Knowledge recommendation method and device, storage medium and electronic equipment
By using preset situational models and user behavior recognition models in the enterprise knowledge management system to filter and sort knowledge, the problem that corporate knowledge cannot efficiently serve business work is solved, personalized knowledge recommendation is achieved and knowledge utilization is improved.
Patent Information
- Application Number
- CN202411201586.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-29
- Publication Date
- 2025-02-07
AI Technical Summary
Enterprise knowledge cannot serve business work efficiently, resulting in low knowledge utilization.
Through a knowledge recommendation method, the initial knowledge feature data set is filtered and sorted using preset situational models and user behavior recognition models to generate a personalized knowledge queue, and the target knowledge data is filtered out according to the threshold of the number of knowledge.
It realizes personalized recommendation of task situation information and improves the efficiency of enterprise knowledge utilization.
Smart Images
Figure CN119397085A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of information recommendation, and in particular to a knowledge recommendation method, device, storage medium and electronic device. Background Art
[0002] In recent years, more and more R&D enterprises have carried out knowledge management work in combination with internal needs, and built knowledge management systems through information technology to achieve the collection, storage, management and utilization of enterprise knowledge resources, and promote enterprise knowledge inheritance and knowledge innovation. However, knowledge management systems are usually information islands, and the full life cycle activities of knowledge are separated from work task scenarios. Staff members are required to actively participate in knowledge management activities in the knowledge management system. Enterprise knowledge cannot efficiently serve business work, the value of knowledge is not fully utilized, and the utilization rate of enterprise knowledge is low. Summary of the invention
[0003] In view of this, the present invention provides a knowledge recommendation method, device, storage medium and electronic device, the main purpose of which is to solve the current problem that enterprise knowledge cannot efficiently serve business work and the utilization rate of enterprise knowledge is low.
[0004] To solve the above problems, this application provides a knowledge recommendation method, including:
[0005] Based on the pre-acquired initial situational feature data set, a preset situational model is used to filter the pre-acquired initial knowledge feature data set to obtain a first knowledge data set;
[0006] Based on the pre-acquired user feature data set, the first knowledge data set is filtered using a preset user behavior recognition model to obtain a second knowledge data set;
[0007] Based on the second knowledge data set, the target sorting model is used to sort each piece of knowledge data in the second knowledge data set to obtain a knowledge queue;
[0008] The knowledge data in the knowledge queue are screened based on the knowledge queue and a preset knowledge quantity threshold to obtain target knowledge data for knowledge recommendation service.
[0009] Optionally, before filtering the pre-acquired initial knowledge feature data set using a preset context model based on the pre-acquired initial context feature data set, the method further includes: acquiring the initial context feature data set, the initial knowledge feature data set, and the user feature data set, specifically including:
[0010] Determine a data collection category and each data collection dimension corresponding to the data collection category;
[0011] Perform data collection based on the data collection category and each of the data collection dimensions to obtain initial data corresponding to different data collection dimensions;
[0012] Performing data processing on each of the initial data corresponding to the data collection category to obtain a feature data set corresponding to the data collection category;
[0013] The data collection categories include knowledge data categories, situational data categories, and user data categories;
[0014] The feature data set includes an initial situation feature data set, an initial knowledge feature data set and a user feature data set.
[0015] Optionally, the performing data processing on each of the initial data corresponding to the data collection category to obtain a feature data set corresponding to the data collection category specifically includes:
[0016] Performing abnormal data processing on each of the initial data to obtain each first data set;
[0017] Using a first feature processing rule to process the structured data and the semi-structured data in each of the first data sets to obtain a first feature data set;
[0018] Processing the unstructured data in each of the first data sets using a second feature processing rule to obtain a second feature data set;
[0019] Based on the combination of the first feature data set and the second feature data set, a feature data set corresponding to the same data collection category is obtained.
[0020] Optionally, the filtering of the pre-acquired initial knowledge feature data set using a preset context model based on the pre-acquired initial context feature data set to obtain the first knowledge data set specifically includes:
[0021] Based on the initial situation feature data set, a preset situation model is used to perform feature recognition to obtain a situation feature vector;
[0022] Performing similarity calculation based on each initial knowledge feature vector in the initial knowledge feature data set and the context feature vector to obtain each first similarity;
[0023] The initial knowledge feature vectors are screened based on the first similarities and a first preset threshold to obtain the first knowledge data set.
[0024] Optionally, the filtering of the first knowledge data set by using a preset user behavior recognition model based on the pre-acquired user feature data set to obtain the second knowledge data set specifically includes:
[0025] Extracting knowledge features based on each user behavior feature data in the user feature data set to obtain a knowledge feature vector corresponding to each user behavior feature data;
[0026] Performing vector construction based on each of the knowledge feature vectors to obtain a user feature vector;
[0027] Performing similarity calculation based on each initial knowledge feature vector in the first knowledge data set and the user feature vector to obtain each second similarity;
[0028] The initial knowledge feature vectors in the first knowledge data set are screened based on the second similarities and a second preset threshold to obtain the second knowledge data set.
[0029] Optionally, the step of sorting each piece of knowledge data in the second knowledge data set by using a target sorting model based on the second knowledge data set to obtain a knowledge queue specifically includes:
[0030] The initial knowledge feature vectors in the second knowledge data set are rearranged based on a predetermined knowledge evaluation rule of the target sorting model to generate the knowledge queue.
[0031] Optionally, the screening of each of the knowledge data in the knowledge queue based on the knowledge queue and a preset knowledge quantity threshold to obtain each target knowledge data for the knowledge recommendation service specifically includes:
[0032] Based on the preset knowledge quantity threshold, the knowledge data ranked at the top in the knowledge queue are collected to obtain target knowledge data for knowledge recommendation service.
[0033] To solve the above problems, the present application provides a knowledge recommendation device, comprising:
[0034] A first filtering module, configured to filter the pre-acquired initial knowledge feature data set using a preset context model based on the pre-acquired initial context feature data set to obtain a first knowledge data set;
[0035] A second filtering module, configured to filter the first knowledge data set using a preset user behavior recognition model based on a pre-acquired user feature data set to obtain a second knowledge data set;
[0036] A sorting module, configured to sort each piece of knowledge data in the second knowledge data set by adopting a target sorting model based on the second knowledge data set to obtain a knowledge queue;
[0037] The screening module is used to screen each of the knowledge data in the knowledge queue based on the knowledge queue and a preset knowledge quantity threshold to obtain each target knowledge data for the knowledge recommendation service.
[0038] In order to solve the above problem, the present application provides a storage medium, characterized in that the storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above-mentioned knowledge recommendation method are implemented.
[0039] To solve the above problem, the present application provides an electronic device, characterized in that it includes at least a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps of the above-mentioned knowledge recommendation method when executing the computer program in the memory.
[0040] Beneficial effects of the present application: The present application combines the full life cycle activities of knowledge with knowledge recommendation services, and uses preset situation models and preset user behavior models to gradually screen knowledge, thereby achieving personalized recommendation of task situation information and improving the utilization efficiency of enterprise knowledge.
[0041] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented according to the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS
[0042] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:
[0043] Figure 1 A schematic diagram of a process of a knowledge recommendation method provided in an embodiment of the present application is shown;
[0044] Figure 2 A schematic diagram showing a flow chart of a knowledge recommendation method provided by another embodiment of the present application is shown;
[0045] Figure 3 A structural block diagram of a knowledge recommendation device provided in another embodiment of the present application is shown. DETAILED DESCRIPTION
[0046] Various aspects and features of the present application are described herein with reference to the accompanying drawings.
[0047] It should be understood that various modifications may be made to the embodiments of the present application. Therefore, the above description should not be considered as limiting, but only as an example of an embodiment. Other modifications within the scope and spirit of the present application will occur to those skilled in the art.
[0048] The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate embodiments of the present application and, together with the general description of the present application given above and the detailed description of the embodiments given below, serve to explain the principles of the present application.
[0049] These and other characteristics of the present application will become apparent from the following description of a preferred form of embodiment given as a non-limiting example with reference to the accompanying drawings.
[0050] It should also be understood that although the present application has been described with reference to some specific examples, those skilled in the art will be able to readily implement many other equivalent forms of the present application.
[0051] The above and other aspects, features and advantages of the present application will become more apparent in view of the following detailed description when taken in conjunction with the accompanying drawings.
[0052] Specific embodiments of the present application are described hereinafter with reference to the accompanying drawings; however, it should be understood that the embodiments applied for are merely examples of the present application, which may be implemented in a variety of ways. Well-known and / or repeated functions and structures are not described in detail to avoid unnecessary or redundant details that obscure the present application. Therefore, the specific structural and functional details applied for herein are not intended to be limiting, but merely serve as a basis and representative basis for the claims to teach those skilled in the art to use the present application in a variety of ways with substantially any suitable detailed structure.
[0053] This specification may use the phrases "in one embodiment," "in another embodiment," "in yet another embodiment," or "in other embodiments," all of which may refer to one or more of the same or different embodiments according to the present application.
[0054] This application embodiment provides a knowledge recommendation method, such as Figure 1 As shown, including:
[0055] Step S101: filtering the pre-acquired initial knowledge feature data set using a preset context model based on the pre-acquired initial context feature data set to obtain a first knowledge data set;
[0056] In the specific implementation process of this step, based on the initial situational feature data set, a preset situational model is used to perform feature recognition to obtain a situational feature vector; the preset situational model includes various situational dimensions, each of which includes: task input, task output, task constraint, project and responsible person and other situational dimensions; each of the situational dimensions includes various sub-dimensions; wherein the task input situational dimension includes: input file, input model and input parameter and other sub-dimensions; the task output situational dimension includes: output file, output model and output parameter and other sub-dimensions; the task constraint situational dimension includes: tool constraint, process constraint, procedure constraint and standard constraint; the project situational dimension includes: project type and project stage and other sub-dimensions; the responsible person situational dimension includes: department and profession and other sub-dimensions. Based on each initial knowledge feature vector in the initial knowledge feature data set and the situational feature vector, similarity calculation is performed to obtain each first similarity; based on each first similarity and a first preset threshold, each initial knowledge feature vector is screened and processed to obtain the first knowledge data set.
[0057] Step S102: filtering the first knowledge data set using a preset user behavior recognition model based on a pre-acquired user feature data set to obtain a second knowledge data set;
[0058] In the specific implementation process of this step, knowledge features are extracted based on each user behavior feature data in the user feature data set to obtain a knowledge feature vector corresponding to each user behavior feature data; vector construction is performed based on each knowledge feature vector to obtain a user feature vector; similarity calculation is performed based on each initial knowledge feature vector in the first knowledge data set and the knowledge feature vector to obtain each second similarity; and each initial knowledge feature vector in the first knowledge data set is screened based on each second similarity and a second preset threshold to obtain the second knowledge data set.
[0059] Step S103: sorting each piece of knowledge data in the second knowledge data set using a target sorting model based on the second knowledge data set to obtain a knowledge queue;
[0060] In the specific implementation process of this step, the initial knowledge feature vectors in the second knowledge data set are rearranged based on the predetermined knowledge evaluation rules of the target sorting model to generate the knowledge queue. The target sorting model can be a task context sorting model, a knowledge popularity sorting model, a content relevance model, etc.
[0061] Step S104: Screening each of the knowledge data in the knowledge queue based on the knowledge queue and a preset knowledge quantity threshold to obtain each target knowledge data for knowledge recommendation service.
[0062] In the specific implementation process of this step, the knowledge data ranked first in the knowledge queue are collected based on the preset knowledge quantity threshold to obtain target knowledge data for knowledge recommendation service.
[0063] This application combines the full life cycle activities of knowledge with knowledge recommendation services, and uses preset situation models and preset user behavior models to gradually screen knowledge, thereby achieving personalized recommendation of task situation information and improving the utilization efficiency of enterprise knowledge.
[0064] Another embodiment of the present invention provides another knowledge recommendation method, such as Figure 2 As shown, including:
[0065] Step S201: obtaining an initial context feature data set, an initial knowledge feature data set, and a user feature data set;
[0066] During the specific implementation of this step, the data collection category and the data collection dimensions corresponding to the data collection category are determined; the data collection category includes a knowledge data category, a contextual data category and a user data category; the data collection dimensions corresponding to the knowledge data category include: attribute information, content information, and interactive data and other data collection dimensions; the data collection dimensions corresponding to the contextual data category include: task information, project information, and responsible person information and other data collection dimensions; the data collection dimensions corresponding to the user data category include: personal information, social information, and behavioral data and other data collection dimensions. Based on the data collection category and each of the data collection dimensions, data collection is performed to obtain initial data corresponding to different data collection dimensions; data processing is performed on each of the initial data corresponding to the data collection category to obtain a feature data set corresponding to the data collection category; specifically, abnormal data processing is performed on each of the initial data to obtain each first data set; first, each abnormal value is identified based on each of the initial data using a statistical analysis method or an outlier detection method; each of the abnormal values is deleted or replaced to obtain each first detection data; then, based on each of the first detection data, a preset threshold interval corresponding to each of the first detection data is used for filtering processing to obtain each second detection data; specifically, data outside the preset threshold interval is deleted to obtain each of the second detection data; finally, missing value supplement processing is performed based on each of the second detection data to obtain each of the first data sets. A regression analysis method can be used to predict the value of the missing value using a regression model; or a random forest method can be used to predict the missing value using a random forest algorithm to supplement the missing value in the second detection data to obtain each of the first data sets. The structured data and semi-structured data in each of the first data sets are processed using a first feature processing rule to obtain a first feature data set; specifically, the first feature processing rule includes: feature selection, feature processing and feature dimension reduction to process the structured data and semi-structured data in the first data set to obtain a first feature data set; the unstructured data in each of the first data sets are processed using a second feature processing rule to obtain a second feature data set; specifically, the second feature processing rule includes: one-hot encoding, bag-of-words model, TF-IDF, Word2Vec, knowledge graph embedding and other methods to process the unstructured data in each of the first data sets to obtain a second feature data set; based on the combination of the first feature data set and the second feature data set, a feature data set corresponding to the same data collection category is obtained. The feature data sets include an initial context feature data set, an initial knowledge feature data set and a user feature data set.
[0067] Step S202: performing feature recognition based on the initial situation feature data set using a preset situation model to obtain a situation feature vector;
[0068] In the specific implementation process of this step, the preset situation model includes various situation dimensions, and each of the situation dimensions includes: task input, task output, task constraint, project and responsible person; each of the situation dimensions includes various sub-dimensions; wherein the task input situation dimension includes: input file, input model and input parameter sub-dimensions; the task output situation dimension includes: output file, output model and output parameter sub-dimensions; the task constraint situation dimension includes: tool constraint, process constraint, procedure constraint and standard constraint; the project situation dimension includes: project type and project stage sub-dimensions; the responsible person situation dimension includes: department and profession sub-dimensions. Based on the initial situation feature data set, the preset situation model is used to perform feature recognition, and the sub-situation vectors corresponding to each of the sub-dimensions under each of the situation dimensions are obtained; vector construction is performed based on each of the sub-situation vectors to obtain the situation feature vector.
[0069] Step S203: performing similarity calculation based on each initial knowledge feature vector in the initial knowledge feature data set and the context feature vector to obtain each first similarity;
[0070] During the specific implementation of this step, the Pearson correlation coefficient method or the Euclidean distance method is used to calculate the similarity of each initial knowledge feature vector and the situational feature vector in the initial knowledge feature data set to obtain each first similarity, thereby laying the foundation for subsequent screening and processing of each initial knowledge feature vector based on each first similarity and a first preset threshold to obtain the first knowledge data set.
[0071] Step S204: screening each of the initial knowledge feature vectors based on each of the first similarities and a first preset threshold to obtain the first knowledge data set;
[0072] During the specific implementation of this step, a comparison is performed based on each of the first similarities and the first preset threshold to obtain a first comparison result; the initial knowledge feature vectors whose first comparison result is that the first similarity is less than the first preset threshold are deleted; and a knowledge data set consisting of initial knowledge feature vectors whose first comparison result is that the first similarity is greater than or equal to the first preset threshold is determined as the first knowledge data set, wherein the first preset threshold can be set according to actual needs.
[0073] Step S205: extracting knowledge features based on each user behavior feature data in the user feature data set to obtain a knowledge feature vector corresponding to each user behavior feature data;
[0074] In the specific implementation process of this step, the user behavior feature data includes: user reading behavior, liking behavior, commenting behavior and downloading behavior, etc. Knowledge features are extracted from each of the user behavior feature data to obtain a knowledge feature vector corresponding to each of the user behavior feature data.
[0075] Step S206: constructing a vector based on each of the knowledge feature vectors to obtain a user feature vector;
[0076] In the specific implementation process of this step, each of the knowledge feature vectors is calculated by using a weighted summation method to obtain the user feature vector.
[0077] Step S207: performing similarity calculation based on each initial knowledge feature vector in the first knowledge data set and the user feature vector to obtain each second similarity;
[0078] During the specific implementation of this step, the Pearson correlation coefficient method or the Euclidean distance method is used to calculate the similarity of each initial knowledge feature vector and the situational feature vector in the first knowledge data set to obtain each second similarity, thereby laying the foundation for the subsequent screening and processing of each initial knowledge feature vector in the first knowledge data set based on each second similarity and a second preset threshold to obtain the second knowledge data set.
[0079] Step S208: screening each of the initial knowledge feature vectors in the first knowledge data set based on each of the second similarities and a second preset threshold to obtain the second knowledge data set;
[0080] During the specific implementation of this step, a comparison is performed based on each of the second similarities and the second preset threshold to obtain a second comparison result; the initial knowledge feature vectors whose second comparison result is that the second similarity is less than the second preset threshold are deleted; and the knowledge data set composed of the initial knowledge feature vectors whose second comparison result is that the second similarity is greater than or equal to the second preset threshold is determined as the second knowledge data set, wherein the second preset threshold can be set according to actual needs.
[0081] Step S209: sorting each piece of knowledge data in the second knowledge data set using a target sorting model based on the second knowledge data set to obtain a knowledge queue;
[0082] In the specific implementation process of this step, the initial knowledge feature vectors in the second knowledge data set are rearranged based on the predetermined knowledge evaluation rules of the target sorting model to generate the knowledge queue. The target sorting model can be a task context sorting model, a knowledge popularity sorting model, a content relevance model, etc.
[0083] Step S210: Screening each of the knowledge data in the knowledge queue based on the knowledge queue and a preset knowledge quantity threshold to obtain each target knowledge data for knowledge recommendation service.
[0084] In the specific implementation process of this step, the knowledge data ranked first in the knowledge queue are collected based on the preset knowledge quantity threshold to obtain target knowledge data for knowledge recommendation service.
[0085] This application combines the full life cycle activities of knowledge with knowledge recommendation services, and uses preset situation models and preset user behavior models to gradually screen knowledge, thereby achieving personalized recommendation of task situation information and improving the utilization efficiency of enterprise knowledge.
[0086] Another embodiment of the present application provides a knowledge recommendation device, such as Figure 3 As shown, including:
[0087] A first filtering module 1, configured to filter the pre-acquired initial knowledge feature data set using a preset context model based on the pre-acquired initial context feature data set to obtain a first knowledge data set;
[0088] A second filtering module 2, configured to filter the first knowledge data set using a preset user behavior recognition model based on a pre-acquired user feature data set to obtain a second knowledge data set;
[0089] A sorting module 3, configured to sort each piece of knowledge data in the second knowledge data set by using a target sorting model based on the second knowledge data set to obtain a knowledge queue;
[0090] The screening module 4 is used to screen the knowledge data in the knowledge queue based on the knowledge queue and a preset knowledge quantity threshold to obtain target knowledge data for knowledge recommendation service.
[0091] In the specific implementation process, the device also includes: a data acquisition module, which is specifically used to: acquire an initial context feature data set, an initial knowledge feature data set and a user feature data set, specifically including: determining a data collection category and each data collection dimension corresponding to the data collection category; performing data collection based on the data collection category and each data collection dimension to obtain initial data corresponding to different data collection dimensions; performing data processing on each of the initial data corresponding to the data collection category to obtain a feature data set corresponding to the data collection category; wherein the data collection category includes a knowledge data category, a context data category and a user data category; the feature data set includes an initial context feature data set, an initial knowledge feature data set and a user feature data set.
[0092] During the specific implementation process, the data acquisition module is also used to: perform abnormal data processing on each of the initial data to obtain each first data set; use the first feature processing rule to perform data processing on the structured data and semi-structured data in each of the first data sets to obtain a first feature data set; use the second feature processing rule to perform data processing on the unstructured data in each of the first data sets to obtain a second feature data set; and combine the first feature data set and the second feature data set to obtain a feature data set corresponding to the same data collection category.
[0093] In the specific implementation process, the first filtering module 1 is specifically used to: perform feature recognition based on the initial situational feature data set using a preset situational model to obtain a situational feature vector; perform similarity calculation based on each initial knowledge feature vector in the initial knowledge feature data set and the situational feature vector to obtain each first similarity; and perform screening and processing on each initial knowledge feature vector based on each first similarity and a first preset threshold to obtain the first knowledge data set.
[0094] In the specific implementation process, the second filtering module 2 is specifically used to: extract knowledge features based on each user behavior feature data in the user feature data set to obtain a knowledge feature vector corresponding to each user behavior feature data; construct a vector based on each knowledge feature vector to obtain a user feature vector; perform similarity calculation processing based on each initial knowledge feature vector in the first knowledge data set and the user feature vector to obtain each second similarity; screen each initial knowledge feature vector in the first knowledge data set based on each second similarity and a second preset threshold to obtain the second knowledge data set.
[0095] In a specific implementation process, the sorting module 3 is specifically used to: rearrange the initial knowledge feature vectors in the second knowledge data set based on a predetermined knowledge evaluation rule of the target sorting model to generate the knowledge queue.
[0096] In the specific implementation process, the screening module 4 is specifically used to collect the knowledge data ranked first in the knowledge queue based on the preset knowledge quantity threshold to obtain the target knowledge data for knowledge recommendation service.
[0097] This application combines the full life cycle activities of knowledge with knowledge recommendation services, and uses preset situation models and preset user behavior models to gradually screen knowledge, thereby achieving personalized recommendation of task situation information and improving the utilization efficiency of enterprise knowledge.
[0098] Another embodiment of the present application provides a storage medium, wherein the storage medium stores a computer program, and when the computer program is executed by a processor, the following method steps are implemented:
[0099] Step 1: Based on the pre-acquired initial situational feature data set, a preset situational model is used to filter the pre-acquired initial knowledge feature data set to obtain a first knowledge data set;
[0100] Step 2: Based on the pre-acquired user feature data set, the first knowledge data set is filtered using a preset user behavior recognition model to obtain a second knowledge data set;
[0101] Step 3: sorting each knowledge data in the second knowledge data set using a target sorting model based on the second knowledge data set to obtain a knowledge queue;
[0102] Step 4: Screen the knowledge data in the knowledge queue based on the knowledge queue and a preset knowledge quantity threshold to obtain target knowledge data for knowledge recommendation service.
[0103] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing related hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct RAMbus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM), etc.
[0104] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.
[0105] The specific implementation process of the above method steps can refer to the embodiments of any of the above knowledge recommendation methods, and this embodiment will not be repeated here.
[0106] This application combines the full life cycle activities of knowledge with knowledge recommendation services, and uses preset situation models and preset user behavior models to gradually screen knowledge, thereby achieving personalized recommendation of task situation information and improving the utilization efficiency of enterprise knowledge.
[0107] Another embodiment of the present application provides an electronic device, which may be a server, and the electronic device includes a processor, a memory, a network interface, and a database connected via a system bus. The processor of the electronic device is used to provide computing and control capabilities. The memory of the electronic device includes a non-volatile and / or volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The network interface of the electronic device is used to communicate with an external client via a network connection. When the electronic device program is executed by the processor, it implements functions or steps on the server side of a knowledge recommendation method.
[0108] In one embodiment, an electronic device is provided, which may be a client. The electronic device includes a processor, a memory, a network interface, a display screen, and an input device connected via a system bus. The processor of the electronic device is used to provide computing and control capabilities. The memory of the electronic device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the electronic device is used to communicate with an external server via a network connection. When the electronic device program is executed by the processor, the functions or steps on the client side of a knowledge recommendation method are implemented.
[0109] Another embodiment of the present application provides an electronic device, comprising at least a memory and a processor, wherein the memory stores a computer program, and the processor implements the following method steps when executing the computer program in the memory:
[0110] Step 1: Based on the pre-acquired initial situational feature data set, a preset situational model is used to filter the pre-acquired initial knowledge feature data set to obtain a first knowledge data set;
[0111] Step 2: Based on the pre-acquired user feature data set, the first knowledge data set is filtered using a preset user behavior recognition model to obtain a second knowledge data set;
[0112] Step 3: sorting each knowledge data in the second knowledge data set using a target sorting model based on the second knowledge data set to obtain a knowledge queue;
[0113] Step 4: Screen the knowledge data in the knowledge queue based on the knowledge queue and a preset knowledge quantity threshold to obtain target knowledge data for knowledge recommendation service.
[0114] The specific implementation process of the above method steps can refer to the embodiments of any of the above knowledge recommendation methods, and this embodiment will not be repeated here.
[0115] This application combines the full life cycle activities of knowledge with knowledge recommendation services, and uses preset situation models and preset user behavior models to gradually screen knowledge, thereby achieving personalized recommendation of task situation information and improving the utilization efficiency of enterprise knowledge.
[0116] The above embodiments are only exemplary embodiments of the present application and are not intended to limit the present application. The protection scope of the present application is defined by the claims. Those skilled in the art may make various modifications or equivalent substitutions to the present application within the essence and protection scope of the present application, and such modifications or equivalent substitutions shall also be deemed to fall within the protection scope of the present application.
Claims
1. A knowledge recommendation method, characterized in that: include: Based on the pre-acquired initial situational feature data set, a preset situational model is used to filter the pre-acquired initial knowledge feature data set to obtain a first knowledge data set; Based on the pre-acquired user feature data set, the first knowledge data set is filtered using a preset user behavior recognition model to obtain a second knowledge data set; Based on the second knowledge data set, the target sorting model is used to sort each piece of knowledge data in the second knowledge data set to obtain a knowledge queue; The knowledge data in the knowledge queue are screened based on the knowledge queue and a preset knowledge quantity threshold to obtain target knowledge data for knowledge recommendation service.
2. The method according to claim 1, characterized in that Before filtering the pre-acquired initial knowledge feature data set using a preset context model based on the pre-acquired initial context feature data set, the method further includes: acquiring the initial context feature data set, the initial knowledge feature data set and the user feature data set, specifically including: Determine a data collection category and each data collection dimension corresponding to the data collection category; Perform data collection based on the data collection category and each of the data collection dimensions to obtain initial data corresponding to different data collection dimensions; Performing data processing on each of the initial data corresponding to the data collection category to obtain a feature data set corresponding to the data collection category; The data collection categories include knowledge data categories, situational data categories, and user data categories; The feature data set includes an initial context feature data set, an initial knowledge feature data set, and a user feature data set.
3. The method according to claim 2, characterized in that The performing data processing on each of the initial data corresponding to the data collection category to obtain a feature data set corresponding to the data collection category specifically includes: Performing abnormal data processing on each of the initial data to obtain each first data set; Using a first feature processing rule to process the structured data and the semi-structured data in each of the first data sets to obtain a first feature data set; Processing the unstructured data in each of the first data sets using a second feature processing rule to obtain a second feature data set; Based on the combination of the first feature data set and the second feature data set, a feature data set corresponding to the same data collection category is obtained.
4. The method according to claim 1, characterized in that The method of filtering the pre-acquired initial knowledge feature data set using a preset context model based on the pre-acquired initial context feature data set to obtain a first knowledge data set specifically includes: Based on the initial situation feature data set, a preset situation model is used to perform feature recognition to obtain a situation feature vector; Performing similarity calculation based on each initial knowledge feature vector in the initial knowledge feature data set and the context feature vector to obtain each first similarity; The initial knowledge feature vectors are screened based on the first similarities and a first preset threshold to obtain the first knowledge data set.
5. The method according to claim 1, characterized in that The filtering of the first knowledge data set by using a preset user behavior recognition model based on the pre-acquired user feature data set to obtain a second knowledge data set specifically includes: Extracting knowledge features based on each user behavior feature data in the user feature data set to obtain a knowledge feature vector corresponding to each user behavior feature data; Performing vector construction based on each of the knowledge feature vectors to obtain a user feature vector; Performing similarity calculation based on each initial knowledge feature vector in the first knowledge data set and the user feature vector to obtain each second similarity; The initial knowledge feature vectors in the first knowledge data set are screened based on the second similarities and a second preset threshold to obtain the second knowledge data set.
6. The method according to claim 1, characterized in that The step of sorting each piece of knowledge data in the second knowledge data set by using a target sorting model based on the second knowledge data set to obtain a knowledge queue specifically includes: The initial knowledge feature vectors in the second knowledge data set are rearranged based on a predetermined knowledge evaluation rule of the target sorting model to generate the knowledge queue.
7. The method according to claim 6, characterized in that The step of screening the knowledge data in the knowledge queue based on the knowledge queue and a preset knowledge quantity threshold to obtain target knowledge data for the knowledge recommendation service specifically includes: Based on the preset knowledge quantity threshold, the knowledge data ranked at the top in the knowledge queue are collected to obtain target knowledge data for knowledge recommendation service.
8. A knowledge recommendation device, characterized in that: include: A first filtering module, configured to filter the pre-acquired initial knowledge feature data set using a preset context model based on the pre-acquired initial context feature data set to obtain a first knowledge data set; A second filtering module, configured to filter the first knowledge data set using a preset user behavior recognition model based on a pre-acquired user feature data set to obtain a second knowledge data set; A sorting module, configured to sort each piece of knowledge data in the second knowledge data set by adopting a target sorting model based on the second knowledge data set to obtain a knowledge queue; The screening module is used to screen each of the knowledge data in the knowledge queue based on the knowledge queue and a preset knowledge quantity threshold to obtain each target knowledge data for the knowledge recommendation service.
9. A storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the knowledge recommendation method described in any one of claims 1 to 7 are implemented.
10. An electronic device, characterized in that: The method comprises at least a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the knowledge recommendation method according to any one of claims 1 to 7 when executing the computer program on the memory.