Intelligent robot response system based on user portrait
By building user portraits and extracting features of user internal and external behavior data, personalized interaction of intelligent robot response systems is achieved, and the problem of inability to understand user intentions in the existing technology is solved, flexible and intelligent response capabilities are provided, and the Quality of service.
Patent Information
- Application Number
- CN202411073568.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-06
- Publication Date
- 2024-11-29
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The existing intelligent robot response system cannot accurately understand user intentions, lacks personalized responses, and cannot adapt to complex and changeable user needs and scenarios.
An intelligent robot response system based on user portrait is adopted to obtain user internal and external behavior data, perform feature extraction and jointing, build user portraits, and realize personalized response interaction based on user portraits.
It realizes more flexible and intelligent response, can adapt to the needs of different user types and scenarios, and continuously improves the accuracy and personalization of responses through continuous learning and iteration, providing a better service experience.
Smart Images

Figure CN119048623A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of intelligent robot answering, and in particular to an intelligent robot answering system based on user portrait. Background Art
[0002] With the rapid development of artificial intelligence technology, intelligent robots have been widely used in various fields, especially in customer service, online education, medical consultation and other industries. In these applications, intelligent robots need to have the ability to interact naturally with humans to provide efficient and accurate services. Therefore, it is particularly important to build an intelligent robot response system that can understand user intentions and provide personalized responses.
[0003] User portraits are a comprehensive description of user characteristics, preferences, behaviors, and other information. By collecting and analyzing user data, portraits of different user types can be constructed. The intelligent robot response system based on user portraits can use this portrait information to better understand user intentions and provide more personalized response content.
[0004] In the prior art, traditional intelligent robot answering systems usually use rule-based or template-based matching methods to generate answering content through preset rules or templates. However, this method has limitations and cannot adapt to complex and changing user needs and scenarios.
[0005] Therefore, there is an urgent need for an intelligent robot response system based on user portraits to solve the above problems. Summary of the invention
[0006] The purpose of the present invention is to provide an intelligent robot response system based on user portrait: to solve the technical problem in the existing solution that when the intelligent robot responds to the user, it cannot accurately understand the user's intention due to the lack of user portrait portrayal.
[0007] The purpose of the present invention can be achieved through the following technical solutions:
[0008] An intelligent robot answering system based on user portrait, the system includes a user behavior data acquisition module, a user portrait construction module and an answering service module;
[0009] A user behavior data acquisition module, which acquires user behavior data, wherein the user behavior data includes user internal behavior data and user external behavior data, wherein the user internal behavior data includes the user's individual behavior information based on specific commodities, and the user external behavior data includes the common behavior information with other users;
[0010] The user portrait construction module extracts features from the user's internal behavior data to obtain the first portrait feature C dy, extract features from the user's external behavior data to obtain the second portrait feature C de , and the first image feature C dy With the second image feature C de Combine them to obtain joint portrait features, and build user portraits based on the joint portrait features;
[0011] The response service module implements response interaction with users based on user portraits.
[0012] Furthermore, before extracting features from the user's internal behavior data, the user's internal behavior data is preprocessed, and the preprocessing includes the following processes:
[0013] Generate a personalized behavior information sequence α={A0, A1...A k}, A i is a personalized behavior information package, i ranges from 0 to k, and each personalized behavior information package contains one of the three description statements: the description statement of the product that the user intends to purchase, the description statement of the purchased product, and the description statement of the specific product that the user has browsed. i Indicated as A i = {B0, B1...B n}, B j To describe the statement, the value of j ranges from 0 to n, and both k and n are positive integers.
[0014] Furthermore, before extracting features from the user's external behavior data, the user's external behavior data is preprocessed, and the preprocessing includes the following processes:
[0015] Based on natural language processing technology, a knowledge graph is constructed for each user's external behavior data to determine whether the user meets the association condition. If so, the users are associated to generate an association graph. The association condition is to determine whether the user has the same external behavior data as other users. If so, the association condition is met. If not, the association condition is not met. A graph convolutional network is used to calculate the user's feature representation in the association graph:
[0016]
[0017] Among them, S is the adjacency matrix in the association graph, S = E + V, E is the unit matrix, V is the matrix of all user nodes in the association graph, is the degree matrix of S, σ is the activation matrix, H l+1 is the feature representation of the l+1th layer, W l is the weight matrix of the lth layer.
[0018] Furthermore, feature extraction is performed on the user's internal behavior data to obtain the first portrait feature, which specifically includes the following process:
[0019] To B j Expand or split into fixed length m: If the sentence length does not reach m, fill it with a preset filler; if the sentence length exceeds m, truncate it directly;
[0020] Extract B j The words in the , each word is mapped to a low-dimensional vector, and the word vector representation P of each word is obtained j ∈R v×d , d is the word mapping dimension, v is the vocabulary size;
[0021] Based on the word vector representation P of each word j Use the convolution kernel f along B j Slide the window and represent the word vector in the window P j Perform convolution operation to generate sentence local features C from the sliding window of the sentence j :
[0022] C j =g(P j *(P j :P j+x-1 )+t)
[0023] Where, 1≤j≤m-x+1, x is the window size, * is the convolution operation, t is the bias term, and g is the nonlinear activation function, preferably the hyperbolic tangent function;
[0024] Usually, q convolution kernels are used to obtain different local features of a sentence. The q local features C generated by the q convolution kernels are j Splice them together to generate the first image feature C dy .
[0025] Furthermore, feature extraction is performed on the user's external behavior data to obtain the second portrait feature, including: l+1 Recorded as the second image feature C de ;
[0026] Furthermore, the first image feature C dy With the second image feature C de The joint process is to obtain joint portrait features and build a user portrait based on the joint portrait features. The specific process includes the following:
[0027] The first image feature C dy With the second image feature C de Concatenate into a new feature representation C new , C new ={C dy:C de}, C new Denoted as joint image features;
[0028] The joint portrait features are input into the SoftMax classifier to obtain the probability value of the user in the current attribute category, and the attribute category with the largest probability value is taken as the analysis result. According to the analysis result, the user is labeled accordingly. The label reflects the user's characteristics and preferences in various dimensions.
[0029] Furthermore, implementing response interaction with users based on user portraits includes the following processes:
[0030] Clarify interaction needs: Based on user portraits, clarify the needs and goals of interacting with users. The needs and goals of interaction include providing personalized recommendations, answering user questions, and providing product usage guides.
[0031] Design interaction strategies: Design corresponding interaction strategies based on the characteristics and preferences of user portraits, including interaction methods, interaction content, and interaction frequency;
[0032] Integrated interactive technology: using modern communication technology, natural language processing (NLP) technology, machine learning technology, etc. to achieve intelligent interaction between users and systems;
[0033] Response interaction implementation:
[0034] Receiving user input: receiving user input through a user interface or voice interface, the user input includes text and voice;
[0035] Understanding user intent: Analyze user input based on natural language processing technology to understand user intent and needs;
[0036] Matching user portraits: According to user intent, match the corresponding user portrait to determine the user type or user group to which the user belongs;
[0037] Generate response content: Generate response content that meets user needs based on the matching user profile and interaction strategy;
[0038] Providing response feedback: Feedback the generated response content to the user through the user interface or voice interface to achieve interaction with the user.
[0039] Compared with the existing solutions, the present invention achieves the following beneficial effects:
[0040] The present invention obtains user behavior data, wherein the user behavior data includes user internal behavior data and user external behavior data, extracts features from the user internal behavior data, and obtains a first portrait feature C dy , extract features from the user's external behavior data to obtain the second portrait feature Cde , and the first image feature C dy With the second image feature C de The system combines the features of the joint portrait to obtain the joint portrait features, and builds the user portrait based on the joint portrait features, and implements the response interaction with the user based on the user portrait. By understanding the user's intentions and needs, the system can generate response content that meets the user's expectations. Compared with traditional rule-based or template-based matching methods, this method is more flexible and intelligent, and can adapt to the needs of different user types and scenarios.
[0041] In addition, the intelligent robot response system based on user portraits can also continuously optimize and improve the response strategy based on user feedback and interaction data. Through continuous learning and iteration, the system can continuously improve the accuracy and personalization of responses and provide a better service experience. BRIEF DESCRIPTION OF THE DRAWINGS
[0042] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.
[0043] Figure 1 is a system block diagram of an intelligent robot answering system based on user portrait according to an embodiment of the present invention;
[0044] Figure 2 It is a system workflow diagram of an intelligent robot answering system based on user portrait in an embodiment of the present invention. DETAILED DESCRIPTION
[0045] 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.
[0046] In addition, the described features, structures or characteristics may be combined in one or more example embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the example embodiments of the present disclosure. However, those skilled in the art will appreciate that the technical solutions of the present disclosure may be practiced while omitting one or more of the specific details, or other methods, components, steps, etc. may be adopted. In other cases, well-known structures, methods, implementations or operations are not shown or described in detail to avoid obscuring various aspects of the present disclosure.
[0047] This embodiment provides an intelligent robot response system based on user portraits. Figure 1 is a system block diagram of an intelligent robot answering system based on user portrait according to an embodiment of the present invention. Figure 1 As shown, the system includes a user behavior data acquisition module, a user portrait construction module and a response service module;
[0048] A user behavior data acquisition module, which acquires user behavior data, wherein the user behavior data includes user internal behavior data and user external behavior data, wherein the user internal behavior data includes the user's individual behavior information based on specific commodities, and the user external behavior data includes the common behavior information with other users;
[0049] The user portrait construction module extracts features from the user's internal behavior data to obtain the first portrait feature C dy , extract features from the user's external behavior data to obtain the second portrait feature C de , and the first image feature C dy With the second image feature C de Combine them to obtain joint portrait features, and build user portraits based on the joint portrait features;
[0050] The response service module implements response interaction with users based on user portraits.
[0051] In summary, the present invention obtains user behavior data, wherein the user behavior data includes user internal behavior data and user external behavior data, extracts features from the user internal behavior data, and obtains the first portrait feature C dy , extract features from the user's external behavior data to obtain the second portrait feature C de , and the first image feature C dy With the second image feature C de The system uses user portrait information to understand user intentions and provide personalized response content. The system is flexible, intelligent, and optimizable, and can meet the needs of different user types and scenarios, providing strong support for the application and development of intelligent robots.
[0052] It is worth noting that the user behavior data acquisition module, the user portrait construction module and the response service module can all realize two-to-two communication and complete data transmission.
[0053] In some embodiments, before extracting features from the user's internal behavior data, the user's internal behavior data is preprocessed, and the preprocessing includes the following process:
[0054] Generate a personalized behavior information sequence α={A0, A1...A k}, A i is a personalized behavior information package, i ranges from 0 to k, and each personalized behavior information package contains one of the three description statements: the description statement of the product that the user intends to purchase, the description statement of the purchased product, and the description statement of the specific product that the user has browsed. i Indicated as A i = {B0, B1...B n}, B j To describe the statement, the value of j ranges from 0 to n, and both k and n are positive integers.
[0055] In some embodiments, before extracting features from the user's external behavior data, the user's external behavior data is preprocessed, and the preprocessing includes the following process:
[0056] Based on natural language processing technology, a knowledge graph is constructed for each user's external behavior data to determine whether the user meets the association conditions. If so, the users are associated to generate an association graph. The association graph refers to a technical method that uses a graph model to describe knowledge and model the relationship between all things in the world. Specifically, the association graph consists of nodes and edges. The association condition is to determine whether the user has the same external behavior data with other users. If so, the association condition is met. If not, the association condition is not met. A graph convolutional network is used to calculate the user's feature representation in the association graph:
[0057]
[0058] Among them, S is the adjacency matrix in the association graph, S = E + V, E is the unit matrix, V is the matrix of all user nodes in the association graph, is the degree matrix of S, σ is the activation matrix, H l+1 is the feature representation of the l+1th layer, W l is the weight matrix of the lth layer.
[0059] It is worth noting that the construction of the knowledge graph of each user's external behavior data based on natural language processing technology includes:
[0060] Data preprocessing:
[0061] The collected raw data is cleaned, deduplicated, formatted and other preprocessing operations are performed to improve data quality. The preprocessing process may include the application of natural language processing technologies such as text denoising, word segmentation, and part-of-speech tagging.
[0062] Information extraction:
[0063] Using natural language processing technology, key information such as entities and relationships are extracted from user behavior data. Entities usually include names of people, places, names of organizations, time, date, etc., while relationships describe the interactions and connections between entities.
[0064] Entity extraction:
[0065] Through methods such as part-of-speech tagging, syntactic analysis and entity recognition, entities are automatically extracted from text and classified.
[0066] Entities are the basic elements in knowledge graphs, so entity extraction is an important basis for building knowledge graphs.
[0067] Relation extraction:
[0068] Identify relationships between entities in text and represent them as edges in a graph.
[0069] Natural language processing technology can automatically extract the relationship between entities in the text by analyzing sentence structure, grammatical rules, and contextual information.
[0070] Knowledge Fusion:
[0071] The extracted entities and relationships are integrated to form a complete knowledge graph.
[0072] In the process of knowledge fusion, problems such as entity disambiguation and coreference resolution need to be solved to ensure the accuracy and consistency of the knowledge graph.
[0073] Knowledge storage:
[0074] The constructed knowledge graph is stored for subsequent query and application.
[0075] In some embodiments, extracting features from the user's internal behavior data to obtain the first portrait feature specifically includes the following process:
[0076] To B j Expand or split into fixed length m: If the sentence length does not reach m, fill it with a preset filler; if the sentence length exceeds m, truncate it directly;
[0077] Extract B j The words in the , each word is mapped to a low-dimensional vector, and the word vector representation P of each word is obtained j ∈R v×d , d is the word mapping dimension, v is the vocabulary size;
[0078] Based on the word vector representation P of each word j Use the convolution kernel f along B j Slide the window and represent the word vector in the window Pj Perform convolution operation to generate sentence local features C from the sliding window of the sentence j :
[0079] C j =g(P j *(P j :P j+x-1 )+t)
[0080] Where, 1≤j≤m-x+1, x is the window size, * is the convolution operation, t is the bias term, and g is the nonlinear activation function, preferably the hyperbolic tangent function;
[0081] Usually, q convolution kernels are used to obtain different local features of a sentence. The q local features C generated by the q convolution kernels are j Splice them together to generate the first image feature C dy .
[0082] In some embodiments, extracting features from the user's external behavior data to obtain the second portrait feature includes: extracting H l+1 Recorded as the second image feature C de ;
[0083] In some embodiments, the first image feature C dy With the second image feature C de The joint process is to obtain joint portrait features and build a user portrait based on the joint portrait features. The specific process includes the following:
[0084] The first image feature C dy With the second image feature C de Concatenate into a new feature representation C new , C new ={C dy :C de}, C new Denoted as joint image features;
[0085] The joint portrait features are input into the SoftMax classifier to obtain the probability value of the user in the current attribute category, and the attribute category with the largest probability value is taken as the analysis result. According to the analysis result, the user is labeled accordingly. The label reflects the user's characteristics and preferences in various dimensions.
[0086] In some embodiments, implementing response interaction with the user based on the user portrait includes the following process:
[0087] Clarify interaction needs: Based on user portraits, clarify the needs and goals of interacting with users. The needs and goals of interaction include providing personalized recommendations, answering user questions, and providing product usage guides.
[0088] Design interaction strategies: Design corresponding interaction strategies based on the characteristics and preferences of user portraits, including interaction methods, interaction content, and interaction frequency;
[0089] Integrated interactive technology: using modern communication technology, natural language processing technology, machine learning technology, etc. to achieve intelligent interaction between users and systems;
[0090] Response interaction implementation:
[0091] Receiving user input: receiving user input through a user interface or voice interface, the user input includes text and voice;
[0092] Understanding user intent: Analyze user input based on natural language processing technology to understand user intent and needs;
[0093] Matching user portraits: According to user intent, match the corresponding user portrait to determine the user type or user group to which the user belongs;
[0094] Generate response content: Generate response content that meets user needs based on the matching user profile and interaction strategy;
[0095] Providing response feedback: Feedback the generated response content to the user through the user interface or voice interface to achieve interaction with the user.
[0096] Figure 2 is a system workflow diagram of an intelligent robot answering system based on user portrait according to an embodiment of the present invention. Figure 2 As shown, the system workflow includes the following steps:
[0097] Step S201: Obtaining user behavior data, wherein the user behavior data includes user internal behavior data and user external behavior data, wherein the user internal behavior data includes the user's individual behavior information based on a specific commodity, and the user external behavior data includes common behavior information with other users;
[0098] Step S202: extracting features from the user's internal behavior data to obtain a first portrait feature, extracting features from the user's external behavior data to obtain a second portrait feature, combining the first portrait feature with the second portrait feature to obtain a combined portrait feature, and constructing a user portrait based on the combined portrait feature;
[0099] Step S203: Implement response interaction with the user based on the user portrait.
[0100] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.
[0101] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.
[0102] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.
[0103] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only some logical function divisions. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.
[0104] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0105] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.
Claims
1. An intelligent robot answering system based on user portrait, characterized in that: The system includes a user behavior data acquisition module, a user portrait construction module and a response service module; A user behavior data acquisition module, which acquires user behavior data, wherein the user behavior data includes user internal behavior data and user external behavior data, wherein the user internal behavior data includes the user's individual behavior information based on specific commodities, and the user external behavior data includes the common behavior information with other users; The user portrait construction module extracts features from the user's internal behavior data to obtain the first portrait feature C dy , extract features from the user's external behavior data to obtain the second portrait feature C de , and the first image feature C dy With the second image feature C de Combine them to obtain joint portrait features, and build user portraits based on the joint portrait features; The response service module implements response interaction with users based on user portraits.
2. According to claim 1, the intelligent robot answering system based on user portrait is characterized in that: Before extracting features from user internal behavior data, preprocess the user internal behavior data. The process includes: Generate a personalized behavior information sequence α={A0, A1...A k }, A i is a personalized behavior information package, i ranges from 0 to k, and each personalized behavior information package contains one of the three description statements: the description statement of the product that the user intends to purchase, the description statement of the purchased product, and the description statement of the specific product that the user has browsed. i Indicated as A i = {B0, B1...B n }, B j To describe the statement, the value of j ranges from 0 to n, and both k and n are positive integers.
3. The intelligent robot answering system based on user portrait according to claim 1 is characterized in that: Before extracting features from user external behavior data, preprocess the user external behavior data. The process includes: Based on natural language processing technology, a knowledge graph is constructed for each user's external behavior data to determine whether the user meets the association condition. If so, the users are associated to generate an association graph. The association condition is to determine whether the user has the same external behavior data as other users. If so, the association condition is met. If not, the association condition is not met. A graph convolutional network is used to calculate the user's feature representation in the association graph: Among them, S is the adjacency matrix in the association graph, S = E + V, E is the unit matrix, V is the matrix of all user nodes in the association graph, is the degree matrix of S, σ is the activation matrix, H l+1 is the feature representation of the l+1th layer, W l is the weight matrix of the lth layer.
4. The intelligent robot answering system based on user portrait according to claim 2 is characterized in that: Extracting features from the user's internal behavior data to obtain the first portrait feature specifically includes the following process: To B j Expand or split into fixed length m: If the sentence length does not reach m, fill it with a preset filler; if the sentence length exceeds m, truncate it directly; Extract B j The words in the , each word is mapped to a low-dimensional vector, and the word vector representation P of each word is obtained j ∈R v×d , d is the word mapping dimension, v is the vocabulary size; Based on the word vector representation P of each word j Use the convolution kernel f along B j Slide the window and represent the word vector in the window P j Perform convolution operation to generate sentence local features C from the sliding window of the sentence j : C j =g(P j *(P j :P j+x-1 )+t) Where, 1≤j≤m-x+1, x is the window size, * is the convolution operation, t is the bias term, and g is the nonlinear activation function, preferably the hyperbolic tangent function; Usually, q convolution kernels are used to obtain different local features of a sentence. The q local features C generated by the q convolution kernels are j Splice them together to generate the first image feature C dy .
5. The intelligent robot answering system based on user portrait according to claim 3 is characterized in that: Extracting features from the user's external behavior data to obtain the second portrait features includes: l+1 Recorded as the second image feature C de .
6. The intelligent robot answering system based on user portrait according to claim 1 is characterized in that: The first image feature C dy With the second image feature C de The joint process is to obtain joint portrait features and build a user portrait based on the joint portrait features. The specific process includes the following: The first image feature C dy With the second image feature C de Concatenate into a new feature representation C new , C new ={C dy :C de }, C new It is recorded as joint image feature; The joint portrait features are input into the SoftMax classifier to obtain the probability value of the user in the current attribute category, and the attribute category with the largest probability value is taken as the analysis result. According to the analysis result, the user is labeled accordingly. The label reflects the user's characteristics and preferences in various dimensions.
7. The intelligent robot answering system based on user portrait according to claim 1 is characterized in that: Implementing response interaction with users based on user portraits includes the following processes: Clarify interaction needs: Based on user portraits, clarify the needs and goals of interacting with users. The needs and goals of interaction include providing personalized recommendations, answering user questions, and providing product usage guides. Design interaction strategies: Design corresponding interaction strategies based on the characteristics and preferences of user portraits, including interaction methods, interaction content, and interaction frequency; Integrated interactive technology: using modern communication technology, natural language processing (NLP) technology, machine learning technology, etc. to achieve intelligent interaction between users and systems; Response interaction implementation: Receiving user input: receiving user input through a user interface or voice interface, the user input includes text and voice; Understanding user intent: Analyze user input based on natural language processing technology to understand user intent and needs; Matching user portraits: According to user intent, match the corresponding user portrait to determine the user type or user group to which the user belongs; Generate response content: Generate response content that meets user needs based on the matching user profile and interaction strategy; Providing response feedback: Feedback the generated response content to the user through the user interface or voice interface to achieve interaction with the user.
Citation Information
Patent Citations
Financial-service-oriented chat robot and system
CN108121824A
Early warning analysis method and system based on chronic disease medication big data
CN114283917A
User portrait system of e-commerce re-purchase behavior based on big data
CN114511378A
Power grid user portrait construction method and system
CN116596574A
Ai-based virtual interaction model generation method and apparatus, computer device and storage medium
WO2022095380A1
