Call business card interaction method and device, equipment, storage medium and product

Abstract

The invention discloses a call business card interaction method and device, equipment, a storage medium and a product, and the method comprises the steps: receiving the business card content and at least one business card information point of an opposite user when a call occurs between a local user and the opposite user; wherein the business card information point is generated according to the business card content of the opposite user; and in response to the operation of selecting the business card information point by the local user, obtaining a reply text preset by the opposite user and corresponding to the selected business card information point, and displaying the reply text to the local user, so that the purposes of capturing user feedback and making a response in time are achieved, the information required by the user can be presented more efficiently, and the user experience is improved. And the information bearing capacity of the business card is also improved.

Description

Technical Field

[0001] The present invention relates to the field of call technology, and in particular to a call business card interaction method, device, equipment, storage medium and product. Background Art

[0002] Existing call business card services mainly rely on flash text or video ringback tone to convey information, such as using flash text or video ringback tone to directly convey the contact's name, phone number, company name and email address. These two forms lack an effective interaction mechanism and cannot capture user feedback and make timely responses. Summary of the invention

[0003] Based on this, the present invention provides a call business card interaction method, device, equipment, storage medium and product to solve the defects of the prior art that user feedback cannot be captured and responded to in a timely manner.

[0004] To achieve the above object, an embodiment of the present invention provides a call business card interaction method, including:

[0005] When a call occurs between the user of the local party and the user of the other party, receiving the business card content and at least one business card information point of the user of the other party; wherein the business card information point is generated according to the business card content of the user of the other party;

[0006] In response to the operation of the local user selecting the business card information point, a reply text preset by the other party user corresponding to the selected business card information point is obtained and displayed to the local user.

[0007] As an improvement of the above solution, the business card information point is specifically generated in the following manner:

[0008] Extracting characteristic information of the business card content as a target demand point;

[0009] Taking the target demand point as a search condition, finding the target information point from a pre-constructed relationship map between information points and demand points; wherein the relationship map is constructed based on the demand points extracted from the questions of the pre-acquired second basic question-answer pair data and the information points extracted from the answers of the second basic question-answer pair data;

[0010] The target information point and the business card content are input into a pre-trained business card information point generation model to obtain the business card information point; wherein the business card information point generation model is trained based on the answers and information points in the pre-acquired first basic question-answer pair data.

[0011] As an improvement to the above solution, the business card information point generation model is trained in the following way:

[0012] Acquire interactive data of the call scenario, perform data cleaning and dialogue splitting on the interactive data, and construct the first basic question-answer pair data;

[0013] Using a natural language processing model, extracting summaries of answers in the first basic question-answer pair data to generate information points of the answers in the first basic question-answer pair data;

[0014] The answers and information points in the first basic question and answer data are used to perform model training to obtain a business card information point generation model.

[0015] As an improvement of the above scheme, the number of the target information points is at least one, the semantic matching degree between the target information point and the target demand point is greater than a set similarity threshold, and the semantic matching degree between the target information point and the target demand point is calculated by calling a semantic matching algorithm.

[0016] As an improvement of the above scheme, the reply text is pre-adjusted by the other user according to the target answer template for the business card information point, and the target answer template is retrieved from the relationship map based on the business card information point, wherein the relationship map records the correspondence between the information point and the reply template.

[0017] As an improvement of the above solution, the relationship map is constructed in the following way:

[0018] Acquire call interaction text data, perform data cleaning and dialogue splitting on the call interaction text data, and obtain the second basic question-answer pair data;

[0019] Using natural language processing technology to extract demand points in the questions of the second basic question-answer pair data, and extract information points in the answers of the second basic question-answer pair data, so as to form triple data using the demand points of the second basic question-answer pair data and the information points of the second basic question-answer pair data;

[0020] The triple data is disambiguated, and the disambiguated triple data is imported into a graph database to construct the relationship graph.

[0021] To achieve the above purpose, an embodiment of the present invention further provides a call business card interaction device, including:

[0022] An information receiving module, used for receiving the business card content and at least one business card information point of the other party user when a call occurs between the user of the local party and the other party user; wherein the business card information point is generated according to the business card content of the other party user;

[0023] The text acquisition module is used to respond to the operation of the local user selecting the business card information point, obtain the reply text preset by the other user corresponding to the selected business card information point and display it to the local user.

[0024] To achieve the above objectives, an embodiment of the present invention also provides a call business card interaction device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, and when the processor executes the computer program, it implements the call business card interaction method described in any of the above embodiments.

[0025] To achieve the above objectives, an embodiment of the present invention further provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the call business card interaction method as described in any of the above embodiments.

[0026] To achieve the above objectives, an embodiment of the present invention further provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the call business card interaction method as described in any of the above embodiments.

[0027] Compared with the prior art, the call business card interaction method, device, equipment, storage medium and product disclosed in the embodiments of the present invention receive the business card content and at least one business card information point of the other user when a call occurs between the local user and the other user; wherein the business card information point is generated according to the business card content of the other user; in response to the operation of the local user selecting the business card information point, the reply text preset by the other user corresponding to the selected business card information point is obtained and displayed to the local user, thereby achieving the purpose of capturing user feedback and making timely responses, being able to more efficiently present the information required by the user, and also improving the information carrying capacity of the business card. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the technical solution of the present invention, the drawings required for use in the implementation mode will be briefly introduced below. Obviously, the drawings described below are only some implementation modes of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0029] Figure 1 It is a flowchart of a call business card interaction method provided by an embodiment of the present invention;

[0030] Figure 2 It is a flow chart of call business card interaction provided by an embodiment of the present invention;

[0031] Figure 3 It is a structural schematic diagram of a call business card interaction device provided by an embodiment of the present invention;

[0032] Figure 4 It is a structural diagram of a call business card interaction device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0033] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0034] See also Figure 1 , Figure 1 1 is a flow chart of a call business card interaction method provided by an embodiment of the present invention. Specifically, the call business card interaction method includes steps S11 to S12:

[0035] S11. When a call occurs between a user of the local party and a user of a remote party, receiving business card content and at least one business card information point of the remote party user; wherein the business card information point is generated according to the business card content of the remote party user.

[0036] Specifically, the business card information point is generated based on the business card content preset by the other user, and is a summary of the key points of the business card content. In general, the business card information point is relatively brief and serves as a guide. The user of the local party can obtain more detailed content by clicking on the business card information point, which enriches and improves the information carrying capacity of the call business card. The number of business card information points is determined according to the actual situation, and can be one, two, three or more, and is not limited here. In addition, when receiving the business card information point of the other user, the business card content of the other user can also be received.

[0037] S12: In response to the operation of the local user selecting the business card information point, obtaining a reply text preset by the other party user corresponding to the selected business card information point and displaying it to the local user.

[0038] Specifically, the user can select the business card information point by touch, by key, or by voice, etc. The selection operation of the business card information point is not limited to the above methods and can be set according to actual needs, and is not limited here.

[0039] It is worth noting that the method can be applied to a terminal, and the method can also be applied to a server. The terminal can be a mobile phone, a tablet, a notebook or other devices. The terminal can be the terminal of the called party or the terminal of the calling party. The method is applied in a call scenario, and can occur during a call, during a call, and after the call ends, etc., and is not limited here.

[0040] Exemplarily, taking the method applied to a terminal and the application period being a call as an example, during the call (before the call is connected), the terminal of the local user receives the business card information points of the other user sent by flash message or other means and displays them on the terminal screen of the local user (or plays them on the user's terminal by voice, or presents the business card information points to the local user on the terminal of the local user in other ways), the local user selects the received business card information points by click / voice, and the terminal obtains the reply text corresponding to the selected business card information point pre-set by the other user to present it on the terminal, thereby completing effective interaction of business cards.

[0041] When presenting the other party's information, the prior art usually presents all the content set by the other party user to the local user together. Therefore, the total content set by the other party user is limited in length, the information carrying capacity is limited, and the information required by the local user cannot be presented efficiently. Compared with the prior art, the embodiment of the present invention receives the business card information point of the other party user, and based on the operation of the local user selecting the business card information point, obtains the reply text preset by the other party user corresponding to the selected business card information point; wherein the business card information point is generated according to the business card content preset by the other party user, thereby achieving the purpose of capturing the feedback of the local user and responding in time, and can more efficiently present the information required by the local user, and also improves the information carrying capacity of the business card.

[0042] In one implementation, the business card information point is generated in the following manner:

[0043] Extracting characteristic information of the business card content as a target demand point;

[0044] Taking the target demand point as a search condition, finding the target information point from a pre-constructed relationship map between information points and demand points; wherein the relationship map is constructed based on the demand points extracted from the questions of the pre-acquired second basic question-answer pair data and the information points extracted from the answers of the second basic question-answer pair data;

[0045] The target information point and the business card content are input into a pre-trained business card information point generation model to obtain the business card information point; wherein the business card information point generation model is trained based on the answers and information points in the pre-acquired first basic question-answer pair data.

[0046] Preferably, the type of the relationship graph is a knowledge graph. A knowledge graph is a graph-based data structure that represents entities and their relationships in the real world. This data structure represents the relationships and attributes between things through a graphical model, thereby structuring, organizing and representing knowledge.

[0047] Specifically, the specific method of generating business card information points is as follows:

[0048] 1. Get the target information point according to the preset business card content.

[0049] First, the parsing ability of the large language model is used to parse the content features of the business card from the pre-set business card content as the target demand points (such as the designer's intention, the user's expression style, the demand characteristics of the target group, etc.). Then, the parsed business card content features are combined according to the preset combination rules as the retrieval conditions of the graph data, and the relevant information points are retrieved from the knowledge graph through the graph retrieval technology. Finally, the information points recalled from each channel are fused to obtain the candidate pool of interactive information points of the business card (i.e., the target information point). The specific implementation method is as follows:

[0050] (1) Prompt construction for business card content feature analysis: A large model prompt is constructed based on the text content of the pre-set business card content, and the large model prompt is enhanced with information through the fewshot learning method. Among them, prompt refers to the text or instruction that provides input to the model to guide it to generate specific output. It is a text paragraph provided by the user when interacting with the model, which is used to describe the information, answer, text, etc. that the user wants to obtain from the model. The purpose of the prompt is to guide the model to produce the required response in order to better control the generated output.

[0051] (2) Semantic parsing: Use the prompt-guided open-source large language model Qwen-14B constructed in 1(1) to parse the business card content features such as the designer’s intention, the user’s expression style, and the needs of the target group. Qwen-14B is a Transformer-based large language model trained on ultra-large-scale pre-training data. Transformer is a deep learning model architecture used for natural language processing and sequence-to-sequence tasks. It introduces a self-attention mechanism that can simultaneously consider all positions in the input sequence.

[0052] (3) Multi-channel recall based on knowledge graph: Based on the pre-constructed call information interaction knowledge graph (the relationship graph between information points and demand points), through the semantic matching algorithm (such as the semantic vector matching graph retrieval algorithm), for different combinations (intention, expression style, target demand characteristics), information points whose matching degree meets the preset threshold are retrieved.

[0053] (4) Information fusion: The multi-channel information points recalled in step 1(3) are weightedly fused by a weighted fusion algorithm to generate a candidate pool of interactive information points (i.e., target information points) for the calling business card.

[0054] 2. Obtain the business card information point based on the target information point and the business card content.

[0055] The preset business card content is used as personalized guidance information, and the content of the interactive information point candidate pool generated above is used as basic information. The two parts of information are spliced ​​using the large model prompt engineering technology to obtain the information point generation prompt, and then input into the fine-tuned large language model (i.e., the business card information point generation model), and the large model text generation capability is used to automatically generate personalized information point text (i.e., business card information point) that conforms to the spatiotemporal characteristics of the business card. The specific implementation includes:

[0056] (1) Information point generation prompt construction: A large model prompt is constructed based on the business card content pre-set by the user and the content of the interactive information point candidate pool of the call business card, and the large model prompt is enhanced with information through the fewshot method.

[0057] (2) Text generation: The prompt constructed in 2(1) is input into the fine-tuned Qwen-14B large language model. The large model's text generation capability is used to obtain personalized interaction information that is consistent with the user's preset content characteristics and suitable for the spatiotemporal characteristics of the call business card.

[0058] In one implementation, the business card information point generation model is trained in the following manner:

[0059] Acquire interactive data of the call scenario, perform data cleaning and dialogue splitting on the interactive data, and construct the first basic question-answer pair data;

[0060] Using a natural language processing model, extracting summaries of answers in the first basic question-answer pair data to generate information points of the answers in the first basic question-answer pair data;

[0061] The answers and information points in the first basic question and answer data are used to perform model training to obtain a business card information point generation model.

[0062] Specifically, the space available for displaying content in call cards is limited, and (in most scenarios) the time available for interaction is limited to the short time before the call is connected, so the interactive content of call cards must be concise, accurate, and attractive. In view of the special time and space limitations of call card application scenarios, Alibaba's open source Qwen-14B was selected as the basic large model in the selection of large language models. At the same time, the large language model was fine-tuned through the processed multi-scenario business question and answer dataset, so that the large language model has the ability to generate concise information points (i.e., business card information points) based on the edited content of the business card. The details are as follows:

[0063] (1) Data collection and organization: Business interaction data from common call scenarios are collected. After data cleaning and conversation splitting, basic question-and-answer data is constructed.

[0064] (2) Fine-tuning dataset construction: Using the text summarization and semantic understanding capabilities of the large language model, the answers to the question-answer data constructed in (1) are summarized and extracted to generate concise and attractive information point description content (i.e., the information points of the answer).

[0065] (3) Model fine-tuning: Using the question-answer pair data constructed in (2) as training data, the open source large language model Qwen-14B is fine-tuned through the Q-LoRa large model fine-tuning technology. After multiple rounds of self-supervised learning optimization, the large language model can generate concise information points that are strongly related to the business card content by inputting the business card content. Among them, Q-LoRa is a new method for fine-tuning large language models, which can save memory while maintaining speed.

[0066] In one implementation, the number of the target information point is at least one, the semantic matching degree between the target information point and the target demand point is greater than a set similarity threshold, and the semantic matching degree between the target information point and the target demand point is calculated by calling a semantic matching algorithm.

[0067] In one embodiment, the reply text is pre-adjusted by the other user based on a target answer template for the business card information point, and the target answer template is retrieved from a relationship map based on the business card information point, wherein the relationship map records the correspondence between the information point and the reply template.

[0068] Specifically, the reply text corresponding to the business card information point is added to the relationship graph in advance to provide the user with corresponding detailed content when the user clicks the business card information point. Taking the relationship graph as a knowledge graph as an example, after completing the interactive answer filling for the generated business card information point and clicking to publish, the system will automatically store the completed interactive configuration in the database and synchronize it to the call business card database in real time. The specific implementation process of the reply text configuration in the database is as follows:

[0069] (1) Interactive answer template filling: Based on each generated business card information point, the system will automatically generate an answer template for each business card information point by retrieving the pre-configured call information interaction knowledge graph (in addition to the correspondence between demand points and information points, the knowledge graph also has a correspondence between information points and answer templates). At the same time, it guides the business card publisher to adjust the content of the answer template according to his or her own situation.

[0070] (2) Storage of interactive configuration: After the publisher of the business card completes the adjustment of the interactive answer, click the publish button to complete the automatic storage of the interactive configuration (i.e., record the improved information in the knowledge graph), and the relevant data will be synchronized to the call business card database in real time. When looking for the corresponding reply text for the business card information point later, you can search in the knowledge graph or in the call business card database.

[0071] In one embodiment, the relationship map is constructed by:

[0072] Acquire call interaction text data, perform data cleaning and dialogue splitting on the call interaction text data, and obtain the second basic question-answer pair data;

[0073] Using natural language processing technology to extract demand points in the questions of the second basic question-answer pair data, and extract information points in the answers of the second basic question-answer pair data, so as to form triple data using the demand points of the second basic question-answer pair data and the information points of the second basic question-answer pair data;

[0074] The triple data is disambiguated, and the disambiguated triple data is imported into a graph database to construct the relationship graph.

[0075] Specifically, taking the relationship graph as a knowledge graph as an example, the construction process is as follows: First, based on the long-term accumulation of massive call interaction text data, a "call information interaction knowledge graph" containing key information nodes in various call scenarios is constructed, and the demand points and information points of different call scenarios and their relationships are stored in a graph, so as to use the efficient relationship retrieval capability of the knowledge graph to realize the recall of information points under various scenario trigger points. Specifically, it includes:

[0076] (1) Data preprocessing: Data cleaning, dialogue segmentation (to obtain basic question-answer pair data), and scene annotation of call interaction text data.

[0077] (2) Information extraction: Natural language processing (NLP) techniques such as entity recognition, relationship extraction, and semantic understanding are used to extract information points in the answers and demand points in the questions from the training data to form triple data.

[0078] (3) Knowledge fusion: Entity disambiguation and relationship disambiguation are performed on the extracted triple data to achieve accurate representation of entity nodes and relationships and remove duplicate data.

[0079] (4) Knowledge graph construction: The final generated triple data is imported into the graph database to complete the construction of the call information interaction knowledge graph, where the node types of the graph include demand points and information points, and the scene identifier where the interaction occurs is attached to the edge attributes of the graph.

[0080] Specifically, see Figure 2 The flowchart of call business card interaction shown in the figure is as follows: 1. Knowledge graph construction: by extracting information and fusing knowledge from call interaction text data, a call information interaction knowledge graph is obtained; 2. Large model fine-tuning: the fine-tuning data set is used to fine-tune the model to obtain a fine-tuned large language model (business card information point generation model); 3. Interaction information point candidate pool generation: the large model semantic analysis is used to generate an interaction information point candidate pool based on the knowledge graph multi-way recall; 4. Personalized interaction information generation: the prompt technology and the fine-tuned large language model are used to generate the text of the business card information point according to the interaction information point candidate pool and the business card content; 5. Interaction information configuration storage: the corresponding answer template is found according to the business card information point and adjusted and added to the call business card database; 6. Real-time interaction of call business cards: the business card information points recorded in the call business card database are displayed on the terminal. After the user clicks the corresponding business card information point, the adjusted answer template corresponding to the clicked business card information point is found from the call business card database and displayed to achieve real-time interaction.

[0081] In order to make the method easier to understand, a specific application scenario is used as an example below.

[0082] Scenario 1: Real estate sales

[0083] Business card content:

[0084] XX Real Estate is committed to providing one-stop solutions from high-end residential, commercial real estate to urban complexes, and its projects cover many hot spots across the country. The hot-selling [Golden Sunshine] high-end community is located in the prosperous area of ​​the city center, with complete supporting facilities and diverse apartment types. We sincerely invite you to come and taste it.

[0085] 1. Use the large language model to analyze the personalized information of the business card content and obtain the demand points of the business card content (i.e., target demand points):

[0086] Business card purpose: real estate sales; expression style: concise and clear; target group: high-end home buyers.

[0087] 2. Use the target demand points to find the corresponding information points in the knowledge graph and form a candidate pool of interactive information points.

[0088] 3. Combine the "candidate pool" and the business card content as the input of the business card information point generation model to obtain the business card information points:

[0089] Hot sale details, preferential measures, if you don’t buy a house, call again on the weekend.

[0090] 4. Click the business card information point [Preferential Measures]: (Use the business card information point to search the call business card database, find the reply text corresponding to [Preferential Measures] associated with the publisher of the business card content, and output the following details):

[0091] A one-time payment will reduce the total house price by 20,000 yuan. If you sign the contract within three days, you can enjoy a total house price discount of 21,000 yuan per unit.

[0092] Scenario 2: Catering customer service hotline

[0093] Business card content:

[0094] YY Food Group is a well-known enterprise that integrates diversified dining experiences. It owns many well-known brand restaurants and special dishes, providing you with safe, delicious and healthy dining enjoyment.

[0095] Personalized information analysis of business card content:

[0096] Business card purpose: restaurant recommendations; Expression style: polite and courteous; Target group: users of specialty foods.

[0097] Business card information point generation:

[0098] Restaurant location, special dishes, recent discounts, please call back for reservations.

[0099] Click on the business card information point [Special dishes]:

[0100] 1. Sweet and Sour Crispy Shrimp Chips (ranked first in reviews) 2. Fresh Crispy Pork (new hot seller) 3. Garlic Crispy Sea Bass (top 3 in reviews).

[0101] Compared with the prior art, the method provided in the embodiment of the present invention receives the business card content and at least one business card information point of the other user when a call occurs between the local user and the other user; wherein the business card information point is generated according to the business card content of the other user; in response to the operation of the local user selecting the business card information point, the reply text preset by the other user corresponding to the selected business card information point is obtained and displayed to the local user, thereby achieving the purpose of capturing user feedback and making timely responses, being able to present the information required by the user more efficiently, and improving the information carrying capacity of the business card.

[0102] See also Figure 3 The embodiment of the present invention further provides a call business card interaction device, including:

[0103] The information receiving module 21 is used to receive the business card content and at least one business card information point of the other party user when a call occurs between the user of the local party and the other party user; wherein the business card information point is generated according to the business card content of the other party user;

[0104] The text acquisition module 22 is used for, in response to the operation of the local user selecting the business card information point, acquiring the reply text preset by the other user corresponding to the selected business card information point and displaying it to the local user.

[0105] In one embodiment, the device further includes a business card information point generation module, which is used to:

[0106] Extracting characteristic information of the business card content as a target demand point;

[0107] Taking the target demand point as a search condition, finding the target information point from a pre-constructed relationship map between information points and demand points; wherein the relationship map is constructed based on the demand points extracted from the questions of the pre-acquired second basic question-answer pair data and the information points extracted from the answers of the second basic question-answer pair data;

[0108] The target information point and the business card content are input into a pre-trained business card information point generation model to obtain the business card information point; wherein the business card information point generation model is trained based on the answers and information points in the pre-acquired first basic question-answer pair data.

[0109] In one embodiment, the device further comprises a model training module, which is used to:

[0110] Acquire interactive data of the call scenario, perform data cleaning and dialogue splitting on the interactive data, and construct the first basic question-answer pair data;

[0111] Using a natural language processing model, extracting summaries of answers in the first basic question-answer pair data to generate information points of the answers in the first basic question-answer pair data;

[0112] The answers and information points in the first basic question and answer data are used to perform model training to obtain a business card information point generation model.

[0113] In one implementation, the number of the target information point is at least one, the semantic matching degree between the target information point and the target demand point is greater than a set similarity threshold, and the semantic matching degree between the target information point and the target demand point is calculated by calling a semantic matching algorithm.

[0114] In one embodiment, the reply text is pre-adjusted by the other user based on a target answer template for the business card information point, and the target answer template is retrieved from a relationship map based on the business card information point, wherein the relationship map records the correspondence between the information point and the reply template.

[0115] In one embodiment, the device further comprises a relationship map construction module, which is used to:

[0116] Acquire call interaction text data, perform data cleaning and dialogue splitting on the call interaction text data, and obtain the second basic question-answer pair data;

[0117] Using natural language processing technology to extract demand points in the questions of the second basic question-answer pair data, and extract information points in the answers of the second basic question-answer pair data, so as to form triple data using the demand points of the second basic question-answer pair data and the information points of the second basic question-answer pair data;

[0118] The triple data is disambiguated, and the disambiguated triple data is imported into a graph database to construct the relationship graph.

[0119] It is worth noting that the working principle of the call business card interaction device provided in the above embodiment can refer to the working process of the call business card interaction method provided in any of the above embodiments, which will not be described in detail here.

[0120] Compared with the prior art, the call business card interaction device provided in the embodiment of the present invention receives the business card content and at least one business card information point of the other user when a call occurs between the local user and the other user; wherein the business card information point is generated according to the business card content of the other user; in response to the operation of the local user selecting the business card information point, the reply text preset by the other user corresponding to the selected business card information point is obtained and displayed to the local user, thereby achieving the purpose of capturing user feedback and responding in a timely manner, being able to present the information required by the user more efficiently, and improving the information carrying capacity of the business card.

[0121] See also Figure 4 The embodiment of the present invention further provides a call card interaction device, including a processor 31, a memory 32, and a computer program stored in the memory 32 and configured to be executed by the processor 31. When the processor 31 executes the computer program, the steps in the above-mentioned call card interaction method embodiment are implemented, for example Figure 1 S11 to S12 in the above description; or, the processor 31 implements the functions of the modules in the above-mentioned device embodiments when executing the computer program.

[0122] Exemplarily, the computer program may be divided into one or more modules, which are stored in the memory 32 and executed by the processor 31 to complete the present invention. The one or more modules may be a series of computer program instruction segments that can complete specific functions, and the instruction segments are used to describe the execution process of the computer program in the call business card interaction device. For example, the computer program may be divided into multiple modules, and the specific working process of each module may refer to the working process of the call business card interaction device described in the above embodiment, which will not be repeated here.

[0123] The call card interaction device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The call card interaction device may include, but is not limited to, a processor 31 and a memory 32. Those skilled in the art may understand that the call card interaction device may also include an input and output device, a network access device, a bus, etc.

[0124] The processor 31 may be a central processing unit (CPU), or other general processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general processor may be a microprocessor or any conventional processor, etc. The processor 31 is the control center of the call business card interaction device, and uses various interfaces and lines to connect various parts of the entire call business card interaction device.

[0125] The memory 32 can be used to store the computer program and / or module. The processor 31 realizes various functions of the call business card interaction device by running or executing the computer program and / or module stored in the memory 32 and calling the data stored in the memory 32. The memory 32 can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required for a function (such as an image playback function, etc.); the data storage area can store data created according to the use of the mobile phone, etc. In addition, the memory 32 can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Med i aCard, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0126] Wherein, if the module integrated in the call business card interaction device is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor 31, the steps of the above-mentioned various method embodiments can be implemented. Wherein, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file or some intermediate form. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium.

[0127] An embodiment of the present invention further provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the call business card interaction method as described in any of the above embodiments.

[0128] The above is a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements and modifications without departing from the principle of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A call business card interaction method, characterized in that: include: When a call occurs between the user of the local party and the user of the other party, receiving the business card content and at least one business card information point of the user of the other party; wherein the business card information point is generated according to the business card content of the user of the other party; In response to the operation of the local user selecting the business card information point, a reply text preset by the other party user corresponding to the selected business card information point is obtained and displayed to the local user.

2. The call business card interaction method according to claim 1, characterized in that: The business card information point is specifically generated in the following manner: Extracting characteristic information of the business card content as a target demand point; Taking the target demand point as a search condition, finding the target information point from a pre-constructed relationship map between information points and demand points; wherein the relationship map is constructed based on the demand points extracted from the questions of the pre-acquired second basic question-answer pair data and the information points extracted from the answers of the second basic question-answer pair data; The target information point and the business card content are input into a pre-trained business card information point generation model to obtain the business card information point; wherein the business card information point generation model is trained based on the answers and information points in the pre-acquired first basic question-answer pair data.

3. The call business card interaction method according to claim 2, characterized in that: The business card information point generation model is specifically trained in the following way: Acquire interactive data of the call scenario, perform data cleaning and dialogue splitting on the interactive data, and construct the first basic question-answer pair data; Using a natural language processing model, extracting summaries of answers in the first basic question-answer pair data to generate information points of the answers in the first basic question-answer pair data; The answers and information points in the first basic question and answer data are used to perform model training to obtain a business card information point generation model.

4. The call business card interaction method according to claim 2, characterized in that: The number of the target information point is at least one, the semantic matching degree between the target information point and the target demand point is greater than a set similarity threshold, and the semantic matching degree between the target information point and the target demand point is calculated by calling a semantic matching algorithm.

5. The call business card interaction method according to claim 1, characterized in that: The reply text is pre-adjusted by the other user according to the target answer template for the business card information point, and the target answer template is retrieved from the relationship map according to the business card information point, wherein the relationship map records the correspondence between the information point and the reply template.

6. The call business card interaction method according to claim 2, characterized in that: The relationship map is constructed in the following way: Acquire call interaction text data, perform data cleaning and dialogue splitting on the call interaction text data, and obtain the second basic question-answer pair data; Using natural language processing technology to extract demand points in the questions of the second basic question-answer pair data, and extract information points in the answers of the second basic question-answer pair data, so as to form triple data using the demand points of the second basic question-answer pair data and the information points of the second basic question-answer pair data; The triple data is disambiguated, and the disambiguated triple data is imported into a graph database to construct the relationship graph.

7. A call business card interaction device, characterized in that: include: An information receiving module, used for receiving the business card content and at least one business card information point of the other party user when a call occurs between the user of the local party and the other party user; wherein the business card information point is generated according to the business card content of the other party user; The text acquisition module is used to respond to the operation of the local user selecting the business card information point, obtain the reply text preset by the other user corresponding to the selected business card information point and display it to the local user.

8. A call business card interaction device, characterized in that: The method comprises a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor implements the call business card interaction method as claimed in any one of claims 1 to 6 when executing the computer program.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program, wherein when the computer program is executed, the device where the computer-readable storage medium is located is controlled to execute the call business card interaction method according to any one of claims 1 to 6.

10. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instruction is executed by a processor, the call business card interaction method as described in any one of claims 1 to 6 is implemented.

Images