Response method and device, computer equipment and storage medium

Abstract

The invention relates to a response method and device, computer equipment and a storage medium. The method is applied to the financial field or other fields and is applied to an intelligent customer service system of a bank, and the method comprises the following steps: carrying out inquiry intention recognition on target inquiry information of a target object based on a general service question-answering system to obtain an intention recognition result, and determining a target professional field indicated by the intention recognition result; calling a professional field question-answering system corresponding to the target professional field, and determining initial response information matched with the target inquiry information; obtaining historical inquiry information of the target object in a historical time period; performing expression style analysis based on the historical inquiry information to obtain an expression style of the target object; and generating target response information conforming to the expression style based on the initial response information, and returning the target response information to the terminal of the target object. By adopting the method, the accuracy of customer service response can be improved.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to a response method, device, computer equipment and storage medium. Background Art

[0002] With the development of artificial intelligence, in order to save labor costs, many companies have launched intelligent customer service systems. The company's customers can directly communicate with the intelligent customer service in the intelligent customer service system to quickly understand relevant business knowledge.

[0003] In conventional technology, preset questions similar to the user's question are usually determined from a preset question-and-answer library, and then the answers corresponding to the preset questions are returned to the user's terminal.

[0004] However, the method of determining preset questions similar to the user's questions from a preset question and answer library and then returning the answers corresponding to the preset questions to the user's terminal often fails to meet the personalized needs of different users in actual business scenarios, resulting in inaccurate responses from the customer service system. Summary of the invention

[0005] Based on this, it is necessary to provide a response method, device, computer equipment, computer-readable storage medium and computer program product that can improve the accuracy of customer service responses in response to the above technical problems.

[0006] In the first aspect, the present application provides a response method. The method includes: based on the general business question-answering system, performing inquiry intent recognition on the target inquiry information of the target object, obtaining the intent recognition result, and determining the target professional field indicated by the intent recognition result; calling the professional field question-answering system corresponding to the target professional field, determining the initial response information matching the target inquiry information; obtaining the historical inquiry information of the target object within the historical time period; performing expression style analysis based on the historical inquiry information to obtain the expression style of the target object; generating target response information that conforms to the expression style based on the initial response information, and returning the target response information to the terminal of the target object.

[0007] In the second aspect, the present application also provides a response device. The device includes: an intention recognition module, which is used to perform inquiry intention recognition on the target inquiry information of the target object based on the general business question and answer system, obtain the intention recognition result, and determine the target professional field indicated by the intention recognition result; an initial information determination module, which is used to call the professional field question and answer system corresponding to the target professional field, and determine the initial response information matching the target inquiry information; a historical information acquisition module, which is used to obtain the historical inquiry information of the target object within the historical time period; a style analysis module, which is used to perform expression style analysis based on the historical inquiry information to obtain the expression style of the target object; an information generation module, which is used to generate target response information that conforms to the expression style based on the initial response information, and return the target response information to the terminal of the target object.

[0008] In some embodiments, the initial information determination module is also used to: call the professional field question and answer system corresponding to the target professional field to generate candidate response information; when there is related field information in the candidate response information, call the professional field question and answer system corresponding to the related field information to generate auxiliary information under the related field; generate initial response information based on the candidate response information and the auxiliary information.

[0009] In some embodiments, the response device also includes a style recommendation module, which is used to: obtain object attribute information of the target object with the authorization of the target object; determine the recommended information style based on the object attribute information; the information generation module is also used to: return the target response information and the recommended information style to the terminal of the target object to instruct the terminal of the target object to display the target response information according to the recommended information style.

[0010] In some embodiments, the intent recognition module is further used to: extract at least one keyword from the target query information; identify a domain keyword from the at least one keyword; and determine the target professional field based on the domain keyword.

[0011] In some embodiments, the initial information determination module is also used to: call the professional field question and answer system corresponding to the target professional field to determine the target corpus corresponding to the target professional field; for each corpus paragraph in the target corpus, calculate the information matching degree between the corpus paragraph and the target inquiry information; determine the corpus paragraph that matches the target inquiry information from the target corpus according to the information matching degree to obtain initial response information.

[0012] In some embodiments, the response device also includes a language identification module, which is used to: receive the initial inquiry information sent by the terminal of the target object; perform language type identification on the initial inquiry information to obtain the object language type corresponding to the initial inquiry information; convert the initial inquiry information into target inquiry information expressed in a preset language type; the information generation module is also used to: convert the target response information into target response information expressed in the object language type, and return the target response information expressed in the object language type to the terminal of the target object.

[0013] In a third aspect, the present application further provides a computer device, wherein the computer device comprises a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps in the above-mentioned response method when executing the computer program.

[0014] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, and the computer program implements the steps in the above-mentioned response method when executed by a processor.

[0015] In a fifth aspect, the present application further provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, the steps in the above-mentioned response method are implemented.

[0016] The above-mentioned answering method, apparatus, computer equipment, storage medium and computer program product obtain the intention recognition result by performing inquiry intent recognition on the target inquiry information of the target object based on the general business question and answer system, and determine the target professional field indicated by the intention recognition result, call the professional field question and answer system corresponding to the target professional field, determine the initial answer information that matches the target inquiry information, and on the basis of determining the more professional initial answer information, perform expression style analysis based on the historical inquiry information of the target object to obtain the expression style of the target object, generate target answer information that conforms to the expression style based on the initial answer information, and return the target answer information to the terminal of the target object, so that the target answer information is more in line with the personalized needs of the target object, thereby improving the accuracy of customer service responses. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] FIG1 is an application environment diagram of a response method in one embodiment;

[0018] FIG2 is a schematic diagram of a flow chart of a response method in one embodiment;

[0019] FIG3 is a schematic flow chart of a response method in another embodiment;

[0020] FIG4 is a block diagram of a response device in one embodiment;

[0021] FIG5 is a diagram showing the internal structure of a computer device in one embodiment;

[0022] FIG. 6 is a diagram showing the internal structure of a computer device in another embodiment. DETAILED DESCRIPTION

[0023] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0024] The answering method provided in the embodiment of the present application can be applied in the application environment shown in FIG1. ​​The application environment includes a terminal 102 and an intelligent customer service system 104. The terminal 102 is a terminal used by a target object, and the target object may be a customer served by a financial institution. The intelligent customer service system 104 is a system for providing automatic customer service to customers of the financial institution, and is deployed on the server of the financial institution. The intelligent customer service system 104 includes a general business question-and-answer system and multiple professional field question-and-answer systems, wherein the terminal 102 communicates with the intelligent customer service system 104 through a network.

[0025] Specifically, the intelligent customer service system 104 identifies the inquiry intent of the target object's target inquiry information based on the general business question and answer system, obtains the intention recognition result, and determines the target professional field indicated by the intention recognition result; then calls the professional field question and answer system corresponding to the target professional field to determine the initial response information that matches the target inquiry information; obtains the historical inquiry information of the target object within the historical time period; performs expression style analysis based on the historical inquiry information to obtain the expression style of the target object; generates target response information that conforms to the expression style based on the initial response information, and returns the target response information to the terminal 102 of the target object.

[0026] The terminal 102 may be, but is not limited to, various personal computers, laptop computers, smart phones, and tablet computers. The server deployed by the intelligent customer service system 104 may be implemented as an independent server or a server cluster consisting of multiple servers.

[0027] Those skilled in the art can understand that the application environment shown in FIG. 1 is only a partial scenario related to the present application solution and does not constitute a limitation on the application environment of the present application solution.

[0028] In some embodiments, as shown in FIG. 2 , a response method is provided, which is described by taking the method applied to the intelligent customer service system 104 in FIG. 1 as an example, including the following steps:

[0029] Step 202 , based on the general business question-answering system, the target query information of the target object is subjected to inquiry intent recognition, an intent recognition result is obtained, and a target professional field indicated by the intent recognition result is determined.

[0030] Among them, the intelligent customer service system includes a general business question and answer system and multiple professional field question and answer systems. The intelligent customer service system belongs to a financial institution and can automatically answer questions raised by customers of the financial institution. The general business question and answer system can identify the inquiry intention of the target object's target inquiry information. The target object is any one of all customers of the financial institution, and the target inquiry information is the inquiry information raised by the target object. The professional field question and answer system is used to provide more professional answers to relevant questions in the professional field to which it belongs. Each professional field question and answer system belongs to a different professional field. The professional field can be a business field of a financial institution, including at least one of the investment and financial management field, loan field, deposit field, credit card field, etc. The intention recognition result represents the target professional field that the target inquiry information intends to inquire about. The target professional field is at least one, and can be one or more.

[0031] Specifically, the general business question-and-answer system receives the target inquiry information sent by the terminal of the target object, and performs inquiry intent recognition on the target inquiry information. For example, the target inquiry information can be input into a trained inquiry intent recognition model to obtain an intention recognition result, and then the target professional field indicated by the intention recognition result is determined. For example, the intention recognition result can include at least one field keyword, and the target professional field can be determined based on the professional field corresponding to each field keyword. For example, if the field keyword is identified as "saving money", the target professional field corresponding to "saving money" is the deposit field. Among them, the trained inquiry intent recognition model can be trained based on the BERT model.

[0032] In some embodiments, when the intent recognition result does not indicate a professional field, the general business question-and-answer system can generate initial response information that matches the target inquiry information. The general business question-and-answer system can input the target inquiry information into the general business response model and output the initial response information. The general business response model is trained based on a general corpus such as conventional business knowledge and daily conversations of financial institutions, and has the function of determining the initial response information that matches the target inquiry information from the general corpus. The general business response model can be a deep learning model, specifically one of a recurrent neural network model, a seq2seq model, or a transformer model.

[0033] Step 204: call the professional field question-answering system corresponding to the target professional field to determine initial response information that matches the target query information.

[0034] The initial response information is text information that matches the target query information, and may be composed of multiple corpus paragraphs determined from a target corpus in the target professional field. The target corpus is a professional corpus corresponding to the target professional field, and each professional field corresponds to a professional corpus, which includes the professional knowledge of the professional field.

[0035] Specifically, when there is only one target professional field, the intelligent customer service system can call the professional field question-and-answer system of the target professional field, and the professional field question-and-answer system of the target professional field can determine the initial response information that matches the target inquiry information, and can input the target inquiry information into the target professional response model, and output the initial response information that matches the target inquiry information. Among them, the target professional response model is obtained by further training based on the target corpus of the target professional field on the basis of the general business response model, and has the function of determining the initial response information that matches the target inquiry information from the target corpus.

[0036] In some embodiments, there are multiple target professional fields, and the intelligent customer service system can call the professional field question and answer systems corresponding to each target professional field. The professional field question and answer system of each target professional field can determine the candidate response information that matches the target inquiry information, and combine the candidate response information corresponding to each target professional field to obtain the initial response information.

[0037] Step 206: Obtain historical query information of the target object within a historical time period.

[0038] The historical time period may be a preset time period, for example, it may be set to within one month or one year. The historical inquiry information includes all inquiry information raised by the target object within the historical time period.

[0039] Specifically, the database of the intelligent customer service system stores historical query information corresponding to each object, and the historical query information of the target object can be found from the database according to the object identifier of the target object. The object identifier is used to uniquely identify the identity of the object, which can be an ID number or a mobile phone number.

[0040] Step 208: Perform expression style analysis based on the historical query information to obtain the expression style of the target object.

[0041] Among them, the expression style represents the target object's style preference in language expression. For example, the expression style can be concise and refined, serious and detailed, enthusiastic and lively, etc.

[0042] Specifically, since different objects have different personality traits and social identities, each object also has different language expression styles. In order to make the response information more suitable for the expression style of the target object, the expression style analysis can be performed on the historical query information of the target object. For example, the historical query information can be input into the expression style analysis model for expression style analysis to obtain the expression style of the target object. The expression style analysis model has the function of analyzing the language expression style corresponding to the text content according to the input text content.

[0043] Step 210: Generate target response information that meets the expression style based on the initial response information, and return the target response information to the terminal of the target object.

[0044] The target response information is response information that conforms to the expression style of the target object.

[0045] Specifically, the intelligent customer service system can input the expression style identifier and the initial response information into the trained text generation model to generate the target response information that matches the expression style of the target object, and then send the target response information to the target object's terminal. The target object's terminal displays the target response information, and the target object can also input the corresponding inquiry information according to the target response information for the next round of inquiry.

[0046] In the above-mentioned answering method, the target inquiry information of the target object is subjected to inquiry intention recognition based on the general business question and answer system to obtain the intention recognition result, and the target professional field indicated by the intention recognition result is determined, and the professional field question and answer system corresponding to the target professional field is called to determine the initial response information that matches the target inquiry information. On the basis of determining the more professional initial response information, the expression style of the target object is obtained by performing expression style analysis based on the historical inquiry information of the target object, and the target response information that conforms to the expression style is generated based on the initial response information, and the target response information is returned to the terminal of the target object, so that the target response information is more in line with the personalized needs of the target object, thereby improving the accuracy of the customer service response.

[0047] In some embodiments, a professional field question and answer system corresponding to the target professional field is called to determine initial response information that matches the target query information, including: calling the professional field question and answer system corresponding to the target professional field to generate candidate response information; when there is related field information in the candidate response information, calling the professional field question and answer system corresponding to the related field information to generate auxiliary information under the related field; generating initial response information based on the candidate response information and the auxiliary information.

[0048] Among them, the candidate answer information is generated by calling the professional field question-answering system corresponding to the target professional field. The related field is other professional fields related to the target professional field, which are determined based on the candidate answer information. The auxiliary information is generated by the professional field question-answering system corresponding to the related field information, and is information belonging to the related field.

[0049] Specifically, first, the professional field question and answer system corresponding to the target professional field can generate candidate response information that matches the target query information, and then perform related field detection on the candidate response information based on the target query information. When there is related field information in the selected response information, the professional field question and answer system corresponding to the related field information is called to generate auxiliary information under the related field, and then the candidate response information and the auxiliary information are combined to obtain the initial response information.

[0050] For example, assuming that the target object wants to obtain investment and financial management recommendations based on its own asset information, the target inquiry information may be: "Personal asset information is: deposits of xxxxxx yuan, no other fixed assets, no loans. How should I plan my investment and financial management if I want to obtain a return of more than 30% within two years?" The professional field question-and-answer system in the field of investment and financial management can generate investment and financial management planning information that meets the target inquiry information as candidate response information. However, if the deposit amount of the target object cannot meet the starting amount of the planning requirement, it is determined that there is related field information, namely loan field information, in the candidate response information, and the professional field question-and-answer system corresponding to the loan field is called to generate loan knowledge as auxiliary information under the related field, so that the investment and financial management planning information and loan knowledge can be returned to the target object together.

[0051] In this embodiment, when there is related field information in the candidate answer information, the professional field question and answer system corresponding to the related field information is called to generate auxiliary information under the related field, and initial answer information is generated based on the candidate answer information and the auxiliary information. This realizes the linkage between the professional field question and answer systems corresponding to different professional fields, can capture the target user's deeper inquiry needs, and improves the accuracy of customer service responses.

[0052] In some embodiments, the response method also includes: obtaining object attribute information of the target object with authorization from the target object; determining a recommended information style based on the object attribute information; generating target response information based on the expression style and the initial response information, and returning the target response information to the terminal of the target object, including: returning the target response information and the recommended information style to the terminal of the target object to instruct the terminal of the target object to display the target response information according to the recommended information style.

[0053] The object attribute information is information related to the personal attributes of the object, including at least one of the user's age, gender, location, occupation, activity, etc. The information style refers to the style of displaying the target response information on the terminal, including at least one of text style, picture style, chart style, audio and video style, etc. The recommended information style is determined based on the object attribute information and the target response information.

[0054] Specifically, the intelligent customer service system can classify the target object according to the object attribute information of the target user to obtain the target object type to which the target object belongs. For example, the object attribute information of the target object can be input into the object portrait model to obtain the target object type. The object type is used to divide the object group. The object type may include: business people, students, retired elderly people, etc. The intelligent customer service system may pre-store a mapping relationship table consisting of each user type and information style, and can search the mapping relationship table to obtain the recommended information style according to the target user type.

[0055] In this embodiment, the recommended information style is determined based on the object attribute information, and the target response information and the recommended information style are returned to the terminal of the target object to instruct the terminal of the target object to display the target response information according to the recommended information style, thereby providing the target object with a suitable information style to meet the needs of different customers and achieve personalized customer service responses.

[0056] In some embodiments, based on the general business question-answering system, the target query information of the target object is subjected to query intent recognition, the intent recognition result is obtained, and the target professional field indicated by the intent recognition result is determined, including: extracting at least one keyword from the target query information; identifying a field keyword from at least one keyword; and determining the target professional field based on the field keyword.

[0057] Among them, keywords are extracted from the target query information, and the keywords include domain keywords and non-domain keywords. Domain keywords are keywords that represent professional fields, and non-domain keywords can be words commonly used in daily conversations.

[0058] Specifically, the intelligent customer service system can extract at least one keyword from the target inquiry information, identify the professional field for each keyword, and determine the keyword as the field keyword when the professional field corresponding to the keyword is identified. If there are multiple field keywords, the professional fields corresponding to the multiple field keywords can be counted. If each field keyword corresponds to the same professional field, the professional field is determined as the target professional field. If there are at least two professional fields corresponding to multiple field keywords, at least two professional fields corresponding to each field keyword are determined as the target professional field.

[0059] In this embodiment, by extracting at least one keyword from the target inquiry information, identifying the domain keyword from the at least one keyword, and determining the target professional field based on the domain keyword, the target professional field intended by the target inquiry information can be accurately identified, thereby improving the accuracy of customer service responses.

[0060] In some embodiments, calling the professional field question and answer system corresponding to the target professional field to determine the initial response information that matches the target query information includes: calling the professional field question and answer system corresponding to the target professional field to determine the target corpus corresponding to the target professional field; for each corpus paragraph in the target corpus, calculating the information matching degree between the corpus paragraph and the target query information; determining the corpus paragraph that matches the target query information from the target corpus according to the information matching degree to obtain the initial response information.

[0061] Among them, the target corpus is a professional corpus corresponding to the target professional field. A professional corpus includes multiple corpus paragraphs, and each corpus paragraph has its own corpus features. The corpus features are obtained by pre-extracting the features of the corpus paragraphs. The information matching degree represents the matching degree between the target query information and the corpus paragraphs.

[0062] Specifically, the professional field question-answering system first determines the target corpus corresponding to the target professional field, and determines the corpus features corresponding to each corpus paragraph in the target corpus, and then extracts features from the target query information to obtain information features. The professional field question-answering system calculates the information matching degree between the information features of the target query information and the corpus features of each corpus paragraph through the target professional response model, and determines the corpus paragraphs that match the target query information from the target corpus according to the information matching degree. For example, a preset number of corpus paragraphs can be selected from the target corpus in order of information matching degree from large to small, and the corpus paragraphs that match the target query information are obtained to form the initial response information. Among them, the preset number is a pre-set positive integer, which can be 3.

[0063] In this embodiment, by calculating the information matching degree between the corpus paragraph and the target query information, the corpus paragraph matching the target query information is determined from the target corpus according to the information matching degree, and the initial response information matching the target query information is obtained, thereby improving the accuracy of the customer service response.

[0064] In some embodiments, before the step of identifying the inquiry intention of the target query information of the target object based on the general business question and answer system, the method also includes: receiving the initial query information sent by the terminal of the target object; identifying the language type of the initial query information to obtain the object language type corresponding to the initial query information; converting the initial query information into target query information expressed in a preset language type; generating target response information that conforms to the expression style based on the initial response information, and returning the target response information to the terminal of the target object, including: converting the target response information into target response information expressed in the object language type, and returning the target response information expressed in the object language type to the terminal of the target object.

[0065] The initial query information is sent by the terminal of the target object, and can be expressed in a preset language or other language. The preset language is pre-set, and the preset language can be one or more, including at least one of Chinese, English, French, Russian, and Spanish. The sentences in the general corpus and each professional corpus are expressed in the preset language. The object language is the language used by the target object, which can be a preset language or other language.

[0066] Specifically, the intelligent customer service system receives an initial inquiry message sent by a terminal of a target object, first identifies the language type of the initial inquiry message to obtain the object language type, and when the object language type belongs to a preset language type, determines the initial inquiry message as the target inquiry message; when the object language type does not belong to the preset language type, translates the initial inquiry message into target inquiry message expressed in any preset language type, and after generating target response information expressed in the preset language type, it is also necessary to translate the target response information into response information expressed in the object language type so that the target object can read it directly.

[0067] In this embodiment, since the objects served by the financial institution can come from different regions and use different languages, and the language types used in the model training in the intelligent customer service system are limited, by identifying the language type of the object and automatically converting the language type, customers do not need to manually switch the language type when making inquiries, thereby realizing multi-language customer service responses.

[0068] In some embodiments, as shown in FIG3 , a response method is provided, which is described by taking the method applied to the intelligent customer service system in FIG1 as an example, and includes the following steps:

[0069] Step 302: Receive the initial inquiry information sent by the terminal of the target object, identify the language type of the initial inquiry information, and obtain the target language type corresponding to the initial inquiry information.

[0070] Step 304: convert the initial query information into target query information expressed in a preset language.

[0071] Step 306 , based on the general business question-answering system, the target query information of the target object is subjected to inquiry intent recognition, an intent recognition result is obtained, and a target professional field indicated by the intent recognition result is determined.

[0072] Step 308: call the professional field question-answering system corresponding to the target professional field to generate candidate answer information.

[0073] Step 310: When there is related field information in the candidate response information, the professional field question answering system corresponding to the related field information is called to generate auxiliary information under the related field.

[0074] Step 312: Generate initial response information based on the candidate response information and the auxiliary information.

[0075] Step 314: Obtain historical query information of the target object within a historical time period, perform expression style analysis based on the historical query information, and obtain the expression style of the target object.

[0076] Step 316: Generate target response information that conforms to the expression style based on the initial response information.

[0077] Step 318, convert the target response information into the target response information expressed in the object language type, and return the target response information expressed in the object language type to the terminal of the target object.

[0078] In this embodiment, by identifying the object language type and automatically converting the language type, the intelligent customer service system can use the object language type to respond to the target object. There is no need for customers to manually switch the language type when making inquiries, thereby realizing multi-language customer service responses; by identifying the inquiry intention of the target inquiry information, the target professional field is obtained, and the professional field question and answer system corresponding to the target professional field is called to determine the candidate response information that matches the target inquiry information, thereby determining more professional candidate response information; when there is related field information in the candidate response information, the professional field question and answer system corresponding to the related field information is called to generate auxiliary information under the related field, and the initial response information is generated based on the candidate response information and the auxiliary information, thereby realizing the linkage between the professional field question and answer systems corresponding to different professional fields, being able to capture the target user's deeper inquiry needs, and generating target response information that conforms to the expression style based on the initial response information, thereby improving the accuracy of the customer service response.

[0079] It should be understood that, although the steps in the flowcharts involved in the above embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above embodiments may include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0080] Based on the same inventive concept, the embodiment of the present application also provides a response device for implementing the response method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more response device embodiments provided below can refer to the limitations on the response method above, and will not be repeated here.

[0081] In some embodiments, as shown in FIG. 4 , a response device is provided, including: an intention recognition module 402, an initial information determination module 404, a historical information acquisition module 406, a style analysis module 408, and an information generation module 410, wherein:

[0082] The intention recognition module 402 is used to perform inquiry intention recognition on the target inquiry information of the target object based on the general business question-answering system, obtain the intention recognition result, and determine the target professional field indicated by the intention recognition result.

[0083] The initial information determination module 404 is used to call the professional field question-answering system corresponding to the target professional field to determine the initial response information that matches the target inquiry information.

[0084] The historical information acquisition module 406 is used to acquire the historical query information of the target object within a historical time period.

[0085] The style analysis module 408 is used to perform expression style analysis based on historical query information to obtain the expression style of the target object.

[0086] The information generation module 410 is used to generate target response information that meets the expression style based on the initial response information, and return the target response information to the terminal of the target object.

[0087] In some embodiments, the initial information determination module 404 is also used to: call the professional field question and answer system corresponding to the target professional field to generate candidate response information; when there is related field information in the candidate response information, call the professional field question and answer system corresponding to the related field information to generate auxiliary information under the related field; generate initial response information based on the candidate response information and the auxiliary information.

[0088] In some embodiments, the response device also includes a style recommendation module, which is used to: obtain object attribute information of the target object with the authorization of the target object; determine the recommended information style based on the object attribute information; the information generation module 410 is also used to: return the target response information and the recommended information style to the terminal of the target object to instruct the terminal of the target object to display the target response information according to the recommended information style.

[0089] In some embodiments, the intent recognition module 402 is further used to: extract at least one keyword from the target query information; identify a domain keyword from the at least one keyword; and determine a target professional field based on the domain keyword.

[0090] In some embodiments, the initial information determination module 404 is also used to: call the professional field question and answer system corresponding to the target professional field, and determine the target corpus corresponding to the target professional field; for each corpus paragraph in the target corpus, calculate the information matching degree between the corpus paragraph and the target query information; determine the corpus paragraph that matches the target query information from the target corpus according to the information matching degree, and obtain the initial response information.

[0091] In some embodiments, the response device also includes a language identification module, which is used to: receive the initial inquiry information sent by the terminal of the target object; identify the language type of the initial inquiry information to obtain the object language type corresponding to the initial inquiry information; convert the initial inquiry information into target inquiry information expressed in a preset language type; the information generation module 410 is also used to: convert the target response information into target response information expressed in the object language type, and return the target response information expressed in the object language type to the terminal of the target object.

[0092] Each module in the above-mentioned response device can be implemented in whole or in part by software, hardware or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute the operations corresponding to each module above.

[0093] In some embodiments, a computer device is provided, which may be a server, and its internal structure diagram may be shown in FIG5. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store relevant data involved in the answering method. The input / output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a response method is implemented.

[0094] In some embodiments, a computer device is provided, which may be a terminal, and its internal structure diagram may be shown in FIG6. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit, and an input device. The processor, the memory, and the input / output interface are connected via a system bus, and the communication interface, the display unit, and the input device are connected to the system bus via the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a response method is implemented. The display unit of the computer device is used to form a visually visible picture, which may be a display screen, a projection device, or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device shell, or an external keyboard, touchpad or mouse.

[0095] Those skilled in the art will understand that the structures shown in FIGS. 5 and 6 are merely block diagrams of partial structures related to the scheme of the present application, and do not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device may include more or fewer components than those shown in the figures, or combine certain components, or have a different arrangement of components.

[0096] In some embodiments, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned response method when executing the computer program.

[0097] In some embodiments, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the steps in the above-mentioned response method are implemented.

[0098] In some embodiments, a computer program product is provided, including a computer program, which implements the steps in the above-mentioned response method when executed by a processor.

[0099] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards of relevant countries and regions.

[0100] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0101] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0102] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A response method, It is characterized in that The method comprises: Based on the general business question-answering system, the target query information of the target object is subjected to inquiry intent recognition, an intention recognition result is obtained, and a target professional field indicated by the intention recognition result is determined; Calling the professional field question-answering system corresponding to the target professional field to determine initial response information matching the target inquiry information; Obtaining historical query information of the target object within a historical time period; Performing expression style analysis based on the historical query information to obtain the expression style of the target object; Target response information that conforms to the expression style is generated based on the initial response information, and the target response information is returned to the terminal of the target object.

2. The method according to claim 1, It is characterized in that The calling of the professional field question-answering system corresponding to the target professional field to determine the initial response information matching the target inquiry information includes: Calling the professional field question-answering system corresponding to the target professional field to generate candidate answer information; In the case where there is related field information in the candidate answer information, calling the professional field question answering system corresponding to the related field information to generate auxiliary information under the related field; Initial response information is generated according to the candidate response information and the auxiliary information.

3. The method according to claim 1, It is characterized in that The method further comprises: Obtaining object attribute information of the target object with authorization from the target object; Determining a recommendation information style based on the object attribute information; The generating target response information based on the expression style and the initial response information, and returning the target response information to the terminal of the target object includes: The target response information and the recommended information style are returned to the terminal of the target object, so as to instruct the terminal of the target object to display the target response information according to the recommended information style.

4. The method according to claim 1, It is characterized in that The general business question-answering system performs inquiry intent recognition on the target inquiry information of the target object to obtain an intent recognition result, and determines that the target professional field indicated by the intent recognition result includes: Extracting at least one keyword from the target query information; From the at least one keyword, identifying a domain keyword; The target professional field is determined based on the field keywords.

5. The method according to claim 1, It is characterized in that The calling of the professional field question-answering system corresponding to the target professional field to determine the initial response information matching the target inquiry information includes: Calling a professional field question answering system corresponding to the target professional field to determine a target corpus corresponding to the target professional field; For each corpus paragraph in the target corpus, calculating the information matching degree between the corpus paragraph and the target query information; According to the information matching degree, a corpus paragraph matching the target query information is determined from the target corpus to obtain initial response information.

6. The method according to claim 1, It is characterized in that Before the step of identifying the inquiry intention of the target inquiry information of the target object based on the general business question-answering system, the method further includes: Receiving initial inquiry information sent by the terminal of the target object; Performing language identification on the initial inquiry information to obtain the target language type corresponding to the initial inquiry information; Converting the initial query information into target query information expressed in a preset language; The generating target response information conforming to the expression style based on the initial response information, and returning the target response information to the terminal of the target object comprises: The target response information is converted into target response information expressed in the target language, and the target response information expressed in the target language is returned to the terminal of the target object.

7. A response device, It is characterized in that The device comprises: An intention recognition module is used to recognize the inquiry intention of the target object's target inquiry information based on the general business question-answering system, obtain the intention recognition result, and determine the target professional field indicated by the intention recognition result; An initial information determination module, used to call the professional field question-answering system corresponding to the target professional field, and determine the initial response information matching the target inquiry information; A historical information acquisition module, used to acquire the historical query information of the target object within a historical time period; A style analysis module, used to perform expression style analysis based on the historical query information to obtain the expression style of the target object; The information generation module is used to generate target response information that conforms to the expression style based on the initial response information, and return the target response information to the terminal of the target object.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program. It is characterized in that When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, It is characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product comprising a computer program, It is characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images