares the word 62 with the candidate words in the candidate word storage unit 32 (S22 in FIG. 5). Although not shown, since there is no candidate word that matches the word 62 (NO in S23 in FIG. 5), the control unit 10 then compares the word 62 with the synonyms in the synonym storage unit 34 (S25 in FIG. 5).

[0047] Fig. 9(B) shows the result of the comparison. As shown in Fig. 9(B), since the synonym 71 matches the word 62 (YES in S26 of Fig. 5), the control unit 10 acquires the representative word corresponding to the synonym 71 as the classification target word 72 (S27 of Fig. 5). Next, since there is only one classification of the candidate word corresponding to the classification target word 72, as shown in FIG. 9(C) (S31 in FIG. 6 is YES), the control unit 10 acquires the classification 73 corresponding to the candidate word in the candidate word storage unit 32, and associates the classification 73 with the classification target word 72 as shown in FIG. 9(D), and stores it in the association storage unit 38 (S32 in FIG. 6).

[0048] <Example 3> Specific example 3 shown in FIG. 10 illustrates an example of a conversation memory unit 37 and an association memory unit 38 stored in the memory unit 30 based on a conversation between a customer and an operator, which includes specific example 1 described based on FIG. 8 and specific example 2 described based on FIG. 9.

[0049] <Example 4> A fourth specific example shown in FIG. 11 shows an example of an operator screen 80 outputted to the operator terminal 5 during a conversation between the customer and the operator in the third specific example shown in FIG. The operator screen 80 includes a conversation area 81 and an assistance area 91 . The conversation area 81 is an area related to conversation with a customer. The conversation area 81 includes a statement output area 82 and a statement input area 83. The comment output area 82 outputs the contents of comments between the operator and the customer in chronological order and in a format that makes it possible to identify the person who made the comment. The comment input area 83 is for the operator to input a comment to the customer, and includes an input field 84 and a send button 85 .

[0050] The support area 91 is an area for outputting information that organizes the contents of comments in order to support an operator. The support area 91 includes a classification information area 92 and a candidate area 93. The classification information area 92 outputs the classification information stored in the association storage unit 38 . The candidate area 93 outputs classification items for which values ​​have not been input, among the classification items in the item storage unit 35. In Fig. 11, the classification items are output together with the priority order, but the priority order may be output, for example, in the order in which the classification items are arranged in the item storage unit 35. Also, the priority order does not have to be output. By referring to the support area 91 while talking to a customer, the operator can understand the contents of the inquiry to the customer while viewing the output of the organized information obtained from the customer, thereby reducing the number of missed questions. In addition, the operator can check the items that need to be asked of the customer, and can talk to the customer efficiently. 11 is merely an example. Other ways of displaying the operator screen 80 are also possible, and for example, a screen showing only the support area 91 may be used.

[0051] As described above, the information acquisition server 1 of this embodiment has the following advantages. (1) A classification for organizing the contents of the conversation is acquired by referring to each storage unit of the storage unit 30 from the contents of the conversation. For example, a classification target word is acquired from the candidate word storage unit 32 that stores a correspondence between a word of the text, a candidate word, and one or more classifications. Next, a classification corresponding to the classification target word is determined using related words in the related word storage unit 33 that stores related words related to the classification. Then, the determined classification is associated with the classification target word. Thus, it is possible to acquire information by improving the accuracy of extracting a classification for the word of the text. In addition, it is possible to check whether the contents of the conversation are related to the classification target word and the classification by using the related words.

[0052] (2) When a word in the text falls into multiple categories, one related word is obtained by comparing the related word memory unit 33, which stores related words related to the category, with the neighboring words, and the category is determined from the one related word obtained. Therefore, a category related to the word in the text can be obtained from the neighboring words, thereby improving the accuracy of the classification.

[0053] (3) Since a set of related words or words to be classified is acquired using the synonym storage unit 34, it is possible to deal with the variations in various expressions used in conversation. (4) The text is divided into phrases, and classification target words are obtained based on the words contained in the phrases, so that words for information extraction can be obtained easily and appropriately. (5) If the utterance data is voice data, it is converted to text through voice recognition processing. Therefore, even if the utterance data is voice data, it is possible to extract information corresponding to the classification. In addition, since it also works when the utterance data is text, it is also possible to handle inquiries from customers via web chat, for example.

[0054] (6) Since the classification information including the association between the classification and the words to be classified is output to the operator terminal 5, the operator can conveniently understand the customer's inquiry at a glance by using the classification information that organizes the contents of the customer's inquiry. (7) Classification items corresponding to the customer's requirements are stored in advance in the item storage unit 35, and when a classification is associated with a word to be classified, the classification items in the item storage unit 35 are updated and output to the operator terminal 5, thereby supporting the operator so that they do not miss any details that need to be confirmed with the customer. In addition, the customer's requirements are inferred from the update status of the classification items in the item storage unit 35, and the classification items in the item storage unit 35 corresponding to the inferred requirements are output to the operator terminal 5, so that the operator can be supported including the customer's requirements.

[0055] Although the embodiments of the present invention have been described above, the present invention is not limited to the above-mentioned embodiments. Furthermore, the effects described in the embodiments are merely a list of the most preferable effects resulting from the present invention, and the effects of the present invention are not limited to those described in the embodiments. The above-mentioned embodiments and the modified forms described below can be used in appropriate combination, but detailed description will be omitted.

[0056] (Variations) (1) In the present embodiment, when the candidate word storage unit has multiple classifications corresponding to the words in the text (classification target words), the related words in the related word storage unit are used, but the present invention is not limited to this. Even if the candidate word storage unit has only one classification corresponding to the classification target words, the related words in the related word storage unit may be used. In this way, the accuracy of text classification can be improved.

[0057] (2) In the present embodiment, a comparison is made between words adjacent to a word in the text (a word to be classified) and related words, but this is not limiting. A comparison may be made between a word having a dependency relationship with the word to be classified and related words. In this way, the accuracy of the text classification can be improved by using words that are highly related to the word to be classified. Alternatively, the classification may be determined by selecting related words according to the frequency of occurrence of words contained in all the texts of the conversation. In this way, the classification can be determined using related words that are common throughout the entire conversation, so that a consistent classification can be obtained throughout the entire conversation.

[0058] (3) In the present embodiment, an example has been described in which utterance data is processed in real time in response to reception thereof, and classification information is output to the operator terminal 5 to support the operator, but the present invention is not limited to this. The utterance data stored in the conversation storage unit may be processed collectively. In this way, the service provider can analyze the contents of inquiries from multiple customers, not just one conversation. The analysis results can be used, for example, to improve or enrich FAQs (Frequently Asked Questions) or to improve the contents for sales promotion.

[0059] (4) In the present embodiment, the words of the text (the words to be classified) are associated with the categories, but this is not limiting. The text itself may be associated with the category. (5) In the present embodiment, an example has been described in which a customer inquires about insurance is handled, but the present invention is not limited to this example and can also be used for other call center operations (inquiries, solicitations, etc.). [Explanation of symbols]

[0060] 1 Information Acquisition Server 10 Control section 11 Speech data reception unit 12 Speaker Association Unit 13 Phrase division section 14 Target Word Acquisition Section 15 Classification acquisition section 16 Classification determination section 17 Association section 18 Item update section 19 Information output section 30 Storage section 31 Program memory section 32 Candidate word storage unit 33 Related Words Memory Section 34 Synonym storage 35 item storage section 37 Conversation Memory Section 38 Association memory unit 39 Communication Interface Section 80 Operator Screen 81 Conversation Area 91 Support area 92 Classification information area 93 Candidate area 100 Information Acquisition System

Claims

1. An information acquisition server that acquires information from a conversation between an operator and a customer, a candidate word storage unit that stores candidate words in association with one or more categories; a related word storage unit that stores related words related to the classification; a target word acquisition means for acquiring a classification target word from the text by comparing words included in the text indicating the content of the conversation with the candidate words in the candidate word storage unit; a classification acquisition means for acquiring the classification of the candidate word storage unit corresponding to the classification target word acquired by the target word acquisition means; a classification determination means for determining one of the classifications using the classification target words of the text and the related words of the related word storage unit related to the classification acquired by the classification acquisition means; an association means for associating the one of the classifications determined by the classification determination means with the classification target word; An information acquisition server comprising:

2. 2. The information acquisition server according to claim 1, The classification determination means determines the one classification by using a word having a dependency relationship with the classification target word of the text.

3. 2. The information acquisition server according to claim 1, The classification determination means determines the one classification by using neighboring words located in a vicinity of the classification target word in the text.

4. In the information acquisition server according to any one of claims 1 to 3, The classification determination means selects the related words according to the frequency of appearance of words included in all of the text of the conversation, and determines the one classification.

5. In the information acquisition server according to any one of claims 1 to 4, an information acquisition server that performs processing by the category determination means when there are a plurality of categories acquired by the category acquisition means;

6. 2. The information acquisition server according to claim 1, an information acquisition server, wherein when there are multiple classifications acquired by the classification acquisition means, the classification determination means refers to the related word storage unit, compares nearby words located in the vicinity of the word to be classified in the text with the related words in the related word storage unit corresponding to one of the multiple classifications, acquires one of the related words, and determines the classification in the related word storage unit corresponding to the one related word to be the one classification.

7. 7. The information acquisition server according to claim 6, a synonym storage unit that stores a representative word and a synonym of the representative word in association with each other; An information acquisition server, wherein the classification determination means, when unable to obtain the one related word, refers to the synonym memory unit, obtains the one related word based on the synonym that matches the nearby word, and determines the classification of the related word memory unit corresponding to the one related word to be the one classification.

8. In the information acquisition server according to any one of claims 1 to 7, a synonym storage unit that stores a representative word and a synonym of the representative word in association with each other; The target word acquisition means, when unable to acquire the target word for classification by comparing the words contained in the text with the candidate words in the candidate word storage unit, refers to the synonym storage unit and acquires the target word for classification based on the synonym that matches the word contained in the text.

9. In the information acquisition server according to any one of claims 1 to 8, A phrase dividing means for dividing the text into phrases, The target word acquisition means acquires the classification target words based on the words included in the phrases.

10. In the information acquisition server according to any one of claims 1 to 9, A speech data receiving means is provided for receiving speech data from a speaker, The target word acquisition means acquires the classification target word from the text corresponding to the utterance data in response to the utterance data acceptance means accepting the utterance data.

11. 11. The information acquisition server according to claim 10, The utterance data receiving means receives the text indicating the content of the utterance.

12. 11. The information acquisition server according to claim 10, The utterance data receiving means receives voice data including the utterance content, The information acquisition server includes a text acquisition means for performing a voice recognition process on the voice data received by the utterance data reception means and acquiring the text.

13. In the information acquisition server according to any one of claims 10 to 12, The information acquisition server includes a speaker associating means for associating the text with speaker identification information for identifying the speaker.

14. A program for causing a computer to function as the information acquisition server according to any one of claims 1 to 13.

15. An information acquisition server according to any one of claims 1 to 13; A terminal used by the operator; An information acquisition system comprising: The information acquisition server comprises an output unit that outputs classification information including the classification and the classification target word associated by the association unit to the terminal.

16. The information acquisition system according to claim 15, The information acquisition server includes: an item storage unit that stores in advance classification items corresponding to the customer's requirements; an item update output means for updating the classification items in the item storage unit corresponding to the classifications associated by the associating unit, and outputting the classification items to the terminal; An information acquisition system comprising:

17. 17. The information acquisition system according to claim 16, the information acquisition server includes a requirement inference means for inferring requirements of the customer by referring to the item storage unit based on the category and the classification target word associated by the associating means; The item update output means outputs the classification items corresponding to the requirements inferred by the requirements inferring means to the terminal.

Images