Business negotiation content analysis system

Abstract

To enable easy identification of items related to the content of conversations in business negotiations and items related to the characteristics of products or services to be transacted, in a system 1 that analyzes the content of business negotiations S between a seller P and a customer C in product or service purchase transactions, based on audio recording data of the negotiations.SOLUTION: A business negotiation content analysis system converts audio recording data of business negotiations into text data; performs natural language analysis to generate keyword-searchable dialogue text data (5, 6); sets conversation content keywords related to the content of conversations and product content keywords related to the products or services (10); searches the dialogue text data for the conversation content keywords and the product content keywords (8); and displays the conversation content keywords and the product content keywords found in the dialogue text data separately (9).SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to a system for analyzing the contents of business negotiations recorded in audio, and more specifically to a system for extracting keywords related to the success or failure of a transaction from the recording of business negotiations, allowing users to easily reflect on the contents of the business negotiations. [Background technology]

[0002] With the spread of artificial intelligence (AI) technology using machine learning, it has become possible to generate text data of utterances in audio data from audio data recorded with human speech or conversation using speech recognition technology, and natural language analysis including morphological analysis has become possible for the text data, so that technologies for analyzing utterances in audio data in various ways have been proposed. For example, Patent Document 1 proposes a technology for checking compliance when conducting sales activities for financial products, which involves searching for dialogues that require monitoring from the content of explanations to customers in order to check that sales representatives explain matters that should be explained during business negotiations with customers and do not state matters that should not be mentioned, and a technology for making the monitoring work more efficient. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent Publication No. 2019-28910 Summary of the Invention [Problem to be solved by the invention]

[0004] The technology described above for converting recorded speech into text data and then analyzing the speech using natural language processing can also be used to analyze the content of negotiations between sellers and customers in the sale and purchase of goods and services, thereby determining the outcome of the negotiations, i.e., the reasons for the success or failure of the transaction. In this regard, the success or failure of a transaction is primarily determined by the content of the conversation between the seller and the customer (e.g., questions and responses, topics, and choice of speech expressions) and the content of the goods or services involved in the transaction (e.g., the features, specifications, and price of the goods or services). Therefore, it is desirable for the results of an analysis of the results of a sales negotiation to easily identify both the content of the conversation and the characteristics of the goods or services involved.

[0005] In view of the above circumstances, the main objective of the present invention is to provide a system that analyzes the content of negotiations between a seller and a customer in a purchase and sale transaction of goods and services from recorded data of the negotiations and analyzes the results of the negotiations, in which it is possible to easily grasp matters related to the content of the conversations in the negotiations and matters related to the characteristics of the goods and services themselves that are the subject of the transaction. [Means for solving the problem]

[0006] According to the present invention, the above problem is solved by a system for analyzing the content of negotiations between a seller and a customer in a sales transaction of a product or service, a dialogue text data generation means configured to generate text data from the recording data of the business negotiation and to perform natural language analysis on the text data to generate dialogue text data that can be searched for by keywords; a keyword setting means configured to set conversation content keywords, which are keywords related to the content of the conversation, and product content keywords, which are keywords related to the products and services; a keyword search means configured to search the dialogue text data for the conversation content keywords and the product content keywords, respectively; a keyword display means configured to separately display the conversation content keywords and the product content keywords found in the dialogue text data by the keyword search means; This is achieved by a system including:

[0007] In the above configuration, "products and services" can be any product (such as automobiles, machinery, or real estate) or any service. The "seller" can be a salesperson or other person directly involved in negotiations with customers. "Recorded data of business negotiations" refers to data recorded during business negotiations. "Text data" refers to data obtained by converting the audio of recorded data into string data that can be processed by a computer using speech recognition technology. "Dialogue text data" refers to text data that has been further processed using natural language analysis technology, including morphological analysis, to identify parts of speech and words, making it keyword searchable. "Keywords related to the content of the conversation" (conversation content keywords) are keywords that enable understanding of the questions and responses between the seller and customer, the topics and choice of utterances, etc., while "keywords related to the product and service" (product content keywords) refer to keywords related to the features, specifications, price, etc. of the product or service being traded. These keywords are generally set arbitrarily by system administrators and users through keyword setting means. Keywords may be words, phrases, clauses, or short conversations (hereinafter referred to as "words, etc."). After multiple pieces of dialogue text data on business negotiations for which the success or failure of a transaction has been determined have been accumulated, the keyword setting means may use any machine learning algorithm or the like to select words that frequently appear when a transaction is concluded or not as conversation content keywords. The "keyword search means" searches the dialogue text data for the set keywords in a typical manner, and when a word, etc. matching the keyword is found, the portion is stored. The keyword display means may be configured to display, in any manner, the conversation content keywords and the product content keywords found in the dialogue text data in a manner that allows them to be distinguished from each other. Specifically, the keyword display means may be configured to highlight portions matching the keywords when displaying the dialogue text data on any display screen or medium, or to display the portions matching the keywords in the form of a word cloud in which the portions matching the keywords are emphasized. Each of the means of the system of the present invention described above is realized by the operation of a computer device according to a program.

[0008] In the system of the present invention, conversation content keywords and product content keywords are searched for in the dialogue text data generated from the recorded data of the business negotiation, and when words matching these keywords are found, the conversation content keywords and the product content keywords are displayed separately. This allows sellers such as salespeople to easily understand and reflect on the factors that determine the success or failure of a transaction by referring to the conversation content keywords in the dialogue text data, and also allows product and service developers to easily understand the factors that determine the success or failure of a transaction related to their product or service by referring to the product content keywords in the dialogue text data.

[0009] The system of the present invention may further include a product / service search means configured to search for products / services that match the customer's attribute information by referencing customer relationship management data (CRM data), and a product / service-related utterance extraction means configured to extract, from the dialogue text data, utterances proposing products / services that match the customer's attribute information or utterances proposing products / services different from the customer's attribute information. The search for products / services that match the customer's attribute information may be achieved using any generative AI technology (e.g., chatGPT (registered trademark)). This allows a seller, system administrator, or user to review whether products / services corresponding to the customer's attribute information recorded in the customer relationship management data were proposed during the negotiation, or whether products / services different from the products / services that match the customer's attribute information were proposed.

[0010] In the above system, the dialogue text data generating means may be configured to generate dialogue text data in a state in which the speaker can be identified, and the keyword display means may be configured to display, for each speaker, conversation content keywords and product content keywords found in the dialogue text data. This makes it easy for the seller, system administrator, or user to consider the appropriateness of the seller's utterances in business negotiations.

[0011] Furthermore, in the above system, the keyword setting means may be configured to set positive and negative keywords related to products and services as product content keywords, and the keyword display means may be configured to display the product content keywords found in the dialogue text data in a manner that allows the positive and negative keywords to be distinguished. This is advantageous in that it makes it easier for product and service developers to understand the factors that determine the success or failure of a transaction related to a product or service, and makes it easier for them to understand the advantages and disadvantages of the product, etc. that are focused on during business negotiations.

[0012] The keyword display means of the above system may be configured to display the conversation content keywords and product content keywords found in the dialogue text data in a manner that allows the frequency of their appearance to be ascertained. This allows the seller, system administrator, or user to obtain information such as which words, etc., affected the success or failure of a transaction during a business negotiation, and which words, etc., appeared more frequently or less frequently when a transaction was concluded or not. [Effects of the Invention]

[0013] Thus, the system of the present invention makes it easy to understand what was said in the conversation between the seller and the customer during a business negotiation, and what was said about the product or service itself. This allows the seller to easily reflect on which words were used in the negotiation that led to the successful or unsuccessful transaction, and allows the developer of the product or service to easily reflect on what aspects of the product or service were evaluated or not evaluated in the negotiation, and how this affected the success or failure of the transaction. The system of the present invention may be applied to reviewing various business negotiations.

[0014] Other objects and advantages of the present invention will become apparent from the following description of preferred embodiments of the invention. [Brief explanation of the drawings]

[0015] [Figure 1] FIG. 1 is a block diagram showing the configuration of the system according to this embodiment. [Figure 2] Fig. 2(A) is a flowchart showing the process up to the generation of dialogue text data in the system of this embodiment, and Fig. 2(B) is a flowchart showing the process of analyzing dialogue text data in the system of this embodiment. [Figure 3] 3A, 3B, and 3C are diagrams showing typical examples of displaying the results of analysis of business negotiations in the system of this embodiment. [Explanation of symbols]

[0016] 1...sales negotiation result analysis system, 2...microphone, 3...recording unit, 4...recorded data memory, 5...speech recognition and speaker identification unit, 6...natural language analysis unit, 7...dialogue text data memory, 8...keyword search and analysis unit, 9...user terminal, 9a...display, 10...keyword setting unit, 11...CRM database, 12...CRM data analysis unit, S...sales negotiation site, C...customer, P...seller (sales representative) BEST MODE FOR CARRYING OUT THE INVENTION

[0017] The present invention will now be described in detail with reference to some preferred embodiments thereof with reference to the accompanying drawings, in which like reference numerals indicate like parts.

[0018] System configuration As shown in Figure 1, the business negotiation result analysis system 1 of this embodiment is configured to search recorded conversations during a business negotiation S between a seller P and a customer C in a product or service purchase transaction for keywords related to the content of the conversation (conversation content keywords) and keywords related to the product or service (product content keywords). When words, phrases, clauses, or short conversations (words, etc.) that match the keywords are found, the system highlights the words, etc., and displays them in a manner that is easily understandable to the system user. This configuration makes it easy to examine and reflect on the factors that led to the success or failure of a business negotiation. For example, if a deal is not concluded due to the salesperson's negotiating skills, the salesperson can reflect on points for improvement. If the cause is a problem with the product or service design, functionality, or equipment, feedback to the product or service developer can be easily provided.

[0019] More specifically, in the configuration of the negotiation result analysis system 1 of this embodiment, first, the voice of a negotiation between seller P and customer C is collected by microphone 2, recorded by recording unit 3, and stored as recorded data in recorded data memory 4. When analyzing the content of the negotiation, the speech in the recorded data stored in recorded data memory 4 is converted into text (transcription) using any speech recognition technology by speech recognition and speaker identification unit 5, generating text data processable by a computer and identifying the speaker of each utterance. Next, natural language analysis, including morphological analysis, is performed on the text data by natural language analysis unit 6, identifying parts of speech, words, and the like in the text data, generating dialogue text data that can be searched for by keywords, and storing the resulting dialogue text data in dialogue text data memory 7.

[0020] The dialogue text data of the business negotiation generated as described above may be sent to the keyword search and analysis unit 8, where keyword search and other analyses may be performed on the dialogue text data for each of the conversation content keywords and product content keywords, as described above. The keywords used for the search may be input by the user through the terminal device 9, set by the keyword setting unit 10, and provided to the keyword search and analysis unit 8. When information about customer C is stored in the CRM database 11, the CRM data analysis unit 12 may refer to the information about customer C to estimate information about products and services that match the attributes of customer C, and this information may be sent to the keyword search and analysis unit 8 and used to analyze the dialogue text data. The results of the analysis by the keyword search and analysis unit 8 may be displayed on the display 9a of the user terminal device 9 in various ways, as described below.

[0021] As already mentioned, each part of the above-mentioned negotiation result analysis system 1 is realized by the operation of a computer device in accordance with a program.

[0022] System Operation In operation of the system of this embodiment, referring also to Figure 2(A), in the process SC up to the generation of dialogue text data for business negotiation S, as described above, the following steps are executed in order: generation of text data by transcribing recorded audio data into text (S1), identification of the speaker (S2), and natural language analysis (S3).

[0023] Then, as shown in Fig. 2(B), keyword search and analysis (SR) is performed on the generated dialogue text data of negotiation S. In this embodiment, as described above, conversation content keywords and product content keywords are set as keywords.

[0024] More specifically, product content keywords are keywords related to the features, specifications, price, etc., of the products and services involved in the transaction. As mentioned above, the success or failure of a business transaction is considered to depend on the above-mentioned product and service-related factors. Therefore, in analyzing business negotiations using the system of this embodiment, product and service-related factors from the dialogue during the negotiation can be extracted, making it easy to provide feedback to the product and service developer. In this case, keywords may be selected that represent the advantages and positive evaluations of the product and service, and the disadvantages and negative evaluations of the product and service, respectively, when the transaction is concluded and when it is not concluded. The results of this analysis are expected to be useful for product and service developers when examining how product and service content factors in business negotiations affected the success or failure of the transaction.

[0025] Conversation content keywords are keywords that enable understanding of the content of questions and responses between the seller and the customer, the topics of conversation, and the choice of speech expressions. As already mentioned, the success or failure of a business negotiation is considered to depend on the negotiation skills of the salesperson, i.e., the content of questions and responses, the topics of conversation, and the speech expressions. Therefore, in the analysis of business negotiations by the system of this embodiment, conversation content keywords may be set and keywords may be extracted from the dialogue text data, focusing on the content of such conversations, so that the situation of the business negotiation can be easily understood. In this case, more specifically, keywords that are frequently used when a transaction is concluded and words that are frequently used when a transaction is not concluded may be selected.

[0026] The above keywords may basically be input and set by the system administrator, designer, or user as appropriate through the terminal device 9. Furthermore, when the analysis results of the dialogue text data of business negotiations are accumulated, the system may select, from the accumulated dialogue text data, words and the like that are frequently used when a deal is concluded and when a deal is not concluded as keywords.

[0027] Once the keywords are set as described above (S4), keyword extraction and analysis (A) specialized for the content of the conversation using the conversation content keywords and keyword extraction and analysis (B) specialized for product and service-related matters using product and service content keywords are performed. In the conversation content-specific keyword extraction and analysis (A), first, words matching the conversation content keywords are extracted from the dialogue text data (S5). As mentioned above, when information about customer C is stored in the CRM database 11, the information about customer C is referenced to estimate products and services that match customer C's attributes, and it may be determined whether the seller has proposed the estimated products and services, or whether they have proposed products and services different from the estimated products and services (S6). The estimation of products and services that match customer C's attributes may be performed using any AI technology that automatically generates answers to any questions, or using a generative AI technology that automatically generates answers to questions about products and services that match customer C's attributes. For example, in a car sales negotiation, the dialogue text data may detect whether a car model was proposed based on the customer's family structure and life events, whether a minivan was proposed to a customer with a family of two, whether a light car was proposed to a customer with a family of five, etc. Meanwhile, in the keyword extraction and analysis (B) specialized for matters related to products and services, words etc. matching product content keywords etc. are extracted from the dialogue text data (S8).

[0028] In either case, the results of the keyword extraction and analysis are organized in various ways and displayed on the display 9a of the terminal device 9 (S7, S9). Specifically, as shown in FIG. 3(A), all or part of the dialogue text data may be displayed, and words matching the conversation content keywords and words matching the product content keywords may be highlighted or otherwise emphasized. Alternatively, as shown in FIG. 3(B), the frequency of appearance of positive or negative words in the dialogue text data may be counted for each of the words matching the conversation content keywords and the product content keywords, and each word may be displayed along with its rank. Alternatively, as shown in FIG. 3(C), the analysis results may be displayed in the form of a word cloud, in which words appearing frequently in the dialogue text data are displayed larger and words that appeared recently are positioned closer together, making it easier to understand the content of the business negotiation. Furthermore, although not shown, the determination result as to whether or not the seller proposed the estimated product or service, or whether or not the seller proposed a product or service different from the estimated product or service, may be displayed in any manner.

[0029] Thus, the system of this embodiment makes it possible to analyze separately the content of the conversation and the content of the product or service in the utterances during the business negotiation, allowing the seller, such as the sales representative who conducted the business negotiation, and the developer of the product or service to easily reflect on the content of the business negotiation. For example, the system of this embodiment can be advantageously used to review the reasons why a business negotiation did not result in a contract after the business negotiation is over to find out the reasons.

[0030] The above description has been made in relation to the embodiments of the present invention, but it will be apparent that many modifications and changes will be readily apparent to those skilled in the art, and the present invention is not limited to the above-described exemplary embodiments, but can be applied to various devices without departing from the concept of the present invention.

Claims

1. A system for analyzing the content of negotiations between a seller and a customer in a sales transaction of a product or service, a dialogue text data generation means configured to generate text data from the recording data of the business negotiation and to perform natural language analysis on the text data to generate dialogue text data that can be searched for by keywords; a keyword setting means configured to set conversation content keywords, which are keywords relating to the content of the conversation, and product content keywords, which are keywords relating to the products and services; a keyword search means configured to search the dialogue text data for the conversation content keywords and the product content keywords, respectively; a keyword display means configured to separately display the conversation content keywords and the product content keywords found in the dialogue text data by the keyword search means; A system including:

2. 10. The system of claim 1, further comprising: a product / service searching means configured to search for products / services that match the attribute information of the customer by referring to the customer relationship management data; a product / service related utterance extraction means configured to extract, from the dialogue text data, utterances proposing products / services that match the attribute information of the customer or utterances proposing products / services different from the products / services that match the attribute information of the customer; a speech display means configured to display the speech extracted by the product / service related speech extraction means; Including system.

3. 2. The system of claim 1, wherein the dialogue text data generation means is configured to generate dialogue text data in a state in which the speaker can be identified, and the keyword display means is configured to display the conversation content keywords and product content keywords found in the dialogue text data for each speaker.

4. 2. The system of claim 1, wherein the keyword setting means is configured to set positive keywords and negative keywords related to the product / service as the product / service content keywords, and the keyword display means is configured to display the product / service content keywords found in the dialogue text data in a manner that allows the positive keywords and the negative keywords to be distinguished.

5. 2. The system of claim 1, wherein the keyword display means is configured to display the conversation content keywords and product content keywords found in the dialogue text data in a manner that allows the frequency of their occurrence to be grasped.

Images