CN117473070B - Multi-channel application method of intelligent robot, intelligent robot and storage medium - Google Patents
Multi-channel application method of intelligent robot, intelligent robot and storage medium Download PDFInfo
- Publication number
- CN117473070B CN117473070B CN202311812870.4A CN202311812870A CN117473070B CN 117473070 B CN117473070 B CN 117473070B CN 202311812870 A CN202311812870 A CN 202311812870A CN 117473070 B CN117473070 B CN 117473070B
- Authority
- CN
- China
- Prior art keywords
- knowledge base
- intelligent robot
- user
- access
- channel
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 51
- 230000004044 response Effects 0.000 claims abstract description 57
- 238000012544 monitoring process Methods 0.000 claims description 26
- 238000004458 analytical method Methods 0.000 claims description 16
- 238000004364 calculation method Methods 0.000 claims description 16
- 238000007781 pre-processing Methods 0.000 claims description 5
- 238000012216 screening Methods 0.000 claims description 5
- 238000004140 cleaning Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 abstract description 7
- 238000004422 calculation algorithm Methods 0.000 description 6
- 238000004891 communication Methods 0.000 description 6
- 238000007621 cluster analysis Methods 0.000 description 4
- 238000007405 data analysis Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 230000006399 behavior Effects 0.000 description 3
- 230000008901 benefit Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000000605 extraction Methods 0.000 description 3
- 230000003993 interaction Effects 0.000 description 3
- 238000010801 machine learning Methods 0.000 description 3
- 238000007726 management method Methods 0.000 description 3
- 230000001360 synchronised effect Effects 0.000 description 3
- 238000012937 correction Methods 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 241000238558 Eucarida Species 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000013145 classification model Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000012905 input function Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000000153 supplemental effect Effects 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000000844 transformation Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/332—Query formulation
- G06F16/3329—Natural language query formulation or dialogue systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3346—Query execution using probabilistic model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/205—Parsing
- G06F40/216—Parsing using statistical methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/02—Knowledge representation; Symbolic representation
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Mathematical Physics (AREA)
- Probability & Statistics with Applications (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The application discloses a multichannel application method of an intelligent robot, the intelligent robot and a computer readable storage medium, and belongs to the field of electric digital data processing. The method comprises the following steps: receiving and analyzing an access request of a user, and acquiring problem description information and user access information; determining an access channel of the user according to the user access information; and generating response content corresponding to the problem description information based on the target branch knowledge base corresponding to the access channel, and outputting the response content. According to the intelligent robot, the intelligent robot is applied through multiple channels, the answer content of the relevant knowledge base is provided based on different access channels, the requirements of users are met, and the application breadth and flexibility of the intelligent robot and the answer efficiency of the question-answer service are improved.
Description
Technical Field
The present application relates to the field of electronic digital data processing technology, and in particular, to a multi-channel application method for an intelligent robot, and a computer readable storage medium.
Background
An intelligent robot is a robot system with the capabilities of sensing, understanding, reasoning, decision making and the like by combining artificial intelligence with robot technology. The method can understand the problems of the user and provide corresponding knowledge question-answering service by utilizing the techniques of natural language processing, knowledge graph, machine learning and the like.
In the related intelligent robot application, the intelligent robot is focused on the design and development of specific application scenes or fields to provide knowledge content collection and self-help knowledge solution, only one answer can be given for different channels, the application breadth and flexibility of the intelligent robot are limited, so that the intelligent robot cannot meet diversified requirements, and the response efficiency of the intelligent robot question-answer service is reduced by a single answer.
The foregoing is merely provided to facilitate an understanding of the principles of the present application and is not admitted to be prior art.
Disclosure of Invention
The main purpose of the present application is to provide a multi-channel application method of an intelligent robot, an intelligent robot and a computer readable storage medium, which aim to solve the problems that the intelligent robot focuses on knowledge collection and solution of a specific scene or field, limits the application breadth and flexibility thereof, leads to a single answer and reduces the response efficiency of a question-answer service.
In order to achieve the above object, the present application provides a multi-channel application method of an intelligent robot, the multi-channel application method of the intelligent robot includes:
receiving and analyzing an access request of a user, and acquiring problem description information and user access information;
determining an access channel of the user according to the user access information;
and generating response content corresponding to the problem description information based on the target branch knowledge base corresponding to the access channel, and outputting the response content.
Optionally, before the step of generating the response content corresponding to the problem description information and outputting the response content based on the target branch knowledge base corresponding to the access channel, the method further includes:
acquiring a history problem description corresponding to the user and history response content corresponding to the history problem description;
determining an intention label according to the problem description information based on the historical problem description and the historical response content;
and determining the target branch knowledge base according to the intention labels and the access channels.
Optionally, after the step of determining the intention label according to the problem description information based on the historical problem description and the historical response content, the method includes:
when the intention label is an artificial service, the problem description information, the history problem description and the history response content are transmitted to an artificial service end in an associated manner;
otherwise, executing the step of determining the target branch knowledge base according to the intention labels and the access channels.
Optionally, the step of determining the target branch knowledge base according to the intention label and the access channel comprises:
performing similarity matching on the intention labels and the common problem solution sets configured by the main knowledge base to obtain a similarity calculation result;
according to the similarity calculation result, taking the maximum value in the calculation result as an intention index, and taking the problem corresponding to the maximum value in the calculation result as a matching problem;
comparing the intention index with a preset similarity threshold;
if the intention index is lower than the similarity threshold, the problem description information, the historical problem description and the historical response content are transmitted to a manual server in an associated mode;
and if the intention index is higher than the similarity threshold, determining a target branch knowledge base according to the matching problem and the access channel.
Optionally, the step of generating the response content corresponding to the problem description information based on the target branch knowledge base corresponding to the access channel and outputting the response content includes:
based on the matching problem, searching in the target branch knowledge base according to the intention label, and outputting the response content;
if the searching in the target branch knowledge base fails, searching is carried out in the main knowledge base according to the intention labels based on the matching problem, and the response content is output.
Optionally, before receiving and analyzing the access request of the user and obtaining the problem description information and the user access information, the method further includes:
creating a branch knowledge base corresponding to the access channel based on the main knowledge base;
screening target knowledge corresponding to the access channel from the main knowledge base, and synchronizing the target knowledge to the branch knowledge base;
optimizing knowledge content of the branch knowledge base based on the access channel;
and monitoring the branch knowledge base, and updating the branch knowledge base according to the monitoring result.
Optionally, the specific step of optimizing the knowledge content of the branch knowledge base based on the access channel includes:
text cleaning and preprocessing are carried out on the knowledge content, and standardized data meeting the requirements of the channel are obtained;
identifying repeated records of the standardized data, and merging the data according to the requirements of the channels;
and expanding the combined data based on the access channel.
Optionally, the monitoring the branch knowledge base, and the specific step of updating the branch knowledge base according to the monitoring result includes:
collecting monitoring data of the branch knowledge base, and performing monitoring index calculation according to the monitoring data;
comparing and analyzing the monitoring index with a preset threshold value, and correspondingly updating the branch knowledge base according to an analysis result;
and synchronizing the updated contents of the branch knowledge base into the main knowledge base.
In addition, in order to achieve the above object, the present application also provides an intelligent robot including: the system comprises a memory, a processor and a multi-channel application program of the intelligent robot stored on the memory and capable of running on the processor, wherein the multi-channel application program of the intelligent robot is configured to realize the steps of the multi-channel application method of the intelligent robot.
In addition, in order to achieve the above object, the present application further provides a computer readable storage medium having stored thereon a multi-channel application program of an intelligent robot, which when executed by a processor, implements the steps of the multi-channel application method of an intelligent robot as described above.
According to the method and the device, the branch knowledge bases of multiple channels are provided, the access channels are determined according to the user access information, the corresponding branch knowledge bases can be determined for different channels, and more accurate and targeted response content is provided. The intelligent robot can flexibly adapt to different application scenes and channels, and the application range of the intelligent robot is widened. Meanwhile, the user experience is improved, the user is helped to find the solution or the required information more quickly, and the response efficiency of the intelligent robot question-answering service is further improved.
Drawings
FIG. 1 is a schematic flow chart of a first embodiment of a multi-channel application method of an intelligent robot of the present application;
FIG. 2 is a schematic flow chart of a second embodiment of a multi-channel application method of the intelligent robot of the present application;
FIG. 3 is a schematic flow chart of a third embodiment of a multi-channel application method of the intelligent robot of the present application;
FIG. 4 is a schematic structural diagram of a smart robot of a hardware operating environment according to an embodiment of the present application;
the realization, functional characteristics and advantages of the present application will be further described with reference to the embodiments, referring to the attached drawings.
Detailed Description
It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the present application.
The personalized response is provided by acquiring and outputting answers to the questions from the corresponding branch knowledge base according to the questions and access channels of the user. By determining the access channels of the users, a suitable knowledge base can be provided for different platforms or modes, and more accurate and customized answers can be provided. The satisfaction degree and experience of the users are improved, the users are helped to obtain the required information or solution more quickly, and the response efficiency of the question-answering service of the intelligent robot is also improved.
In order to better understand the above technical solution, exemplary embodiments of the present application will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present application are shown in the drawings, it should be understood that the present application may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
First embodiment
In this embodiment, the video service integration method proposed in the present application is further explained based on the flowchart shown in fig. 1.
Referring to fig. 1, the multi-channel application method of the intelligent robot includes:
step S100: receiving and analyzing an access request of a user, and acquiring problem description information and user access information;
in this embodiment, the intelligent robot receives the user's access request through a network interface or other suitable communication means, for example, the user inputs a question on a web page, sends a request through a mobile application, interacts with a voice assistant, and so on.
For example, the user may input a question or a demand in an input box of the web page, and after submitting a transmission request, the intelligent robot receives and parses the user's input by listening to a request interface on the web page. Also, the user may use a mobile application associated with the intelligent robot to raise a question through an input box or a voice input function on the application interface. The mobile application sends the user input questions to the intelligent robot via a network connection. Similarly, the intelligent robot may be integrated with a voice Assistant, such as Siri, google Assistant, amazon Alexa, etc. The user may speak with the voice assistant via voice and communicate the question to the intelligent robot. The voice assistant converts the voice of the user into a text form and transmits the text form to the intelligent robot for analysis through a network.
It should be noted that, the access request of the user may be, but not limited to, text format, voice format, video format, picture format, or data format. The intelligent robot can perform corresponding processing and analysis according to the requests in different formats so as to extract the problem description information and the user access information.
After receiving the access request of the user, the intelligent robot analyzes the access request.
For example, if the input content of the user is in text format, the text content may be parsed; if the user inputs a voice format, voice needs to be converted into text by using voice recognition technology; if the user inputs a picture or video, the user can parse the picture or video using computer vision techniques to extract relevant text information.
By parsing the user input content, the problem description information can be obtained. Corresponding user access information, such as request header, user agent, etc., can be obtained by parsing the access request.
Step S200: determining an access channel of the user according to the user access information;
in this embodiment, channels for accessing the intelligent robot mainly include a web channel, a mobile application channel, a real-time chat platform channel, a voice assistant channel, and the like. Different access channels may exhibit different user habits, needs and characteristics. The method can analyze according to the relevant bytes in the access information of the user or further analyze and interpret the relevant bytes of the access information of the user, and finally obtain the access channel of the user.
As an alternative implementation manner for determining the access channel, according to a User Agent (User-Agent) header in the User access information, and using a User-Agent parsing library, the client information used by the access User, such as related information of an operating system, a browser or an application program, contained in the User access information is analyzed, and by identifying different User-Agent values, the access channel of the User can be determined.
Alternatively, the access channel may also be determined by analyzing request header information, URL (Uniform Resource Locator ) parameters of the request or routing information, using a third party analysis tool, or the like.
As another alternative embodiment of determining the access channel, in determining the access channel, a data analysis tool of a third party, such as Google analysis (Google analysis), adobe analysis (Adobe analysis), etc., may be used. The third party data analysis tool may be integrated into a robotic system first, custom parameters set, defining parameters for identifying user access channels. Different parameters may be defined to represent different channels as desired. For example, the web page channel may set the custom parameter to "utm _source" with a value of an identifier corresponding to the source of the web page, such as direct access, search engine, social media, etc.; in the mobile application channel, the user-defined parameter of "app_source" may be used, and the value may be an application store channel, an in-application promotion or push notification, or the like. Secondly, at key points where the user interacts with the intelligent robot, event tracking is configured to capture user behavior and transformations. Corresponding events are defined according to different channels and interaction modes, such as submitting forms, clicking buttons, selecting menus and the like. And finally, analyzing and exploring the data through the report and the index which are provided by the third-party data analysis tool and are used for analyzing the access information of the user. And deducing the channel of the user accessing the intelligent robot by accessing information such as sources, reference websites, user behaviors and the like.
In this embodiment, the third party data analysis tool provides rich functionality and metrics that can help determine information about access sources, reference websites, user behavior, etc., and thus infer the user's access channel.
Step S300: and generating response content corresponding to the problem description information based on the target branch knowledge base corresponding to the access channel, and outputting the response content.
In this embodiment, different channels have corresponding branch knowledge bases, and the intelligent robot can provide corresponding response content for the user aiming at the access channel, so as to meet diversified user requirements. According to different access channels, the problem description information is searched in the corresponding target branch knowledge base, and response content meeting the needs of the user is generated through corresponding search rules or algorithms.
As an alternative embodiment of generating the response content, a corresponding template library is created for each access channel when generating the response content. And according to the problems or requirements set by the user, selecting the best matched template by matching the problem description information with the templates in the template library. And filling the corresponding knowledge content in the target knowledge base into the template to generate final response content. The accurate and reasonable response content is ensured, and the display requirements of all access channels are met.
In the technical scheme disclosed in the embodiment, the access channel of the user is determined by analyzing the user access request, and the response content of the problem description information is generated based on the corresponding target branch knowledge base and output. According to the scheme, customized response content is provided according to different access channels and user problem descriptions, user experience is enhanced, and problem solving efficiency is improved.
Second embodiment
Based on the same inventive concept, the present solution also provides a second embodiment. In this embodiment, a multi-channel application method of the intelligent robot of the present application is further explained based on the flow chart shown in fig. 2.
Referring to fig. 2, before the step of generating the response content corresponding to the problem description information and outputting the response content based on the target branch knowledge base corresponding to the access channel, the method further includes:
step S210: acquiring a history problem description corresponding to the user and history response content corresponding to the history problem description;
step S220: determining an intention label according to the problem description information based on the historical problem description and the historical response content;
in this embodiment, the intelligent robot may determine the intent tag of the user according to the problem description information by acquiring the history data of the user performing the question-answering service and combining the history data.
As an alternative embodiment to acquiring historical data, the intelligent robot uses context-aware techniques in acquiring the historical data. The context information of the current interaction is captured by observing and recording the conversation history, user operations, environmental information, etc. This may include previous user questions, answers to the robot, specific instructions, user geographic location, etc.
Optionally, the history data corresponding to the user may also be obtained by one or more of retrieving a database record, parsing a log file, querying a session state, calling a corresponding external system or a cloud service interface to obtain a corresponding history record, and the like.
It should be noted that, the method for acquiring the historical data by the intelligent robot generally depends on the specific implementation of the intelligent robot and the storage mode and storage location of the historical data.
As an alternative embodiment of determining the user intent tag, the problem description information is semantically identified when determining the intent tag. Firstly, preprocessing the problem description information, including removing special characters, punctuation coincidence and stop words, performing case-to-case conversion and the like. Feature extraction is then performed, and semantic relationships between words are captured by some feature extraction methods, such as a bag of words model, TF-IDF (term frequency-inverse text frequency) vector, or word embedding technique, and the features are converted into semantic representations. The corresponding semantic representations are then weighted according to the historical data, according to the similarity of the context, the temporal distance, or other relevance metric. Finally, a semantic recognition method of a traditional machine learning algorithm or a deep learning-based method, such as a random forest algorithm or a convolutional neural network, is used for predicting and deducing semantic representation of the problem description information through an intention classification model to generate an intention label.
Optionally, the intention label can be determined by one or more of keyword matching, a preset rule engine or cluster analysis and the like.
As another alternative embodiment for determining the user intent tag, the problem description information is subjected to cluster analysis and semantic recognition at the same time when the intent tag is determined. First, the problem description information is preprocessed and features are extracted. And secondly, clustering the problem description characteristic information into different groups by using a clustering analysis method, such as a K-means (K-means) clustering algorithm, a hierarchical clustering algorithm and the like, and carrying out characteristic classification through semantic recognition. And finally, comprehensively judging the final intention label by combining the contexts.
In the present embodiment, by preprocessing and feature extraction of the problem description information and then applying it to cluster analysis and semantic recognition, the user intention label can be effectively determined. The cluster analysis helps cluster the problem description feature information into different groups so that similar problem descriptions fall into the same category. And the semantic recognition further performs feature classification on the problem description, and predicts the intention category of each problem cluster by using a machine learning or deep learning method.
Further, when the intention label is an artificial service, the problem description information, the history problem description and the history response content are transmitted to an artificial service end in an associated manner; otherwise, the subsequent steps are performed.
In this embodiment, when the intention of the user is the manual service, the question description of the user and the history data of the user are sent to the manual customer service together, and the corresponding question-answering service is continuously provided by switching the manual customer service. The manual customer service can know the demands of the user according to the historical data, and further provides response content which meets the requirements better for the user.
Step S230: and determining the target branch knowledge base according to the intention labels and the access channels.
In this embodiment, the target branch knowledge base is determined according to the intention labels and the access channels in combination with the available knowledge base resources. This may be a pre-built knowledge base containing answers and solutions that are preferentially provided for a particular intent and access channel. Or a dynamically generated knowledge base, providing corresponding answers and support in real-time or near real-time according to specific intents and access channels.
Further, as an alternative implementation manner for determining the target branch knowledge base, when determining the target branch knowledge base, similarity matching needs to be performed on the intention labels and the common problem solution sets configured by the main knowledge base, so as to obtain a similarity calculation result. And secondly, taking the maximum value in the calculation result as an intention index according to the similarity calculation result, and taking the problem corresponding to the maximum value in the calculation result as a matching problem. Then comparing the intention index with a preset similarity threshold; if the intention index is lower than the similarity threshold, the problem description information, the history problem description and the history response content of the user are transmitted to the manual server in a related manner; and if the intention index is higher than the similarity threshold, determining a target branch knowledge base according to the matching problem and the access channel.
In this embodiment, the degree of matching between the user's intention and the main knowledge base is determined by calculating a similarity index, and whether to use the target branch knowledge base is determined according to a similarity threshold.
Further, after determining the target branch knowledge base, as an alternative implementation manner for generating the response content, searching in the target branch knowledge base according to the intention label based on the matching problem, and outputting the response content; if the searching in the target branch knowledge base fails, searching is carried out in the main knowledge base according to the intention labels based on the matching problem, and response content is output.
In this embodiment, a two-stage search method is proposed, in which search is performed in a target branch knowledge base, and response contents are generated using knowledge specific to the branch. If the search fails, the user intention is searched in the main knowledge base to ensure that relevant content about the user intention can be provided.
In the technical scheme disclosed by the embodiment, the intention labels of the users are obtained by combining the historical data of the user question-answering service, and the response modes are selected according to the intention labels, so that the demands and the backgrounds of the users are better understood, more personalized response content with high correlation is provided, and the user satisfaction degree and the response efficiency of the question-answering service are improved.
Third embodiment
Based on the same inventive concept, the present solution also provides a third embodiment. In this embodiment, a multi-channel application method of the intelligent robot of the present application is further explained based on the flowchart shown in fig. 3.
Referring to fig. 3, before receiving and analyzing the access request of the user and obtaining the problem description information and the user access information, the method further includes:
step S010: creating a branch knowledge base corresponding to the access channel based on the main knowledge base;
in the present embodiment, in order to enable a user to more conveniently acquire information and a solution related to the use of an access channel thereof, a corresponding branch knowledge base is created in a main knowledge base according to the access channel.
As an alternative embodiment of creating the branch knowledge base, the creation of the branches is performed by the Git tool while creating the branch knowledge base. Access channels are determined, and based on the subject knowledge base, a new branch is created to accommodate the access channels using the Git commands of the 'Git branch'.
In this embodiment, creating a branch knowledge base using the Git tool may provide version control, parallel development, customization, conflict resolution, and traceability benefits. This allows for more flexible and reliable management and maintenance of the knowledge base while also helping to accommodate different access channel requirements.
Alternatively, creation of subspaces or sub-sites may also be performed using a specialized knowledge management system to effect creation of branch databases.
Step S020: screening target knowledge corresponding to the access channel from the main knowledge base, and synchronizing the target knowledge to the branch knowledge base;
in this embodiment, the corresponding target knowledge needs to be synchronized to the branch knowledge base according to the channel, so as to implement a targeted branch database related to the channel. First, it is necessary to determine target knowledge, which may be a specific topic, a common question, an operation guide, etc. related to an access channel. And secondly, screening the knowledge content in the main knowledge base, and screening the content related to the target knowledge in a classification, label or keyword mode. And finally, synchronizing the target content into the branch knowledge base according to the knowledge base management tool.
Further, after synchronizing the target knowledge to the branch knowledge base, the relevant content is synchronized periodically according to the update of the main knowledge base. Ensuring that the content in the branch knowledge base remains synchronized with the subject knowledge base in order to provide up-to-date, accurate knowledge services.
Step S030: optimizing knowledge content of the branch knowledge base based on the access channel;
in this embodiment, in order to effectively meet the user requirements of a specific channel, a more accurate and targeted knowledge service is provided. By associating knowledge content with channels and optimizing the content, better user experience can be provided and users can be helped to find the information they need more easily.
As an alternative implementation scheme of data optimization, when data content optimization of a branch database is carried out, firstly, text cleaning and preprocessing are carried out on knowledge content in a knowledge base to obtain standardized data meeting channel requirements. Secondly, identifying repeated records of standardized data, and merging the data according to the requirements of channels; and finally, expanding the combined data based on the access channel.
In this embodiment, since different access channels have different characteristics, functions and constraints, standardized processing of data contents is required. Merging duplicate records can ensure consistency of the data. Repeated recordings may contain the same information but with subtle differences such as misspellings, different data formats, or misentries. By merging the records, these differences can be eliminated and the uniformity of the data can be ensured. By expanding the data, the practicability and application value of the data can be improved. For example, adding information such as relevant links, citations, case analysis, etc., can provide deeper learning and research resources, increase the practicality and applicability of the data, etc.
Step S040: and monitoring the branch knowledge base, and updating the branch knowledge base according to the monitoring result.
In this embodiment, in order to ensure accuracy and timeliness of data, the branch knowledge base needs to be monitored in real time, monitoring indexes are collected, and data is selectively updated according to different monitoring results.
As an alternative implementation of data updating, when the data of the branch database is updated, monitoring data of the branch database is firstly collected, and monitoring index calculation is performed according to the monitoring data. And comparing and analyzing the monitoring index with a preset threshold value, and correspondingly updating the analysis knowledge base according to the analysis result. And finally, synchronizing the updated content of the branch knowledge base into the main knowledge base.
In this embodiment, a monitoring system may be set up to collect relevant data of the branch knowledge base, such as access volume, error rate, user feedback, etc. And applying a proper algorithm and an index calculation formula according to the collected monitoring data to obtain a specific monitoring index. For example, metrics such as average access to the knowledge base, percentage of error rate, content update frequency, etc. may be calculated.
And comparing the calculated monitoring index with a preset threshold value for analysis. The preset threshold may be set according to the service requirement and the actual situation, for example, when the error rate exceeds a certain threshold (e.g. 5%), or the access amount drops by more than a certain percentage (e.g. 10%), further analysis and processing are required.
And correspondingly updating the branch knowledge base according to the comparison and analysis result. For example, if the error rate exceeds a threshold, the error information needs to be checked and corrected; if the access volume is reduced, it may be necessary to optimize the content or promotion policy. The updates may include correction data, supplemental information, correction errors, and the like.
And finally, synchronizing the updated content of the branch knowledge base to the main knowledge base to ensure that the content in the main knowledge base is consistent with the branch knowledge base.
In the technical scheme disclosed by the embodiment, a branch knowledge base is established according to different channel requirements, knowledge content in the knowledge base is maintained, diversified channel requirements are met, the effectiveness and coverage range of knowledge propagation are improved, high-quality user experience is provided for users, and the response efficiency of the intelligent robot question-answering service is improved.
Referring to fig. 4, fig. 4 is a schematic structural diagram of an intelligent robot of a hardware running environment according to an embodiment of the present application.
As shown in fig. 4, the intelligent robot may include: a processor 1001, such as a central processing unit (Central Processing Unit, CPU), a communication bus 1002, a user interface 1003, a network interface 1004, a memory 1005. Wherein the communication bus 1002 is used to enable connected communication between these components. The user interface 1003 may include a Display, an input unit such as a Keyboard (Keyboard), and the optional user interface 1003 may further include a standard wired interface, a wireless interface. The network interface 1004 may optionally include a standard wired interface, a WIreless interface (e.g., a WIreless-FIdelity (WI-FI) interface). The Memory 1005 may be a high-speed random access Memory (Random Access Memory, RAM) Memory or a stable nonvolatile Memory (NVM), such as a disk Memory. The memory 1005 may also optionally be a storage device separate from the processor 1001 described above.
Those skilled in the art will appreciate that the configuration shown in fig. 4 is not limiting of the intelligent robot and may include more or fewer components than shown, or may combine certain components, or may be arranged in a different arrangement of components.
As shown in fig. 4, an operating system, a data storage module, a network communication module, a user interface module, and a multi-channel application of the intelligent robot may be included in the memory 1005 as one type of storage medium.
In the intelligent robot shown in fig. 4, the network interface 1004 is mainly used for data communication with other devices; the user interface 1003 is mainly used for data interaction with a user; the processor 1001 and the memory 1005 in the intelligent robot can be arranged in the intelligent robot, and the intelligent robot calls the multi-channel application program of the intelligent robot stored in the memory 1005 through the processor 1001 and executes the multi-channel application method of the intelligent robot provided by the embodiment of the application.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or system. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or system that comprises the element.
The foregoing embodiment numbers of the present application are merely for describing, and do not represent advantages or disadvantages of the embodiments.
From the above description of embodiments, it will be clear to a person skilled in the art that the above embodiment method may be implemented by means of software plus a necessary general hardware platform, but may of course also be implemented by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (such as ROM/RAM, magnetic disk, optical disk) as described above, including several instructions for causing a terminal device (which may be a mobile phone, a computer, a server, or a network device, etc.) to perform the method described in the embodiments of the present application.
The foregoing description is only of the preferred embodiments of the present application, and is not intended to limit the scope of the claims, and all equivalent structures or equivalent processes using the descriptions and drawings of the present application, or direct or indirect application in other related technical fields are included in the scope of the claims of the present application.
Claims (6)
1. The multi-channel application method of the intelligent robot is characterized by comprising the following steps of:
creating a branch knowledge base corresponding to the access channel based on the main knowledge base;
screening target knowledge corresponding to the access channel from the main knowledge base, and synchronizing the target knowledge to the branch knowledge base;
optimizing knowledge content of the branch knowledge base based on the access channel;
monitoring the branch knowledge base, and updating the branch knowledge base according to a monitoring result, wherein the method comprises the following steps: collecting monitoring data of the branch knowledge base, and performing monitoring index calculation according to the monitoring data; comparing and analyzing the monitoring index with a preset threshold value, and correspondingly updating the branch knowledge base according to an analysis result; synchronizing the updated content of the branch knowledge base to the main knowledge base;
receiving and analyzing an access request of a user, and acquiring problem description information and user access information;
determining an access channel of the user according to the user access information;
acquiring a history problem description corresponding to the user and history response content corresponding to the history problem description;
determining an intention label according to the problem description information based on the historical problem description and the historical response content;
determining a target branch knowledge base according to the intention labels and the access channels, wherein the target branch knowledge base comprises the following steps: performing similarity matching on the intention labels and the common problem solution sets configured by the main knowledge base to obtain a similarity calculation result; according to the similarity calculation result, taking the maximum value in the calculation result as an intention index, and taking the problem corresponding to the maximum value in the calculation result as a matching problem; comparing the intention index with a preset similarity threshold; if the intention index is lower than the similarity threshold, the problem description information, the historical problem description and the historical response content are transmitted to a manual server in an associated mode; if the intention index is higher than the similarity threshold, determining a target branch knowledge base according to the matching problem and the access channel;
and generating response content corresponding to the problem description information based on the target branch knowledge base corresponding to the access channel, and outputting the response content.
2. The multi-channel application method of an intelligent robot according to claim 1, wherein after the step of determining an intention label from the question description information based on the history question description and the history response content, it comprises:
when the intention label is an artificial service, the problem description information, the history problem description and the history response content are transmitted to an artificial service end in an associated manner;
otherwise, executing the step of determining the target branch knowledge base according to the intention labels and the access channels.
3. The multi-channel application method of an intelligent robot according to claim 1, wherein the generating response contents corresponding to the problem description information based on the target branch knowledge base corresponding to the access channel, and outputting the response contents comprises:
based on the matching problem, searching in the target branch knowledge base according to the intention label, and outputting the response content;
if the searching in the target branch knowledge base fails, searching is carried out in the main knowledge base according to the intention labels based on the matching problem, and the response content is output.
4. The multi-channel application method of an intelligent robot according to claim 1, wherein the step of optimizing knowledge contents of the branched knowledge base based on the access channel comprises:
text cleaning and preprocessing are carried out on the knowledge content, and standardized data meeting the requirements of the channel are obtained;
identifying repeated records of the standardized data, and merging the data according to the requirements of the channels;
and expanding the combined data based on the access channel.
5. An intelligent robot, characterized in that the intelligent robot comprises: memory, a processor and a multi-channel application of the intelligent robot stored on the memory and executable on the processor, the multi-channel application of the intelligent robot being configured to implement the steps of the multi-channel application method of the intelligent robot as claimed in any one of claims 1 to 4.
6. A computer readable storage medium, wherein a multi-channel application of a smart robot is stored on the computer readable storage medium, which when executed by a processor, implements the steps of the multi-channel application method of a smart robot as claimed in any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311812870.4A CN117473070B (en) | 2023-12-27 | 2023-12-27 | Multi-channel application method of intelligent robot, intelligent robot and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311812870.4A CN117473070B (en) | 2023-12-27 | 2023-12-27 | Multi-channel application method of intelligent robot, intelligent robot and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117473070A CN117473070A (en) | 2024-01-30 |
| CN117473070B true CN117473070B (en) | 2024-04-02 |
Family
ID=89624106
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311812870.4A Active CN117473070B (en) | 2023-12-27 | 2023-12-27 | Multi-channel application method of intelligent robot, intelligent robot and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117473070B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106649874A (en) * | 2017-01-04 | 2017-05-10 | 上海智臻智能网络科技股份有限公司 | Data interaction method, device and system |
| CN108052615A (en) * | 2017-12-15 | 2018-05-18 | 泰康保险集团股份有限公司 | Processing method, device, medium and the electronic equipment of access request |
| CN112836521A (en) * | 2020-12-23 | 2021-05-25 | 平安普惠企业管理有限公司 | Question-answer matching method and device, computer equipment and storage medium |
| CN113011175A (en) * | 2021-03-02 | 2021-06-22 | 中国工商银行股份有限公司 | Semantic identification method and system based on dual channel feature matching |
| WO2021164286A1 (en) * | 2020-02-21 | 2021-08-26 | 平安科技（深圳）有限公司 | User intention recognition method, apparatus and device, and computer-readable storage medium |
-
2023
- 2023-12-27 CN CN202311812870.4A patent/CN117473070B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106649874A (en) * | 2017-01-04 | 2017-05-10 | 上海智臻智能网络科技股份有限公司 | Data interaction method, device and system |
| CN108052615A (en) * | 2017-12-15 | 2018-05-18 | 泰康保险集团股份有限公司 | Processing method, device, medium and the electronic equipment of access request |
| WO2021164286A1 (en) * | 2020-02-21 | 2021-08-26 | 平安科技（深圳）有限公司 | User intention recognition method, apparatus and device, and computer-readable storage medium |
| CN112836521A (en) * | 2020-12-23 | 2021-05-25 | 平安普惠企业管理有限公司 | Question-answer matching method and device, computer equipment and storage medium |
| CN113011175A (en) * | 2021-03-02 | 2021-06-22 | 中国工商银行股份有限公司 | Semantic identification method and system based on dual channel feature matching |
Non-Patent Citations (1)
| Title |
|---|
| 互联网多渠道接入场景下的在线智能客服方案研究;董宇;;信息与电脑(理论版);20160623(12);第76-77页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117473070A (en) | 2024-01-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111444428B (en) | Information recommendation method and device based on artificial intelligence, electronic equipment and storage medium | |
| CN110147437B (en) | Knowledge graph-based searching method and device | |
| US20200301954A1 (en) | Reply information obtaining method and apparatus | |
| CN108874992B (en) | Public opinion analysis method, system, computer equipment and storage medium | |
| KR102288249B1 (en) | Information processing method, terminal, and computer storage medium | |
| CN107590224B (en) | Big data based user preference analysis method and device | |
| CN110909182B (en) | Multimedia resource searching method, device, computer equipment and storage medium | |
| CN106815356B (en) | Precision target user message method for pushing and system based on semantic analysis | |
| CN109710739B (en) | Information processing method and device and storage medium | |
| CN112035638B (en) | Information processing method, device, storage medium and equipment | |
| CN111324713A (en) | Automatic replying method and device for conversation, storage medium and computer equipment | |
| KR20180072049A (en) | User customized system and method for searching and matching expert | |
| KR101450453B1 (en) | Method and apparatus for recommending contents | |
| CN114564595A (en) | Knowledge graph updating method and device and electronic equipment | |
| US11200264B2 (en) | Systems and methods for identifying dynamic types in voice queries | |
| CN111078855A (en) | Information processing method, information processing device, electronic equipment and storage medium | |
| CN110597951A (en) | Text parsing method and device, computer equipment and storage medium | |
| CN112671885B (en) | Information analysis method based on cloud computing and big data and digital financial service platform | |
| CN116956183A (en) | Multimedia resource recommendation method, model training method, device and storage medium | |
| CN117473070B (en) | Multi-channel application method of intelligent robot, intelligent robot and storage medium | |
| CN113961811B (en) | Event map-based conversation recommendation method, device, equipment and medium | |
| CN114298767A (en) | Live broadcast platform information pushing method and device, equipment, medium and product thereof | |
| CN114328947A (en) | Knowledge graph-based question and answer method and device | |
| CN112395490B (en) | Method and device for generating information | |
| CN112286916A (en) | Data processing method, device, equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |