CN115905702B - Data recommendation method and system based on user demand analysis - Google Patents
Data recommendation method and system based on user demand analysis Download PDFInfo
- Publication number
- CN115905702B CN115905702B CN202211553047.1A CN202211553047A CN115905702B CN 115905702 B CN115905702 B CN 115905702B CN 202211553047 A CN202211553047 A CN 202211553047A CN 115905702 B CN115905702 B CN 115905702B
- Authority
- CN
- China
- Prior art keywords
- knowledge vector
- user demand
- interest element
- mining result
- variable
- 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
Classifications
-
- 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
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Abstract
According to the data recommendation method and the data recommendation system based on the user demand analysis, provided by the embodiment of the invention, through setting rule adjustment processing, the variables such as knowledge features generated by the AI machine learning model are free from missing and errors, and the accuracy of the obtained data pushing decision information can be ensured in the process of calling the AI machine learning model to conduct pushing decision analysis under different scenes, so that the flexibility of conducting the data pushing decision analysis based on the user demand is improved, the errors in conducting the data pushing decision analysis and feature translation under different scenes are reduced, and progressive data mining analysis can be realized based on the user demand knowledge vectors and linkage demand description fields, so that the data pushing decision information is accurately, completely and reasonably obtained.
Description
Technical Field
The invention relates to the technical field of data pushing, in particular to a data recommendation method and system based on user demand analysis.
Background
User demand analysis is popular in the big data age, and is widely applied to different business fields, such as front-edge fields of blockchain, cloud game, intelligent medical treatment, electronic commerce and the like, and the front-edge fields of the user demand analysis and mining are started, so that guidance can be provided for later service optimization or data pushing. Under the above conditions, user demand analysis is common by applying artificial intelligence, but the precision and flexibility of the traditional technology are difficult to guarantee when analyzing user demand and data push.
Disclosure of Invention
In order to improve the technical problems in the related art, the invention provides a data recommendation method and system based on user demand analysis.
In a first aspect, an embodiment of the present invention provides a data recommendation method based on user demand analysis, which is applied to a data recommendation processing system, where the method includes: mining a first user demand knowledge vector and a first linkage demand description field of a first digital user feedback record to be subjected to data recommendation analysis; obtaining a data pushing related variable by using the first user demand knowledge vector and the first linkage demand description field, wherein the data pushing related variable is a feature extraction result obtained by an AI machine learning model after setting rule adjustment processing, and the setting rule adjustment processing is used for enabling a generation result of the AI machine learning model to be an expert knowledge vector; and carrying out pushing decision analysis operation by using the data pushing related variable, the first linkage demand description field and the first user demand knowledge vector to obtain first data pushing decision information of the first digital user feedback record.
According to the data recommendation method based on user demand analysis, through setting rule adjustment processing, the variables such as knowledge features generated by the AI machine learning model are free from loss and errors, the accuracy of obtained data pushing decision information can be ensured in the process of calling the AI machine learning model to conduct pushing decision analysis under different scenes, the flexibility of conducting data pushing decision analysis based on user demands is improved, errors in conducting data pushing decision analysis and feature translation under different scenes are reduced, and progressive data mining analysis can be achieved based on user demand knowledge vectors and linkage demand description fields, so that the data pushing decision information is accurately, completely and reasonably obtained.
In some independent embodiments, the mining the first user-demand knowledge vector and the first linkage-demand description field of the first digitized user-feedback record to be subjected to data recommendation analysis includes: loading the first digital user feedback record into a first feature mining model to obtain a first interest element mining result; loading the first interest element mining result to a second feature mining model to obtain a third interest element mining result; performing feature transformation operation on the third interest element mining result to obtain the first linkage demand description field; and performing feature transformation operation on the first interest element mining result to obtain the first user demand knowledge vector.
By the design, the first linkage demand description field and the first user demand knowledge vector of which the feature members are numerical values (such as integers) under the specified rule can be obtained, so that the resource waste of feature mining can be reduced, and the knowledge feature operation deviation caused by non-integers can be avoided as much as possible.
In some independent embodiments, the obtaining the data push related variable using the first user requirement knowledge vector and the first linkage requirement description field includes: loading the first linkage requirement description field into a first characteristic translation model subjected to setting rule adjustment processing to obtain a second interest element mining result; and obtaining the data pushing related variable by using the second interest element mining result and the first user demand knowledge vector.
By the design, the first characteristic translation model after the rule adjustment processing can be set to obtain the mining result of the second interest element, so that the deviation of operation data is reduced, and the accuracy and the credibility of the data pushing related variables are improved.
In some independent embodiments, the obtaining the data push related variable by using the second interest element mining result and the first user demand knowledge vector includes: obtaining a first-stage interest element mining result and a second-stage interest element mining result by using the second interest element mining result and the first user demand knowledge vector, wherein the first-stage interest element mining result and the second-stage interest element mining result respectively correspond to part of knowledge vectors of the first user demand knowledge vector; obtaining a first push correlation regression variable and a second push correlation regression variable by using the first stage interest element mining result, the second stage interest element mining result and the regression analysis model after the setting rule adjustment treatment; and obtaining the data push correlation variable by using the first push correlation variable and the second push correlation variable.
By the design, variable regression analysis can be performed on the mining result of the second interest element based on the first user demand knowledge vector, and the accuracy and the reliability of the obtained data pushing related variables are improved.
In some independent embodiments, the obtaining the first stage interest element mining result and the second stage interest element mining result by using the second interest element mining result and the first user requirement knowledge vector includes: splitting the first user demand knowledge vector to obtain a second user demand knowledge vector and a third user demand knowledge vector; obtaining the first-stage interest element mining result by using the second user demand knowledge vector and the second interest element mining result; and obtaining the second-stage interest element mining result by using the third user demand knowledge vector and the second interest element mining result.
In the design, in the obtained two user demand knowledge vectors (the second user demand knowledge vector and the third user demand knowledge vector), continuous feature members of the mining result of the other interest element exist respectively, so that the relation between the feature members in the two user demand knowledge vectors is conveniently obtained, and the efficiency and the precision of classification mining can be improved.
In some independent embodiments, the splitting the first user-required knowledge vector to obtain a second user-required knowledge vector and a third user-required knowledge vector includes: and carrying out regularized splitting on the first user demand knowledge vector to obtain the second user demand knowledge vector and the third user demand knowledge vector.
By the design, the first-stage interest element mining result and the second-stage interest element mining result can be obtained through regularizing the mapping condition (corresponding relation) of the feature members in the split first user demand knowledge vector, so that the relation between the feature members of the first-stage interest element mining result and the second-stage interest element mining result can be conveniently obtained, and the classification mining efficiency and precision can be improved.
In some independent embodiments, obtaining a first push correlation regression variable and a second push correlation regression variable by using the first stage interest element mining result, the second stage interest element mining result, and the regression analysis model after the setting rule adjustment process includes: loading the first stage interest element mining result to a regression analysis model subjected to set rule adjustment processing to obtain a first push associated regression variable; loading the second user demand knowledge vector into a category regression model subjected to set rule adjustment processing to obtain a first interest category analysis result; and loading the first interest category analysis result and the second-stage interest element mining result into the regression analysis model subjected to the setting rule adjustment processing to obtain a second pushing associated regression variable.
By the design, when the data pushing associated variable is determined, the content of the relation among the continuous characteristic members can be considered, the first interest class analysis result is obtained, and the accuracy and the reliability of the data pushing associated variable can be improved.
In some independent embodiments, the performing a push decision analysis operation using the data push related variable, the first linkage requirement description field, and the first user requirement knowledge vector to obtain first data push decision information of the first digitized user feedback record includes: loading the data push related variable into a possibility prediction model after a statistical updating operation to obtain a first possibility variable, wherein the statistical updating operation enables the first possibility variable to be a determined variable; performing push preference mining processing by using the first possibility variable and the first user demand knowledge vector to obtain a push preference knowledge field; and obtaining the first data push decision information by using the push preference knowledge field and the first linkage demand description field.
By the design, the determined first probability variable can be output through the statistical updating operation possibility prediction model, pushing decision mining is performed based on the first probability variable, so that the accuracy of data pushing decision information can be guaranteed, the feature recognition degree of the data pushing decision information can be improved, the requirement-pushing progressive mining under a flexible scene is realized, and deviation in the feature translation process can be reduced (corresponding digital user feedback records are obtained based on the translation of the data pushing decision information).
In some independent embodiments, the method further comprises: acquiring a second linkage demand description field from second data push decision information to be processed; performing feature translation operation by using the second linkage requirement description field to obtain a fourth user requirement knowledge vector; and performing feature translation operation on the fourth user demand knowledge vector to obtain a second digitized user feedback record.
By the design, the fourth interest element mining result can be obtained based on the first feature translation model after the setting rule adjustment processing, so that the fourth interest element mining result is accurate information, and errors in the feature translation process are reduced.
In some independent embodiments, the performing the feature translation operation using the second linkage requirement description field to obtain a fourth user requirement knowledge vector includes: loading the second linkage demand description field into a first characteristic translation model subjected to set rule adjustment processing to obtain a fourth interest element mining result; obtaining a fifth user demand knowledge vector according to a fourth interest element mining result, wherein the fifth user demand knowledge vector corresponds to a part of knowledge vectors of the fourth interest element mining result; and obtaining the fourth user demand knowledge vector by using the fifth user demand knowledge vector and the fourth interest element mining result.
In some independent embodiments, the obtaining a fifth user demand knowledge vector according to the fourth interest element mining result includes: splitting the fourth interest element mining result to obtain a third-stage interest element mining result, wherein the third-stage interest element mining result is a partial knowledge vector of the fourth interest element mining result; loading the third-stage interest element mining result to the regression analysis model subjected to the setting rule adjustment processing to obtain a third push associated regression variable; loading the third push associated regression variable into a possibility prediction model after the statistical updating operation to obtain a second possibility variable; and performing feature translation operation on the second data push decision information by using the obtained second possibility variable to obtain a fifth user demand knowledge vector.
By the design, the fifth user demand knowledge vector corresponding to a part of all the user demand knowledge vectors can be obtained by setting the regression analysis model after the rule adjustment processing and the possibility prediction model after the statistical updating operation, and the accuracy of the fifth user demand knowledge vector is improved.
In some independent embodiments, the obtaining the fourth user-required knowledge vector using the fifth user-required knowledge vector and the fourth interest element mining result includes: splitting the fourth interest element mining result to obtain a fourth-stage interest element mining result, wherein the fourth-stage interest element mining result is a partial knowledge vector of the fourth interest element mining result; obtaining a fourth push associated regression variable by using the knowledge vector of the fifth user requirement and the fourth stage interest element mining result; obtaining a sixth user demand knowledge vector by using the fourth push associated regression variable; and obtaining the fourth user demand knowledge vector by using the fifth user demand knowledge vector and the sixth user demand knowledge vector.
By the design, the sixth user demand knowledge vector can be obtained based on the content of the relation between the adjacent feature members, so that the fourth user demand knowledge vector is obtained, and the accuracy and the reliability of the fourth user demand knowledge vector can be improved.
In some independent embodiments, the obtaining a fourth push associative regression variable using the fifth user requirement knowledge vector and the fourth stage interest element mining result includes: loading the knowledge vector required by the fifth user to a category regression model subjected to set rule adjustment processing to obtain a second interest category analysis result; and loading the second interest category analysis result and the fourth-stage interest element mining result into the regression analysis model subjected to the setting rule adjustment processing to obtain the fourth push associated regression variable.
By the design, the analysis result of the second interest category can be obtained, in other words, the content of the relation between the adjacent characteristic members is expressed, so that the accuracy and the reliability of the fourth push associated regression variable are improved.
In some independent embodiments, the method further comprises: performing feature transformation operation on the sliding filter unit variables by using a first target value and a second target value of the sliding filter unit variables with a plurality of sequential priorities of the first feature translation model to obtain sliding filter unit variables with a plurality of sequential priorities after the feature transformation operation; determining a plurality of sequential priorities of the first feature translation model respectively, and aiming at a first target value and a second target value in calculation data of set verification information, wherein the first linkage demand description field is loaded to the first feature translation model after the adjustment processing of the set rule to obtain a second interest element mining result, and the method comprises the following steps: obtaining raw material data of a plurality of sequential priorities by using a generating result of the plurality of sequential priorities of the first feature translation model after the setting rule adjustment processing and a first target value and a second target value in the calculation data, wherein the generating result of the plurality of sequential priorities comprises a processing result of the plurality of sequential priorities of the first feature translation model after the setting rule adjustment processing on the first linkage demand description field; and obtaining the second interest element mining result by using the raw material data with the sequence priority and the sliding filter unit variables with the sequence priority after the feature transformation operation.
By means of the design, errors of knowledge characteristic operation can be reduced through setting rule adjustment processing, flexibility of a scheme is improved, the structure of the AI machine learning model cannot be changed, performance of the AI machine learning model is guaranteed, and timeliness and accuracy of the model in an easy-to-use process can be improved.
In some independent embodiments, the method further comprises: carrying out quantization processing on the set statistical algorithm to obtain a quantization processing result; and obtaining the possibility prediction model after the statistical updating operation by using the quantization processing result.
By the design, the first possibility variable determined by the output of the possibility prediction model can be obtained through the statistical updating operation, so that the calling capacity under different scenes is obtained, and the error in the characteristic translation process is reduced. And the structure of the AI machine learning model is not required to be changed, so that the timeliness and the precision of the model in the easy-to-use process can be improved.
In a second aspect, the invention also provides a data recommendation processing system, which comprises a processor and a memory; the processor is in communication with the memory, and the processor is configured to read and execute a computer program from the memory to implement the method described above.
In a third aspect, the present invention also provides a computer-readable storage medium having stored thereon a program which, when executed by a processor, implements the method described above.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and together with the description, serve to explain the principles of the invention.
Fig. 1 is a flow chart of a data recommendation method based on user demand analysis according to an embodiment of the present invention.
Fig. 2 is a schematic diagram of a communication architecture of an application environment of a data recommendation method based on user demand analysis according to an embodiment of the present invention.
Detailed Description
Reference will now be made in detail to exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, the same numbers in different drawings refer to the same or similar elements, unless otherwise indicated. The implementations described in the following exemplary examples do not represent all implementations consistent with the invention. Rather, they are merely examples of apparatus and methods consistent with aspects of the invention as detailed in the accompanying claims.
It should be noted that the terms "first," "second," and the like in the description and the claims of the present invention and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order.
The method embodiments provided by the embodiments of the present invention may be performed in a data recommendation processing system, a computer device, or a similar computing device. Taking the example of running on a data recommendation processing system, the data recommendation processing system 10 may include one or more processors 102 (the processor 102 may include, but is not limited to, a processing device such as a microprocessor MCU or programmable logic device FPGA) and a memory 104 for storing data, and optionally the data recommendation processing system may also include a transmission device 106 for communication functions. It will be appreciated by those of ordinary skill in the art that the above-described structure is merely illustrative and is not intended to limit the structure of the data recommendation processing system. For example, the data recommendation processing system 10 may also include more or fewer components than those shown above, or have a different configuration than those shown above.
The memory 104 may be used to store a computer program, for example, a software program of application software and a module, such as a computer program corresponding to a data recommendation method based on user requirement analysis in an embodiment of the present invention, and the processor 102 executes the computer program stored in the memory 104 to perform various functional applications and data processing, that is, implement the method described above. Memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, memory 104 may further include memory remotely located with respect to processor 102, which may be connected to data recommendation processing system 10 via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The transmission means 106 is arranged to receive or transmit data via a network. Specific examples of the network described above may include a wireless network provided by a communication provider of data recommendation processing system 10. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, simply referred to as NIC) that can connect to other network devices through a base station to communicate with the internet. In one example, the transmission device 106 may be a Radio Frequency (RF) module, which is used to communicate with the internet wirelessly.
Based on this, referring to fig. 1, fig. 1 is a flowchart of a data recommendation method based on user demand analysis according to an embodiment of the present invention, where the method is applied to a data recommendation processing system, and further may include the technical solutions described in step 1 to step 3.
And step 1, mining a first user demand knowledge vector and a first linkage demand description field of a first digital user feedback record to be subjected to data recommendation analysis.
For example, the first digitized user feedback record may be a user feedback text or user feedback graphic information of the existing e-commerce service, and the first user requirement knowledge vector may be understood as a feature vector corresponding to the user requirement, such as "e-commerce interface optimization requirement", "service security detection requirement", "service push requirement", and the like. The first linkage requirement description field can be understood as a traditional connection feature, and is used for recording association information between different first user requirement knowledge vectors.
And 2, obtaining a data pushing related variable by using the first user demand knowledge vector and the first linkage demand description field, wherein the data pushing related variable is a characteristic extraction result obtained by an AI machine learning model after setting rule adjustment processing, and the setting rule adjustment processing is used for enabling a generation result of the AI machine learning model to be an expert knowledge vector.
For example, the data push related variable reflects the relevance between the requirements of the user in the data push requirement level, and the AI machine learning model after the adjustment processing of the setting rule can be understood as an expert knowledge vector/array for performing the integer processing of the feature value or the feature member on the AI machine learning model, that is, the integration of the feature value or the feature member into zero, so that the output information of the AI machine learning model can be ensured to be in an integer state.
And step 3, performing pushing decision analysis operation by using the data pushing related variable, the first linkage demand description field and the first user demand knowledge vector to obtain first data pushing decision information of the first digital user feedback record.
For example, the push decision analysis operation may be understood as performing progressive feature mining or big data analysis, so that progressive push decision mining analysis may be performed based on three levels, i.e., a requirement level, a push requirement level, and a requirement association level, so as to completely and accurately obtain first data push decision information, where the first data push decision information may be used to instruct subsequent information recommendation or service push, thereby implementing "push on demand", improving the intelligentization degree of push, and guaranteeing the utilization rate of limited channel resources. In addition, through setting rule adjustment processing (feature member integer processing), the variables such as knowledge features generated by the AI machine learning model are free from loss and errors, the accuracy of the obtained data pushing decision information can be ensured in the process of calling the AI machine learning model to conduct pushing decision analysis under different scenes, so that the flexibility of conducting data pushing decision analysis based on user demands is improved, errors in conducting data pushing decision analysis and feature translation under different scenes are reduced, progressive data mining analysis can be achieved based on user demand knowledge vectors and linkage demand description fields, and accordingly the data pushing decision information is accurately, completely and reasonably obtained.
In some possible embodiments, the mining the first user-requirement knowledge vector and the first linkage-requirement description field of the first digitized user-feedback record to be subjected to data recommendation analysis includes: loading the first digital user feedback record into a first feature mining model (which can be a coding model built based on a traditional convolutional neural network) to obtain a first interest element mining result; and loading the first interest element mining result to a second feature mining model to obtain a third interest element mining result.
Further, performing feature transformation operation (such as feature value integer processing) on the third interest element mining result to obtain the first linkage requirement description field; and performing feature transformation operation on the first interest element mining result to obtain the first user demand knowledge vector. By the design, the first linkage demand description field and the first user demand knowledge vector of which the feature members are numerical values (such as integers) under the specified rule can be obtained, so that the resource waste of feature mining can be reduced, and the knowledge feature operation deviation caused by non-integers can be avoided as much as possible.
Under some possible design ideas, the step a and the step b are included in the step a and the step b, in which the data push related variables are obtained by using the first user requirement knowledge vector and the first linkage requirement description field.
And a step a of loading the first linkage requirement description field into a first feature translation model (which can be understood as a decoding model or a feature mapping model and is used for carrying out prediction processing or derivative processing of the interest element) after the adjustment processing of the set rule to obtain a second interest element mining result.
And b, obtaining the data pushing related variable by using the second interest element mining result and the first user demand knowledge vector.
Further, the second interest element mining result focuses on the interest features of the subjective experience level of the user, and the first interest element mining result focuses on the interest features of the system function level. By adopting the design, the first characteristic translation model after the rule adjustment treatment can be set to obtain the mining result of the second interest element, so that the deviation of operation data is reduced, and the accuracy and the credibility of the data pushing related variables are improved
In another possible embodiment of writing, the step of obtaining the data push related variable by using the second interest element mining result and the first user demand knowledge vector includes steps s 1-s 3.
Step s1, obtaining a first-stage interest element mining result and a second-stage interest element mining result by using the second interest element mining result and the first user demand knowledge vector, wherein the first-stage interest element mining result and the second-stage interest element mining result respectively correspond to part of knowledge vectors of the first user demand knowledge vector.
And step s2, utilizing the first stage interest element mining result, the second stage interest element mining result and the regression analysis model (variable prediction analysis network) after the setting rule adjustment processing to obtain a first push associated regression variable and a second push associated regression variable.
Step s3, obtaining the data push associated variable by using the first push associated regression variable and the second push associated regression variable.
Based on the steps s 1-s 3, variable regression analysis can be performed on the second interest element mining result based on the first user demand knowledge vector, and the accuracy and the reliability of the obtained data pushing related variables are improved.
For some optional embodiments, the obtaining the first stage interest element mining result and the second stage interest element mining result by using the second interest element mining result and the first user requirement knowledge vector may include step p 1-step p3.
And step p1, splitting the first user demand knowledge vector to obtain a second user demand knowledge vector and a third user demand knowledge vector.
And p2, obtaining the first-stage interest element mining result by using the second user demand knowledge vector and the second interest element mining result.
And p3, obtaining the second-stage interest element mining result by using the third user demand knowledge vector and the second interest element mining result.
Where phase interests may be understood as local interests. In the design, in the obtained two user demand knowledge vectors (the second user demand knowledge vector and the third user demand knowledge vector), continuous feature members of the mining result of the other interest element exist respectively, so that the relation between the feature members in the two user demand knowledge vectors is conveniently obtained, and the efficiency and the precision of classification mining can be improved.
Illustratively, the splitting the first user demand knowledge vector to obtain a second user demand knowledge vector and a third user demand knowledge vector includes: and carrying out regularized splitting (such as decomposition processing or segmentation processing based on rule patterning) on the first user demand knowledge vector to obtain the second user demand knowledge vector and the third user demand knowledge vector. By the design, the first-stage interest element mining result and the second-stage interest element mining result can be obtained through regularizing the mapping condition (corresponding relation) of the feature members in the split first user demand knowledge vector, so that the relation between the feature members of the first-stage interest element mining result and the second-stage interest element mining result can be conveniently obtained, and the classification mining efficiency and precision can be improved
Under some possible design ideas, obtaining a first push correlation regression variable and a second push correlation regression variable by using the first stage interest element mining result, the second stage interest element mining result and the regression analysis model after the setting rule adjustment processing, including: loading the first-stage interest element mining result to a regression analysis model subjected to set rule adjustment processing to obtain a first push associated regression variable (which can be understood as a prediction result of the push associated variable); loading the second user demand knowledge vector into a category regression model (which can be understood as a DNN model for category prediction and analysis) after the adjustment processing of the set rule, and obtaining a first interest category analysis result (which can be understood as category information of interest features, such as a video interest category 1, a text interest category 2 and the like); and loading the first interest category analysis result and the second-stage interest element mining result into the regression analysis model subjected to the setting rule adjustment processing to obtain a second pushing associated regression variable. Therefore, when the data pushing associated variable is determined, the content of the relation among the continuous characteristic members is considered, the first interest class analysis result is obtained, and the accuracy and the reliability of the data pushing associated variable can be improved.
And under some possible design ideas, performing pushing decision analysis operation by using the data pushing related variable, the first linkage requirement description field and the first user requirement knowledge vector to obtain first data pushing decision information of the first digital user feedback record, wherein the first data pushing decision information comprises steps 01-03.
And step 01, loading the data push related variable into a possibility prediction model after a statistical updating operation to obtain a first possibility variable (prediction probability parameter), wherein the statistical updating operation enables the first possibility variable to be a determined variable (a variable without error).
And 02, performing push preference mining processing by using the first probability variable and the first user demand knowledge vector to obtain a push preference knowledge field (which can be understood as mining characteristics of push preferences).
And step 03, obtaining the first data push decision information by using the push preference knowledge field and the first linkage demand description field.
Based on the steps 01-03, the determined first probability variable can be output through the statistical updating operation possibility prediction model, and the push decision mining is performed based on the first probability variable, so that the precision of the data push decision information can be ensured, the feature recognition degree of the data push decision information can be improved, the requirement-push progressive mining under a flexible scene is realized, and the deviation in the feature translation process (corresponding digital user feedback records are obtained based on the data push decision information) can be reduced.
In some possible embodiments, the method further comprises: acquiring a second linkage demand description field from second data push decision information to be processed; performing feature translation operation by using the second linkage requirement description field to obtain a fourth user requirement knowledge vector; and performing feature translation operation on the fourth user demand knowledge vector to obtain a second digitized user feedback record. By the design, the fourth interest element mining result can be obtained based on the first feature translation model after the setting rule adjustment processing, so that the fourth interest element mining result is accurate information, and errors in the feature translation process are reduced.
In other possible embodiments, the performing a feature translation operation using the second linkage requirement description field to obtain a fourth user requirement knowledge vector includes: loading the second linkage demand description field into a first characteristic translation model subjected to set rule adjustment processing to obtain a fourth interest element mining result; obtaining a fifth user demand knowledge vector according to a fourth interest element mining result, wherein the fifth user demand knowledge vector corresponds to a part of knowledge vectors of the fourth interest element mining result; and obtaining the fourth user demand knowledge vector by using the fifth user demand knowledge vector and the fourth interest element mining result.
In still other possible embodiments, the obtaining a fifth user demand knowledge vector according to the fourth interest element mining result includes: splitting the fourth interest element mining result to obtain a third-stage interest element mining result, wherein the third-stage interest element mining result is a partial knowledge vector of the fourth interest element mining result; loading the third-stage interest element mining result to the regression analysis model subjected to the setting rule adjustment processing to obtain a third push associated regression variable; loading the third push associated regression variable into a possibility prediction model after the statistical updating operation to obtain a second possibility variable; and performing feature translation operation on the second data push decision information by using the obtained second possibility variable to obtain a fifth user demand knowledge vector. By the design, the fifth user demand knowledge vector corresponding to a part of all the user demand knowledge vectors can be obtained by setting the regression analysis model after the rule adjustment processing and the possibility prediction model after the statistical updating operation, and the accuracy of the fifth user demand knowledge vector is improved.
Optionally, the obtaining the fourth user demand knowledge vector by using the fifth user demand knowledge vector and the fourth interest element mining result includes: splitting the fourth interest element mining result to obtain a fourth-stage interest element mining result, wherein the fourth-stage interest element mining result is a partial knowledge vector of the fourth interest element mining result; obtaining a fourth push associated regression variable by using the knowledge vector of the fifth user requirement and the fourth stage interest element mining result; obtaining a sixth user demand knowledge vector by using the fourth push associated regression variable; and obtaining the fourth user demand knowledge vector by using the fifth user demand knowledge vector and the sixth user demand knowledge vector. By the design, the sixth user demand knowledge vector can be obtained based on the content of the relation between the adjacent feature members, so that the fourth user demand knowledge vector is obtained, and the accuracy and the reliability of the fourth user demand knowledge vector can be improved.
Further, the obtaining a fourth push associated regression variable by using the knowledge vector of the fifth user requirement and the fourth stage interest element mining result includes: loading the knowledge vector required by the fifth user to a category regression model subjected to set rule adjustment processing to obtain a second interest category analysis result; and loading the second interest category analysis result and the fourth-stage interest element mining result into the regression analysis model subjected to the setting rule adjustment processing to obtain the fourth push associated regression variable. By the design, the analysis result of the second interest category can be obtained, in other words, the content of the relation between the adjacent characteristic members is expressed, so that the accuracy and the reliability of the fourth push associated regression variable are improved.
In some independent embodiments, the method further comprises: performing feature transformation operation on the sliding filter unit variables by using a first target value (maximum value) and a second target value (minimum value) of the sliding filter unit variables (convolution kernel parameters) of a plurality of sequential priorities of the first feature translation model to obtain sliding filter unit variables of a plurality of sequential priorities after the feature transformation operation; determining a plurality of sequential priorities of the first feature translation model, and aiming at a first target value and a second target value in calculation data (feature operation result) of set verification information (verification data), wherein the first linkage requirement description field is loaded to the first feature translation model after the setting rule adjustment processing to obtain a second interest element mining result, and the method comprises the following steps: obtaining raw material data (input information) of a plurality of sequential priorities by using a generating result of the plurality of sequential priorities of the first feature translation model after the setting rule adjustment processing and a first target value and a second target value in the calculation data, wherein the generating result of the plurality of sequential priorities comprises a processing result of the first linkage demand description field by the plurality of sequential priorities of the first feature translation model after the setting rule adjustment processing; and obtaining the second interest element mining result by using the raw material data with the sequence priority and the sliding filter unit variables with the sequence priority after the feature transformation operation.
By means of the design, errors of knowledge characteristic operation can be reduced through setting rule adjustment processing, flexibility of a scheme is improved, the structure of the AI machine learning model cannot be changed, performance of the AI machine learning model is guaranteed, and timeliness and accuracy of the model in an easy-to-use process can be improved.
In some independent embodiments, the method further comprises: carrying out quantization processing on the set statistical algorithm to obtain a quantization processing result; and obtaining the possibility prediction model after the statistical updating operation by using the quantization processing result.
By the design, the first possibility variable determined by the output of the possibility prediction model can be obtained through the statistical updating operation, so that the calling capacity under different scenes is obtained, and the error in the characteristic translation process is reduced. And the structure of the AI machine learning model is not required to be changed, so that the timeliness and the precision of the model in the easy-to-use process can be improved.
It can be appreciated that the training process of the neural network/machine learning model can be selected according to actual requirements, for example, feedback training can be performed based on a cross entropy loss function, and a person skilled in the art can flexibly select a model training scheme, which is not limited herein.
Based on the same or similar inventive concept, please refer to fig. 2, a schematic architecture diagram of an application environment 30 of a data recommendation method based on user requirement analysis is further provided, including a data recommendation processing system 10 and a digital user device 20 that communicate with each other, where the data recommendation processing system 10 and the digital user device 20 implement or partially implement the technical scheme described in the above method embodiments at runtime.
Further, there is also provided a computer-readable storage medium having stored thereon a program which, when executed by a processor, implements the above-described method.
In the embodiments provided in the present invention, it should be understood that the disclosed apparatus and method may be implemented in other manners. The apparatus and method embodiments described above are merely illustrative, for example, flow diagrams and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
In addition, functional modules in the embodiments of the present invention may be integrated together to form a single part, or each module may exist alone, or two or more modules may be integrated to form a single part.
The functions, if implemented in the form of software functional modules and sold or used as a stand-alone product, may be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a network device, or the like) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes. 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 apparatus 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 apparatus. 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 apparatus that comprises the element.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (5)
1. A data recommendation method based on user demand analysis, applied to a data recommendation processing system, the method comprising:
mining a first user demand knowledge vector and a first linkage demand description field of a first digital user feedback record to be subjected to data recommendation analysis;
obtaining a data pushing related variable by using the first user demand knowledge vector and the first linkage demand description field; the data pushing related variable is a feature extraction result obtained by an AI machine learning model after setting rule adjustment processing, wherein the setting rule adjustment processing is used for enabling a generation result of the AI machine learning model to be an expert knowledge vector;
performing pushing decision analysis operation by using the data pushing related variable, the first linkage demand description field and the first user demand knowledge vector to obtain first data pushing decision information of the first digital user feedback record;
The mining of the first user demand knowledge vector and the first linkage demand description field of the first digitized user feedback record to be subjected to data recommendation analysis includes:
loading the first digital user feedback record into a first feature mining model to obtain a first interest element mining result;
loading the first interest element mining result to a second feature mining model to obtain a third interest element mining result;
performing feature transformation operation on the third interest element mining result to obtain the first linkage demand description field;
performing feature transformation operation on the first interest element mining result to obtain the first user demand knowledge vector;
the obtaining the data push related variable by using the first user requirement knowledge vector and the first linkage requirement description field includes:
loading the first linkage requirement description field into a first characteristic translation model subjected to setting rule adjustment processing to obtain a second interest element mining result;
obtaining the data pushing related variable by using the second interest element mining result and the first user demand knowledge vector;
the obtaining the data pushing related variable by using the second interest element mining result and the first user demand knowledge vector includes:
Obtaining a first-stage interest element mining result and a second-stage interest element mining result by using the second interest element mining result and the first user demand knowledge vector, wherein the first-stage interest element mining result and the second-stage interest element mining result respectively correspond to part of knowledge vectors of the first user demand knowledge vector;
obtaining a first push correlation regression variable and a second push correlation regression variable by using the first stage interest element mining result, the second stage interest element mining result and the regression analysis model after the setting rule adjustment treatment;
obtaining the data push correlation variable by utilizing the first push correlation variable and the second push correlation variable;
the obtaining a first stage interest element mining result and a second stage interest element mining result by using the second interest element mining result and the first user demand knowledge vector includes:
splitting the first user demand knowledge vector to obtain a second user demand knowledge vector and a third user demand knowledge vector;
obtaining the first-stage interest element mining result by using the second user demand knowledge vector and the second interest element mining result;
Obtaining the second-stage interest element mining result by using the third user demand knowledge vector and the second interest element mining result;
splitting the first user demand knowledge vector to obtain a second user demand knowledge vector and a third user demand knowledge vector, including: the first user demand knowledge vector is subjected to regularized splitting to obtain the second user demand knowledge vector and the third user demand knowledge vector;
the method for obtaining the first push correlation regression variable and the second push correlation regression variable by using the first stage interest element mining result, the second stage interest element mining result and the regression analysis model after the setting rule adjustment processing comprises the following steps: loading the first stage interest element mining result to a regression analysis model subjected to set rule adjustment processing to obtain a first push associated regression variable; loading the second user demand knowledge vector into a category regression model subjected to set rule adjustment processing to obtain a first interest category analysis result; and loading the first interest category analysis result and the second-stage interest element mining result into the regression analysis model subjected to the setting rule adjustment processing to obtain a second pushing associated regression variable.
2. The method of claim 1, wherein performing a push decision analysis operation using the data push correlation variable, the first linkage demand description field, and the first user demand knowledge vector to obtain first data push decision information for the first digitized user feedback record comprises:
loading the data push related variable into a possibility prediction model after a statistical updating operation to obtain a first possibility variable, wherein the statistical updating operation enables the first possibility variable to be a determined variable;
performing push preference mining processing by using the first possibility variable and the first user demand knowledge vector to obtain a push preference knowledge field;
and obtaining the first data push decision information by using the push preference knowledge field and the first linkage demand description field.
3. The method according to claim 2, wherein the method further comprises: acquiring a second linkage demand description field from second data push decision information to be processed; performing feature translation operation by using the second linkage requirement description field to obtain a fourth user requirement knowledge vector; performing feature translation operation on the fourth user demand knowledge vector to obtain a second digitized user feedback record;
The feature translation operation is performed by using the second linkage requirement description field to obtain a fourth user requirement knowledge vector, which includes: loading the second linkage demand description field into a first characteristic translation model subjected to set rule adjustment processing to obtain a fourth interest element mining result; obtaining a fifth user demand knowledge vector according to a fourth interest element mining result, wherein the fifth user demand knowledge vector corresponds to a part of knowledge vectors of the fourth interest element mining result; obtaining a fourth user demand knowledge vector by using the fifth user demand knowledge vector and the fourth interest element mining result;
the obtaining a fifth user demand knowledge vector according to the fourth interest element mining result includes: splitting the fourth interest element mining result to obtain a third-stage interest element mining result, wherein the third-stage interest element mining result is a partial knowledge vector of the fourth interest element mining result; loading the third-stage interest element mining result to the regression analysis model subjected to the setting rule adjustment processing to obtain a third push associated regression variable; loading the third push associated regression variable into a possibility prediction model after the statistical updating operation to obtain a second possibility variable; performing feature translation operation on the second data push decision information by using the obtained second possibility variable to obtain a fifth user demand knowledge vector;
The obtaining the fourth user demand knowledge vector by using the fifth user demand knowledge vector and the fourth interest element mining result includes: splitting the fourth interest element mining result to obtain a fourth-stage interest element mining result, wherein the fourth-stage interest element mining result is a partial knowledge vector of the fourth interest element mining result; obtaining a fourth push associated regression variable by using the knowledge vector of the fifth user requirement and the fourth stage interest element mining result; obtaining a sixth user demand knowledge vector by using the fourth push associated regression variable; obtaining the fourth user demand knowledge vector by using the fifth user demand knowledge vector and the sixth user demand knowledge vector;
the obtaining a fourth push associated regression variable by using the knowledge vector of the fifth user requirement and the fourth stage interest element mining result includes: loading the knowledge vector required by the fifth user to a category regression model subjected to set rule adjustment processing to obtain a second interest category analysis result; and loading the second interest category analysis result and the fourth-stage interest element mining result into the regression analysis model subjected to the setting rule adjustment processing to obtain the fourth push associated regression variable.
4. The method according to claim 1, wherein the method further comprises:
performing feature transformation operation on the sliding filter unit variables by using a first target value and a second target value of the sliding filter unit variables with a plurality of sequential priorities of the first feature translation model to obtain sliding filter unit variables with a plurality of sequential priorities after the feature transformation operation;
determining a plurality of sequential priorities of the first feature translation model respectively, and aiming at a first target value and a second target value in calculation data of set verification information, wherein the first linkage demand description field is loaded to the first feature translation model after the adjustment processing of the set rule to obtain a second interest element mining result, and the method comprises the following steps:
obtaining raw material data of a plurality of sequential priorities by using a generating result of the plurality of sequential priorities of the first feature translation model after the setting rule adjustment processing and a first target value and a second target value in the calculation data, wherein the generating result of the plurality of sequential priorities comprises a processing result of the plurality of sequential priorities of the first feature translation model after the setting rule adjustment processing on the first linkage demand description field; and obtaining the second interest element mining result by using the raw material data with the sequence priority and the sliding filter unit variables with the sequence priority after the feature transformation operation.
5. A data recommendation processing system, comprising a processor and a memory; the processor is communicatively connected to the memory, the processor being configured to read a computer program from the memory and execute the computer program to implement the method of any of claims 1-4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211553047.1A CN115905702B (en) | 2022-12-06 | 2022-12-06 | Data recommendation method and system based on user demand analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211553047.1A CN115905702B (en) | 2022-12-06 | 2022-12-06 | Data recommendation method and system based on user demand analysis |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115905702A CN115905702A (en) | 2023-04-04 |
| CN115905702B true CN115905702B (en) | 2023-10-10 |
Family
ID=86493624
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211553047.1A Active CN115905702B (en) | 2022-12-06 | 2022-12-06 | Data recommendation method and system based on user demand analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115905702B (en) |
Citations (24)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2009059199A2 (en) * | 2007-11-02 | 2009-05-07 | Hunch Inc. | Interactive machine learning advice facility |
| CN103019673A (en) * | 2012-11-14 | 2013-04-03 | 北京仟手莲科技有限公司 | Intelligent decision-making and entity recommending union system based on internet and work flow |
| CN107491992A (en) * | 2017-08-25 | 2017-12-19 | 哈尔滨工业大学（威海） | A kind of intelligent Service proposed algorithm based on cloud computing |
| CN107656938A (en) * | 2016-07-26 | 2018-02-02 | 北京搜狗科技发展有限公司 | It is a kind of to recommend method and apparatus, a kind of device for being used to recommend |
| CN109597940A (en) * | 2018-12-06 | 2019-04-09 | 上海哔哩哔哩科技有限公司 | Based on target group's determination of commercial interest and information-pushing method and system |
| CN110136838A (en) * | 2019-04-29 | 2019-08-16 | 平安科技（深圳）有限公司 | Data Matching decision-making technique and system based on multiple knowledge base reasoning |
| CN110245874A (en) * | 2019-03-27 | 2019-09-17 | 中国海洋大学 | A kind of Decision fusion method based on machine learning and knowledge reasoning |
| CN111475637A (en) * | 2020-06-24 | 2020-07-31 | 支付宝(杭州)信息技术有限公司 | Data processing and training method and device for pushing knowledge points |
| CN111737451A (en) * | 2020-05-14 | 2020-10-02 | 上海交通大学 | Expert recommendation method based on hyper-network model |
| CN111949711A (en) * | 2020-08-18 | 2020-11-17 | 齐鲁工业大学 | Commodity recommendation system based on decision-making high-utility negative sequence rule mining and working method thereof |
| CN112256962A (en) * | 2020-10-21 | 2021-01-22 | 网娱互动科技(北京)股份有限公司 | Method for intelligently recommending house decoration scheme |
| US11017321B1 (en) * | 2020-11-23 | 2021-05-25 | Accenture Global Solutions Limited | Machine learning systems for automated event analysis and categorization, equipment status and maintenance action recommendation |
| CN113111250A (en) * | 2020-01-09 | 2021-07-13 | 中国移动通信有限公司研究院 | Service recommendation method and device, related equipment and storage medium |
| CN113343106A (en) * | 2021-06-29 | 2021-09-03 | 山东建筑大学 | Intelligent student recommendation method and system |
| CN113420152A (en) * | 2021-08-23 | 2021-09-21 | 北京好欣晴移动医疗科技有限公司 | Service processing method, device and system based on fuzzy logic |
| CN113889209A (en) * | 2021-09-26 | 2022-01-04 | 浙江禾连网络科技有限公司 | Recommendation system and storage medium for health management service products |
| CN114579875A (en) * | 2022-03-03 | 2022-06-03 | 中国科学院大学 | Equipment fault diagnosis and maintenance knowledge recommendation system based on knowledge graph |
| CN114840486A (en) * | 2022-06-28 | 2022-08-02 | 广州趣米网络科技有限公司 | User behavior data acquisition method and system and cloud platform |
| CN115168453A (en) * | 2022-07-26 | 2022-10-11 | 赵雨 | Interest analysis method and system for e-commerce data push |
| CN115203498A (en) * | 2022-06-30 | 2022-10-18 | 寇占香 | Data information pushing analysis method and system applying expert system |
| CN115238197A (en) * | 2022-09-26 | 2022-10-25 | 中国电子科技集团公司第十研究所 | Expert thinking model-based field business auxiliary analysis method |
| CN115329204A (en) * | 2022-03-08 | 2022-11-11 | 张伟斌 | Cloud business service pushing method and pushing processing system based on big data mining |
| CN115329205A (en) * | 2022-09-15 | 2022-11-11 | 郭振东 | Big data mining method serving personalized push service and AI recommendation system |
| CN115422463A (en) * | 2022-09-27 | 2022-12-02 | 高冬 | Big data-based user analysis push processing method and system |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105446970A (en) * | 2014-06-10 | 2016-03-30 | 华为技术有限公司 | Item recommendation method and device |
| US11687801B2 (en) * | 2018-04-23 | 2023-06-27 | Qliktech International Ab | Knowledge graph data structures and uses thereof |
| US11386358B2 (en) * | 2019-04-17 | 2022-07-12 | International Business Machines Corporation | Intelligent decision support system |
| US20210133670A1 (en) * | 2019-11-05 | 2021-05-06 | Strong Force Vcn Portfolio 2019, Llc | Control tower and enterprise management platform with a machine learning/artificial intelligence managing sensor and the camera feeds into digital twin |
-
2022
- 2022-12-06 CN CN202211553047.1A patent/CN115905702B/en active Active
Patent Citations (24)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2009059199A2 (en) * | 2007-11-02 | 2009-05-07 | Hunch Inc. | Interactive machine learning advice facility |
| CN103019673A (en) * | 2012-11-14 | 2013-04-03 | 北京仟手莲科技有限公司 | Intelligent decision-making and entity recommending union system based on internet and work flow |
| CN107656938A (en) * | 2016-07-26 | 2018-02-02 | 北京搜狗科技发展有限公司 | It is a kind of to recommend method and apparatus, a kind of device for being used to recommend |
| CN107491992A (en) * | 2017-08-25 | 2017-12-19 | 哈尔滨工业大学（威海） | A kind of intelligent Service proposed algorithm based on cloud computing |
| CN109597940A (en) * | 2018-12-06 | 2019-04-09 | 上海哔哩哔哩科技有限公司 | Based on target group's determination of commercial interest and information-pushing method and system |
| CN110245874A (en) * | 2019-03-27 | 2019-09-17 | 中国海洋大学 | A kind of Decision fusion method based on machine learning and knowledge reasoning |
| CN110136838A (en) * | 2019-04-29 | 2019-08-16 | 平安科技（深圳）有限公司 | Data Matching decision-making technique and system based on multiple knowledge base reasoning |
| CN113111250A (en) * | 2020-01-09 | 2021-07-13 | 中国移动通信有限公司研究院 | Service recommendation method and device, related equipment and storage medium |
| CN111737451A (en) * | 2020-05-14 | 2020-10-02 | 上海交通大学 | Expert recommendation method based on hyper-network model |
| CN111475637A (en) * | 2020-06-24 | 2020-07-31 | 支付宝(杭州)信息技术有限公司 | Data processing and training method and device for pushing knowledge points |
| CN111949711A (en) * | 2020-08-18 | 2020-11-17 | 齐鲁工业大学 | Commodity recommendation system based on decision-making high-utility negative sequence rule mining and working method thereof |
| CN112256962A (en) * | 2020-10-21 | 2021-01-22 | 网娱互动科技(北京)股份有限公司 | Method for intelligently recommending house decoration scheme |
| US11017321B1 (en) * | 2020-11-23 | 2021-05-25 | Accenture Global Solutions Limited | Machine learning systems for automated event analysis and categorization, equipment status and maintenance action recommendation |
| CN113343106A (en) * | 2021-06-29 | 2021-09-03 | 山东建筑大学 | Intelligent student recommendation method and system |
| CN113420152A (en) * | 2021-08-23 | 2021-09-21 | 北京好欣晴移动医疗科技有限公司 | Service processing method, device and system based on fuzzy logic |
| CN113889209A (en) * | 2021-09-26 | 2022-01-04 | 浙江禾连网络科技有限公司 | Recommendation system and storage medium for health management service products |
| CN114579875A (en) * | 2022-03-03 | 2022-06-03 | 中国科学院大学 | Equipment fault diagnosis and maintenance knowledge recommendation system based on knowledge graph |
| CN115329204A (en) * | 2022-03-08 | 2022-11-11 | 张伟斌 | Cloud business service pushing method and pushing processing system based on big data mining |
| CN114840486A (en) * | 2022-06-28 | 2022-08-02 | 广州趣米网络科技有限公司 | User behavior data acquisition method and system and cloud platform |
| CN115203498A (en) * | 2022-06-30 | 2022-10-18 | 寇占香 | Data information pushing analysis method and system applying expert system |
| CN115168453A (en) * | 2022-07-26 | 2022-10-11 | 赵雨 | Interest analysis method and system for e-commerce data push |
| CN115329205A (en) * | 2022-09-15 | 2022-11-11 | 郭振东 | Big data mining method serving personalized push service and AI recommendation system |
| CN115238197A (en) * | 2022-09-26 | 2022-10-25 | 中国电子科技集团公司第十研究所 | Expert thinking model-based field business auxiliary analysis method |
| CN115422463A (en) * | 2022-09-27 | 2022-12-02 | 高冬 | Big data-based user analysis push processing method and system |
Non-Patent Citations (2)
| Title |
|---|
| Adaptive KNN based Recommender System through Mining of User Preferences;V. Subramaniyaswamy;《Wireless Personal Communications 》;第97卷;2229-2247 * |
| 基于多源数据融合的移动商务情景推荐研究;方卉;《中国优秀硕士学位论文全文数据库 信息科技》;I138-950 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115905702A (en) | 2023-04-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9990558B2 (en) | Generating image features based on robust feature-learning | |
| Li et al. | Deep unsupervised image hashing by maximizing bit entropy | |
| US10795645B2 (en) | Neural network for program synthesis | |
| EP3767549A1 (en) | Delivery of compressed neural networks | |
| CN110717099A (en) | Method and terminal for recommending film | |
| CN112381216A (en) | Training and predicting method and device for mixed graph neural network model | |
| CN113408668A (en) | Decision tree construction method and device based on federated learning system and electronic equipment | |
| CN113468432A (en) | Mobile internet-based user behavior big data analysis method and system | |
| CN113627207A (en) | Bar code identification method and device, computer equipment and storage medium | |
| CN113673260A (en) | Model processing method, device, storage medium and processor | |
| CN115016965A (en) | Method, device, equipment and storage medium for detecting faults of metering automation master station | |
| CN115905702B (en) | Data recommendation method and system based on user demand analysis | |
| Ezekannagha et al. | Dnasmart: Multiple attribute ranking tool for dna data storage systems | |
| CN112417874A (en) | Named entity recognition method and device, storage medium and electronic device | |
| CN110275991B (en) | Hash value determination method and device, storage medium and electronic device | |
| Huang et al. | Flow of renyi information in deep neural networks | |
| CN113963241B (en) | FPGA hardware architecture, data processing method thereof and storage medium | |
| CN112395864B (en) | Text error correction model training method, text error correction method and related device | |
| WO2019234291A1 (en) | An apparatus, a method and a computer program for selecting a neural network | |
| CN115392234A (en) | Text representation method, word representation method, corresponding device, medium and equipment | |
| CN115880024A (en) | Method for recommending lightweight graph convolutional neural network based on pre-training | |
| CN115205736A (en) | Video data identification method and device, electronic equipment and storage medium | |
| KR102305981B1 (en) | Method for Training to Compress Neural Network and Method for Using Compressed Neural Network | |
| CN114372383A (en) | Scene fast switching method and system based on VR simulation scene | |
| EP3767548A1 (en) | Delivery of compressed neural networks |
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 | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20230913 Address after: Unit 601, No. 768 Gaoqi Road, Xiamen Area, China (Fujian) Pilot Free Trade Zone, Xiamen, Fujian 361000 Applicant after: Cifnews (Xiamen) Cross-border E-commerce Co.,Ltd. Address before: No. 123, Renmin Street, Juancheng County, Heze City, Shandong Province, 274600 Applicant before: Juancheng Xinning Network Technology Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |