CN112286977A - Data pushing method, electronic equipment and system based on cloud computing - Google Patents
Data pushing method, electronic equipment and system based on cloud computing Download PDFInfo
- Publication number
- CN112286977A CN112286977A CN202011182629.4A CN202011182629A CN112286977A CN 112286977 A CN112286977 A CN 112286977A CN 202011182629 A CN202011182629 A CN 202011182629A CN 112286977 A CN112286977 A CN 112286977A
- Authority
- CN
- China
- Prior art keywords
- data
- vector
- information
- network
- node
- 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.)
- Withdrawn
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2457—Query processing with adaptation to user needs
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/55—Push-based network services
Abstract
The application relates to a data pushing method, electronic equipment and system based on cloud computing. By applying the scheme, additional feature dimensions can be further mined on the basis of the first feature vector of the user behavior data, the efficiency and comprehensiveness of feature analysis on the user behavior data are improved, and the feature dimensions are prevented from being omitted. Therefore, the target characteristic vector corresponding to the user behavior data can be accurately and completely determined, the expected data can be accurately determined and pushed based on the target characteristic vector, the pushing accuracy of the cloud server in the data pushing process is improved, and the data is prevented from being pushed by mistake.
Description
Technical Field
The application relates to the technical field of cloud computing data pushing, in particular to a data pushing method, electronic equipment and system based on cloud computing.
Background
With the development of cloud computing, data interaction has been widely applied to various industries. Taking the data pushing technology as an example, the cloud server may analyze the user behavior data of the intelligent terminal, so as to determine expected data of a user corresponding to the intelligent terminal, and then determine the expected data from the database and push the expected data to the intelligent terminal. However, in the actual data pushing process, the cloud server often has the problem of low pushing accuracy or mistaken pushing.
Disclosure of Invention
The application provides a data pushing method, electronic equipment and system based on cloud computing, and aims to solve the problems of low pushing accuracy or mistaken pushing in the prior art.
According to an aspect of the present application, there is provided a data pushing method based on cloud computing, applied to an electronic device, the method including:
acquiring user behavior data generated by processing an operation instruction input by a user through the intelligent terminal from the intelligent terminal; the operation instruction comprises a touch instruction, a voice instruction or a facial expression instruction; the user behavior data includes the following four categories of behavior data: behavior parameters, generation time information corresponding to the behavior parameters, parameter category information corresponding to the behavior parameters and data format information corresponding to the intelligent terminal;
extracting a first feature vector corresponding to the user behavior data and second feature vectors corresponding to the behavior parameters, the generation time information, the parameter category information and the data format information respectively; the first feature vector is an N-dimensional vector, the second feature vector is an N + 1-dimensional vector, N is a positive integer, and a last one-dimensional feature value of the second feature vector is used for representing a behavior data category corresponding to the second feature vector;
determining a first vector network corresponding to the first feature vector and a second vector network of each second feature vector; the first vector network and each second vector network comprise at least a plurality of network nodes with different network response weights, each network node corresponds to a vector value, the network response weights are convergence influence coefficients of the network nodes on the vector networks, the convergence influence coefficients are used for representing dimension expansion values of characteristic vectors corresponding to the vector networks, and the number of the network nodes in the first vector network and the number of the network nodes in each second vector network are both N;
determining an initial network node of any network response weight of the first feature vector in the first vector network, and determining a network node with the largest network response weight in each second vector network as a target network node; acquiring node information similarity between the initial network node and each target network node, wherein the node information similarity is used for representing the matching degree of the initial network node and each target network node on the vector value level;
determining a dimension expansion coefficient of each second feature vector relative to the first feature vector based on the similarity of each node information; carrying out dimension expansion processing on the first feature vector according to all the determined dimension expansion coefficients to obtain a target feature vector with dimensions of N + M, wherein M is a positive integer;
and determining at least one expected data from a preset database according to the target characteristic vector, and pushing the expected data to the intelligent terminal.
Preferably, the extracting a first feature vector corresponding to the user behavior data includes:
sequentially listing a plurality of data fields of the user behavior data according to a sequence of field lengths from large to small to obtain a data field sequence corresponding to the user behavior data; determining a field length difference value between two adjacent data fields in the data field sequence aiming at the data field sequence; determining the sequence distribution characteristics of the data field sequence according to the determined length difference values of all the fields; the data field is divided according to time intervals, and the sequence distribution characteristics are used for representing the field length distribution condition of the data field sequence;
extracting a sequence distribution value used for representing field distribution discrete degree of the data field in the sequence distribution characteristics, and generating a data field track corresponding to the data field sequence based on the sequence distribution value, wherein the data field track comprises a plurality of track nodes, the track nodes correspond to the data field one by one, each track node is connected with at least one track node except the track node in the data field track, a relevance weight exists between two mutually connected track nodes, the relevance weight has a priority from large to small, and the priority is used for representing an influence factor between the two mutually connected track nodes;
listing every two track nodes with interconnection relation according to the sequence of the priority from low to high to obtain a track node sequence, and removing repeated track nodes in the track node sequence to obtain a target track node sequence;
for each target track node in the target track node sequence, mapping field information in a data field corresponding to the target track node to a preset coordinate plane according to a preset mapping relation to obtain a mapping coordinate value, and determining a characteristic value corresponding to each data field based on the mapping coordinate value; and sequencing the characteristic values according to the data field sequence to obtain a first characteristic vector corresponding to the user behavior data.
Preferably, the extracting the second feature vector corresponding to the behavior parameter includes:
determining at least a plurality of execution functions corresponding to the behavior parameters, wherein the execution functions are used for executing the operation instruction of the user and outputting the behavior parameters corresponding to the operation instruction;
determining input information of each execution function according to the operation instruction and determining output information of each execution function according to the behavior parameters; determining execution logic information corresponding to each execution function based on each input information and the corresponding output information;
for each piece of execution logic information, judging whether a call log exists in the piece of execution logic information, wherein the call log is an operation log of the electronic equipment for reflecting and calling a hook function according to a first function type of the execution function so as to process input information corresponding to the execution function through the hook function to obtain corresponding output information;
when the call log exists in the execution logic information, determining a second function type of the hook function corresponding to the logic execution information; determining a feature extraction list of the behavior parameters according to at least part of the first function types and at least part of the second function types corresponding to the behavior parameters; and extracting the features of the behavior parameters based on the feature extraction list, and obtaining second feature vectors corresponding to the behavior parameters based on the behavior data categories corresponding to the behavior parameters.
Preferably, the obtaining of the node information similarity between the initial network node and each target network node includes:
determining a first information list corresponding to the node information of the initial network node and a second information list corresponding to the node information of each target network node; the first information list and the second information list have the same row number and column number, the list units of the first information list and the second information list have the same number, the first list units in the first information list have different first unit weights, and the second list units in the second information list have different second unit weights;
comparing each first list unit in the first information list with a corresponding second list unit in the second information list one by one to obtain a comparison result; when the comparison result represents that the information in the first list unit is the same as the information in the second list unit, determining the comparison similarity of the comparison result as a first set numerical value; when the comparison result represents that the information in the first list unit is different from the information in the second list unit, weighting a second set numerical value based on a first unit weight corresponding to the first list unit and a second unit weight corresponding to the second list unit to obtain a comparison similarity corresponding to the comparison result; and determining the mean value of the comparison similarity as the node information similarity between the initial network and each target network node.
Preferably, the determining the dimension expansion coefficient of each second feature vector relative to the first feature vector based on each node information similarity includes:
determining a similarity difference value between every two adjacent node information similarities and a median of the node information similarities;
determining an accumulated value of similarity difference values falling into a set interval which takes the median as an interval midpoint and takes the similarity average value of the node information similarity as an interval length;
judging whether the accumulated value reaches a set value; if so, determining the dimension expansion coefficient of the second feature vector corresponding to the node information similarity relative to the first feature vector according to the difference value of the node information similarity and the median; and if not, determining the dimension expansion coefficient of the second feature vector corresponding to the node information similarity relative to the first feature vector according to the difference value of the similarity of each node information similarity and the similarity average value.
Preferably, the performing, according to all the determined dimension expansion coefficients, dimension expansion processing on the first feature vector to obtain an N + M-dimensional target feature vector includes:
determining a dimension reference value for performing dimension expansion processing on the first feature vector according to all the dimension expansion coefficients;
correcting the dimension reference value according to the vector dimension and the vector confidence of the first feature vector to obtain a target dimension value;
and carrying out dimension expansion processing on the first feature vector according to the target dimension value M to obtain an N + M-dimensional target feature vector.
According to an aspect of the present application, there is provided a data pushing apparatus based on cloud computing, applied to an electronic device, the apparatus including:
the data acquisition module is used for acquiring user behavior data generated by processing an operation instruction input by a user from the intelligent terminal; the operation instruction comprises a touch instruction, a voice instruction or a facial expression instruction; the user behavior data includes the following four categories of behavior data: behavior parameters, generation time information corresponding to the behavior parameters, parameter category information corresponding to the behavior parameters and data format information corresponding to the intelligent terminal;
a vector extraction module, configured to extract a first feature vector corresponding to the user behavior data and second feature vectors corresponding to the behavior parameter, the generation time information, the parameter category information, and the data format information, respectively; the first feature vector is an N-dimensional vector, the second feature vector is an N + 1-dimensional vector, N is a positive integer, and a last one-dimensional feature value of the second feature vector is used for representing a behavior data category corresponding to the second feature vector;
a network determining module, configured to determine a first vector network corresponding to the first feature vector and a second vector network of each second feature vector; the first vector network and each second vector network comprise at least a plurality of network nodes with different network response weights, each network node corresponds to a vector value, the network response weights are convergence influence coefficients of the network nodes on the vector networks, the convergence influence coefficients are used for representing dimension expansion values of characteristic vectors corresponding to the vector networks, and the number of the network nodes in the first vector network and the number of the network nodes in each second vector network are both N;
a node determination module, configured to determine an initial network node of any network response weight of the first feature vector in the first vector network, and determine a network node having a largest network response weight in each second vector network as a target network node; acquiring node information similarity between the initial network node and each target network node, wherein the node information similarity is used for representing the matching degree of the initial network node and each target network node on the vector value level;
the vector dimension expansion module is used for determining the dimension expansion coefficient of each second feature vector relative to the first feature vector based on the information similarity of each node; carrying out dimension expansion processing on the first feature vector according to all the determined dimension expansion coefficients to obtain a target feature vector with dimensions of N + M, wherein M is a positive integer;
and the data pushing module is used for determining at least one expected data from a preset database according to the target characteristic vector and pushing the expected data to the intelligent terminal.
According to an aspect of the present application, there is provided an electronic device including: the system comprises a processor, a memory and a network interface, wherein the memory and the network interface are connected with the processor; the network interface is connected with a nonvolatile memory in the access right verification device; when the processor is operated, the computer program is called from the nonvolatile memory through the network interface, and the computer program is operated through the memory so as to execute the method.
According to one aspect of the present application, a readable storage medium applied to a computer is provided, and the readable storage medium is burned with a computer program, and the computer program realizes the method when running in a memory of an electronic device.
According to one aspect of the application, a data pushing system based on cloud computing is provided, and the system comprises an electronic device and an intelligent terminal which are communicated with each other;
the intelligent terminal is used for processing an operation instruction input by a user to generate user behavior data; the operation instruction comprises a touch instruction, a voice instruction or a facial expression instruction; the user behavior data includes the following four categories of behavior data: behavior parameters, generation time information corresponding to the behavior parameters, parameter category information corresponding to the behavior parameters and data format information corresponding to the intelligent terminal;
the electronic equipment is used for collecting the user behavior data from the intelligent terminal;
the electronic device is configured to extract a first feature vector corresponding to the user behavior data and second feature vectors corresponding to the behavior parameter, the generation time information, the parameter category information, and the data format information, respectively; the first feature vector is an N-dimensional vector, the second feature vector is an N + 1-dimensional vector, N is a positive integer, and a last one-dimensional feature value of the second feature vector is used for representing a behavior data category corresponding to the second feature vector;
the electronic device is configured to determine a first vector network corresponding to the first feature vector and a second vector network for each second feature vector; the first vector network and each second vector network comprise at least a plurality of network nodes with different network response weights, each network node corresponds to a vector value, the network response weights are convergence influence coefficients of the network nodes on the vector networks, the convergence influence coefficients are used for representing dimension expansion values of characteristic vectors corresponding to the vector networks, and the number of the network nodes in the first vector network and the number of the network nodes in each second vector network are both N;
the electronic device is configured to determine an initial network node of any network response weight of the first feature vector in the first vector network, and determine a network node having a largest network response weight in each second vector network as a target network node; acquiring node information similarity between the initial network node and each target network node, wherein the node information similarity is used for representing the matching degree of the initial network node and each target network node on the vector value level;
the electronic equipment is used for determining the dimension expansion coefficient of each second feature vector relative to the first feature vector based on the information similarity of each node; carrying out dimension expansion processing on the first feature vector according to all the determined dimension expansion coefficients to obtain a target feature vector with dimensions of N + M, wherein M is a positive integer;
the electronic equipment is used for determining at least one expected data from a preset database according to the target characteristic vector and pushing the expected data to the intelligent terminal.
When the data pushing method, the electronic equipment and the system based on the cloud computing are applied, additional feature dimensions can be further mined on the basis of the first feature vector of the user behavior data, the efficiency and the comprehensiveness of feature analysis on the user behavior data are improved, and the feature dimensions are prevented from being omitted. Therefore, the target characteristic vector corresponding to the user behavior data can be accurately and completely determined, the expected data can be accurately determined and pushed based on the target characteristic vector, the pushing accuracy of the cloud server in the data pushing process is improved, and the data is prevented from being pushed by mistake.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the application.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and together with the description, serve to explain the principles of the application.
Fig. 1 is a flowchart illustrating a data pushing method based on cloud computing according to an exemplary embodiment of the present application.
FIG. 2 is a block diagram of one embodiment of an apparatus shown in the present application according to an exemplary embodiment.
Fig. 3 is a hardware structure diagram of an electronic device in which the apparatus of the present application is located.
Detailed Description
Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present application, as detailed in the appended claims.
The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. As used in this application and the appended claims, the singular forms "a", "an", and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.
It is to be understood that although the terms first, second, third, etc. may be used herein to describe various information, such information should not be limited to these terms. These terms are only used to distinguish one type of information from another. For example, first information may also be referred to as second information, and similarly, second information may also be referred to as first information, without departing from the scope of the present application. The word "if" as used herein may be interpreted as "at … …" or "when … …" or "in response to a determination", depending on the context.
In order to solve the technical problem that the pushing accuracy is low or the pushing is mistaken in the data pushing process of the cloud server, the invention provides a data pushing method, electronic equipment and system based on cloud computing, which can perform multi-dimensional feature analysis on user behavior data corresponding to an intelligent terminal, further mine additional features on the basis of the multi-dimensional features of the user behavior data, improve the efficiency of feature analysis on the user behavior data, and avoid omission of feature dimensions. Therefore, the data characteristics corresponding to the user behavior data can be accurately and completely determined, the expected data can be accurately determined based on the data characteristics and pushed, the pushing accuracy of the cloud server in the data pushing process is improved, and the data is prevented from being pushed by mistake.
Referring to fig. 1, a flowchart of a data pushing method based on cloud computing according to the present invention is shown, where the data pushing method can be applied to an electronic device communicating with an intelligent terminal. Specifically, the electronic device may be understood as a cloud server, and the smart terminal includes, but is not limited to, a mobile phone, a tablet computer, a desktop computer, and the like. It can be understood that the electronic device and the intelligent terminal can be in one-to-one communication interaction or in one-to-many communication interaction. Further, the data pushing method may specifically include the following steps.
Step S21, collecting user behavior data generated by processing an operation instruction input by a user from an intelligent terminal; the operation instruction comprises a touch instruction, a voice instruction or a facial expression instruction; the user behavior data includes the following four categories of behavior data: the intelligent terminal comprises behavior parameters, generation time information corresponding to the behavior parameters, parameter category information corresponding to the behavior parameters and data format information corresponding to the intelligent terminal.
Step S22, extracting a first feature vector corresponding to the user behavior data and second feature vectors corresponding to the behavior parameter, the generation time information, the parameter category information, and the data format information, respectively; the first feature vector is an N-dimensional vector, the second feature vector is an N + 1-dimensional vector, N is a positive integer, and a last one-dimensional feature value of the second feature vector is used for representing a behavior data category corresponding to the second feature vector.
In particular implementation, in order to ensure the accuracy of subsequent feature analysis, when the first feature vector and the plurality of second feature vectors are extracted, the vector value dimensions of the first feature vector and the second feature vectors for characterizing data information may be set to be the same.
Step S23, determining a first vector network corresponding to the first feature vector and a second vector network for each second feature vector; the first vector network and each second vector network comprise at least a plurality of network nodes with different network response weights, each network node corresponds to a vector value, the network response weights are convergence influence coefficients of the network nodes on the vector networks, the convergence influence coefficients are used for representing dimension expansion values of characteristic vectors corresponding to the vector networks, and the number of the network nodes in the first vector network and the number of the network nodes in each second vector network are N.
Step S24, determining an initial network node of any network response weight of the first feature vector in the first vector network, and determining a network node with the maximum network response weight in each second vector network as a target network node; and acquiring node information similarity between the initial network node and each target network node, wherein the node information similarity is used for representing the matching degree of the initial network node and each target network node on the vector value level.
Step S25, determining the dimension expansion coefficient of each second feature vector relative to the first feature vector based on the similarity of each node information; and carrying out dimension expansion processing on the first feature vector according to all the determined dimension expansion coefficients to obtain a target feature vector with dimension of N + M, wherein M is a positive integer.
Step S26, determining at least one expected data from a preset database according to the target characteristic vector, and pushing the expected data to the intelligent terminal.
In specific implementation, according to the technical scheme, firstly, the first characteristic vector and the second characteristic vector are extracted from the collected user behavior data. And secondly, determining a first vector network corresponding to the first feature vector and a second vector network corresponding to the second feature vector. And then, carrying out network node level analysis on the first vector network and the second vector network to obtain node information similarity. And further determining a dimension expansion coefficient based on the node information similarity so as to realize dimension expansion processing of the first feature vector.
Therefore, additional feature dimensions can be further mined on the basis of the first feature vector of the user behavior data, the efficiency and comprehensiveness of feature analysis on the user behavior data are improved, and the feature dimensions are prevented from being omitted. Therefore, the target characteristic vector corresponding to the user behavior data can be accurately and completely determined, the expected data can be accurately determined and pushed based on the target characteristic vector, the pushing accuracy of the cloud server in the data pushing process is improved, and the data is prevented from being pushed by mistake.
In one possible example, in step S22, a first feature vector corresponding to the user behavior data is extracted, which may be implemented by the method described in the following steps.
Step S221, sequentially listing a plurality of data fields of the user behavior data according to a sequence of field lengths from large to small to obtain a data field sequence corresponding to the user behavior data; determining a field length difference value between two adjacent data fields in the data field sequence aiming at the data field sequence; determining the sequence distribution characteristics of the data field sequence according to the determined length difference values of all the fields; the data fields are divided according to time intervals, and the sequence distribution characteristics are used for representing the field length distribution condition of the data field sequence.
Step S222, extracting a sequence distribution value used for representing a field distribution discrete degree of the data field in the sequence distribution feature, and generating a data field trajectory corresponding to the data field sequence based on the sequence distribution value, where the data field trajectory includes a plurality of trajectory nodes, the trajectory nodes correspond to the data field one to one, each trajectory node is connected to at least one trajectory node except the trajectory node in the data field trajectory, a relevance weight exists between two interconnected trajectory nodes, the relevance weight has a priority from large to small, and the priority is used for representing an influence factor between two interconnected trajectory nodes.
And step S223, listing every two track nodes with mutual connection relation according to the sequence of the priority from low to high to obtain a track node sequence, and removing repeated track nodes in the track node sequence to obtain a target track node sequence.
Step S224, for each target track node in the target track node sequence, mapping field information in a data field corresponding to the target track node to a preset coordinate plane according to a preset mapping relationship to obtain a mapping coordinate value, and determining a characteristic value corresponding to each data field based on the mapping coordinate value; and sequencing the characteristic values according to the data field sequence to obtain a first characteristic vector corresponding to the user behavior data.
When the contents described in steps S221 to S224 are applied, the behavior parameters, the generation time information, the parameter category information, and the data format information in the user behavior data can be divided in the form of data fields, and then the data fields are analyzed to determine the first feature vector. Therefore, the difference between different types of behavior data can be reduced, and the first feature vector can be accurately determined from the global perspective.
In another possible example, the implementation principle of extracting the second feature vectors corresponding to the behavior parameters, the generation time information, the parameter category information, and the data format information is similar. For the simplicity of the following description, the second feature vector corresponding to the behavior parameter is extracted for detailed description.
In particular implementation, in order to ensure the accuracy and the discrimination of the second feature vector, the following processing may be performed on the behavior parameter to obtain a corresponding second feature vector.
(1) And determining at least a plurality of execution functions corresponding to the behavior parameters, wherein the execution functions are used for executing the operation instruction of the user and outputting the behavior parameters corresponding to the operation instruction.
(2) Determining input information of each execution function according to the operation instruction and determining output information of each execution function according to the behavior parameters; and determining execution logic information corresponding to each execution function based on each input information and the corresponding output information.
(3) And judging whether a call log exists in the execution logic information or not aiming at each execution logic information, wherein the call log is an operation log of the electronic equipment for reflecting and calling a hook function according to a first function type of the execution function so as to process input information corresponding to the execution function through the hook function to obtain corresponding output information.
(4) When the call log exists in the execution logic information, determining a second function type of the hook function corresponding to the logic execution information; determining a feature extraction list of the behavior parameters according to at least part of the first function types and at least part of the second function types corresponding to the behavior parameters; and extracting the features of the behavior parameters based on the feature extraction list, and obtaining second feature vectors corresponding to the behavior parameters based on the behavior data categories corresponding to the behavior parameters.
It can be understood that, through the above, on one hand, the accuracy of the second feature vector can be ensured through the feature extraction list, and on the other hand, the degree of distinction of the second feature vector can be ensured through the behavior data category corresponding to the behavior parameter.
In a more specific embodiment, in step S24, the obtaining of the node information similarity between the initial network node and each target network node may specifically include what is described in the following substeps.
Step S241, determining a first information list corresponding to the node information of the initial network node and a second information list corresponding to the node information of each target network node; the first information list and the second information list have the same row number and column number, the list units of the first information list and the second information list have the same number, the first list units in the first information list have different first unit weights, and the second list units in the second information list have different second unit weights.
Step S242, comparing each first list unit in the first information list with a corresponding second list unit in the second information list one by one, to obtain a comparison result; when the comparison result represents that the information in the first list unit is the same as the information in the second list unit, determining the comparison similarity of the comparison result as a first set numerical value; when the comparison result represents that the information in the first list unit is different from the information in the second list unit, weighting a second set numerical value based on a first unit weight corresponding to the first list unit and a second unit weight corresponding to the second list unit to obtain a comparison similarity corresponding to the comparison result; and determining the mean value of the comparison similarity as the node information similarity between the initial network and each target network node.
In step S242, the first setting value may be 1, the second setting value may be 0.5, and when the second setting value is weighted, the second setting value may be increased or decreased according to the first cell weight and the second cell weight, which is not limited herein.
It can be understood that based on the content described in the above steps, the node information similarity between the initial network node and the target network node can be determined from the information list level, so as to ensure the accuracy and reliability of the node information similarity.
In an alternative embodiment, the step of determining the dimension expansion coefficient of each second feature vector relative to the first feature vector based on the similarity of each node information described in step S25 may be specifically implemented by the following steps.
Step S2511, determining a similarity difference between the information similarities of every two adjacent nodes and a median of the information similarities of the nodes.
Step S2512, determining an accumulated value of similarity differences falling in a set interval with the median as an interval midpoint and the average value of the similarities of the node information as an interval length.
Step S2513, judging whether the accumulated value reaches a set value; if so, determining the dimension expansion coefficient of the second feature vector corresponding to the node information similarity relative to the first feature vector according to the difference value of the node information similarity and the median; and if not, determining the dimension expansion coefficient of the second feature vector corresponding to the node information similarity relative to the first feature vector according to the difference value of the similarity of each node information similarity and the similarity average value.
It can be understood that, based on the contents described in steps S2511 to S2513, the difference, median and average between the node similarity information can be analyzed, so as to determine the dimension expansion coefficient under different conditions and ensure the accuracy of the dimension expansion coefficient.
Further, on the basis, the step S25 of performing the dimension expansion processing on the first feature vector according to all the determined dimension expansion coefficients to obtain the target feature vector with the dimension of N + M may specifically include the following sub-steps.
Step S2521, determining a dimension reference value for performing dimension expansion processing on the first feature vector according to all the dimension expansion coefficients.
Step S2522, the dimension reference value is corrected according to the vector dimension and the vector confidence of the first feature vector to obtain a target dimension value.
Step S2523, the first feature vector is subjected to dimension expansion processing according to the target dimension value M to obtain an N + M-dimensional target feature vector.
In this embodiment, the target dimension value may be M. The dimension reference value may be a positive integer smaller than M, or may be a positive integer larger than M, and is not limited herein.
Optionally, in step S26, the determining at least one desired datum from a preset database according to the target feature vector may specifically include the following: and determining expected data corresponding to at least one characteristic vector of which the cosine similarity with the target characteristic vector is greater than a preset threshold value from the preset database.
In an alternative embodiment, to further ensure the accuracy of data pushing, the data in the database of the electronic device needs to be updated in real time, and for this purpose, the electronic device may also obtain the latest data from the outside and store the latest data in the database.
In order to ensure that the latest data can be completely, accurately and quickly stored in the database when the electronic device is implemented, the electronic device stores the latest data in the database, and the following substeps may be specifically included.
Step S31, extracting the data characteristic vector and each data segment of the latest data; wherein different data segments correspond to different data information.
Step S32, when it is determined that the latest data includes the data structure type according to the data feature vector, determining a data difference coefficient between each data segment of the latest data in the data information type and each data segment of the latest data in the data structure type based on the data segment of the latest data in the data structure type and the data segment weight thereof.
Step S33, adjusting the data segment with the smallest data difference coefficient between the data segment of the latest data in the data information category and the data segment of the latest data in the data structure category to the data structure category of the latest data.
Step S34, when the data information category of the latest data includes a plurality of data segments, determining a data difference coefficient between the data segments of the latest data in the data structure category based on the data segments of the latest data in the data structure category and the data segment weight thereof, and performing data conversion on each data segment in the data information category based on the data difference coefficient between each data segment to obtain a target data segment corresponding to each data segment in the data information category.
Step S35, encapsulating a paragraph pointing parameter for each target data segment based on the data segment of the latest data in the data structure category and the data segment weight thereof, and transferring each target data segment to a sub-category of the data structure category corresponding to the paragraph pointing parameter.
Step S36, determining a first data structure characteristic of the latest data based on each first data segment in the data structure category; acquiring storage thread information of the database and determining a second data structure characteristic corresponding to the database based on the storage thread information; determining a cosine distance between the first data structure feature and the second data structure feature; and converting the data format of the latest data according to the cosine distance and storing the latest data into the database.
It is understood that through the descriptions of the above steps S31-S36, the corresponding second data structure characteristic of the database can be determined through the corresponding storage thread information of the database, and the second data structure characteristic can be understood as the data structure characteristic of the data in the database. Furthermore, the data feature vector and each data segment of the latest data can be analyzed, and the data information category and the data structure category are taken into account, so that the first data structure feature of the latest data can be accurately determined. In this way, the latest data can be stored after being subjected to data format conversion based on the cosine distance between the first data structure characteristic and the second data structure characteristic. In this way, the latest data can be completely, accurately and quickly stored in the database.
The various technical features in the above embodiments can be arbitrarily combined, so long as there is no conflict or contradiction between the combinations of the features, but the combination is limited by the space and is not described one by one, and therefore, any combination of the various technical features in the above embodiments also belongs to the scope disclosed in the present specification.
Corresponding to the embodiment of the data pushing method based on cloud computing, the present application also provides embodiments of the data pushing apparatus 200 and the electronic device 400 based on cloud computing.
Referring to fig. 2, a block diagram of functional modules of the data pushing apparatus 200 is shown, which specifically includes the following functional modules.
The data acquisition module 201 is used for acquiring user behavior data generated by processing an operation instruction input by a user from an intelligent terminal; the operation instruction comprises a touch instruction, a voice instruction or a facial expression instruction; the user behavior data includes the following four categories of behavior data: the intelligent terminal comprises behavior parameters, generation time information corresponding to the behavior parameters, parameter category information corresponding to the behavior parameters and data format information corresponding to the intelligent terminal.
A vector extraction module 202, configured to extract a first feature vector corresponding to the user behavior data and second feature vectors corresponding to the behavior parameter, the generation time information, the parameter category information, and the data format information, respectively; the first feature vector is an N-dimensional vector, the second feature vector is an N + 1-dimensional vector, N is a positive integer, and a last one-dimensional feature value of the second feature vector is used for representing a behavior data category corresponding to the second feature vector.
A network determining module 203, configured to determine a first vector network corresponding to the first feature vector and a second vector network of each second feature vector; the first vector network and each second vector network comprise at least a plurality of network nodes with different network response weights, each network node corresponds to a vector value, the network response weights are convergence influence coefficients of the network nodes on the vector networks, the convergence influence coefficients are used for representing dimension expansion values of characteristic vectors corresponding to the vector networks, and the number of the network nodes in the first vector network and the number of the network nodes in each second vector network are N.
A node determining module 204, configured to determine an initial network node of any network response weight of the first feature vector in the first vector network, and determine a network node with a largest network response weight in each second vector network as a target network node; and acquiring node information similarity between the initial network node and each target network node, wherein the node information similarity is used for representing the matching degree of the initial network node and each target network node on the vector value level.
The vector dimension expanding module 205 is configured to determine a dimension expanding coefficient of each second feature vector relative to the first feature vector based on each node information similarity; and carrying out dimension expansion processing on the first feature vector according to all the determined dimension expansion coefficients to obtain a target feature vector with dimension of N + M, wherein M is a positive integer.
And the data pushing module 206 is configured to determine at least one expected data from a preset database according to the target feature vector, and push the expected data to the intelligent terminal.
The embodiment of the data pushing apparatus 200 based on cloud computing can be applied to the electronic device 400. The embodiments of the apparatus may be implemented by software, or by hardware, or by a combination of hardware and software. The software implementation is taken as an example, and as a logical device, the processor of the electronic device reads corresponding computer program instructions in the nonvolatile memory to the memory based on the network interface to run. From a hardware aspect, as shown in fig. 3, a hardware structure diagram of an electronic device 400 in which the data pushing apparatus 200 based on cloud computing is located according to the present application is shown, except for the processor 410, the memory 430, the network interface 440, and the nonvolatile memory 420 shown in fig. 3, the device in which the apparatus is located in the embodiment may further include other hardware according to an actual function of the device, which is not shown in fig. 3 one by one.
The implementation process of the functions and actions of each module in the above device is specifically described in the implementation process of the corresponding step in the above method, and is not described herein again.
On the basis, a readable storage medium applied to a computer is further provided, and a computer program is burned on the readable storage medium and realizes the method when the computer program runs in a memory of the electronic device.
On the basis, the data pushing system based on the cloud computing is further provided, and the system comprises electronic equipment and an intelligent terminal which are communicated with each other.
The intelligent terminal is used for processing an operation instruction input by a user to generate user behavior data; the operation instruction comprises a touch instruction, a voice instruction or a facial expression instruction; the user behavior data includes the following four categories of behavior data: the intelligent terminal comprises behavior parameters, generation time information corresponding to the behavior parameters, parameter category information corresponding to the behavior parameters and data format information corresponding to the intelligent terminal.
The electronic equipment is used for collecting the user behavior data from the intelligent terminal.
The electronic device is configured to extract a first feature vector corresponding to the user behavior data and second feature vectors corresponding to the behavior parameter, the generation time information, the parameter category information, and the data format information, respectively; the first feature vector is an N-dimensional vector, the second feature vector is an N + 1-dimensional vector, N is a positive integer, and a last one-dimensional feature value of the second feature vector is used for representing a behavior data category corresponding to the second feature vector.
The electronic device is configured to determine a first vector network corresponding to the first feature vector and a second vector network for each second feature vector; the first vector network and each second vector network comprise at least a plurality of network nodes with different network response weights, each network node corresponds to a vector value, the network response weights are convergence influence coefficients of the network nodes on the vector networks, the convergence influence coefficients are used for representing dimension expansion values of characteristic vectors corresponding to the vector networks, and the number of the network nodes in the first vector network and the number of the network nodes in each second vector network are N.
The electronic device is configured to determine an initial network node of any network response weight of the first feature vector in the first vector network, and determine a network node having a largest network response weight in each second vector network as a target network node; and acquiring node information similarity between the initial network node and each target network node, wherein the node information similarity is used for representing the matching degree of the initial network node and each target network node on the vector value level.
The electronic equipment is used for determining the dimension expansion coefficient of each second feature vector relative to the first feature vector based on the information similarity of each node; and carrying out dimension expansion processing on the first feature vector according to all the determined dimension expansion coefficients to obtain a target feature vector with dimension of N + M, wherein M is a positive integer.
The electronic equipment is used for determining at least one expected data from a preset database according to the target characteristic vector and pushing the expected data to the intelligent terminal.
For the device embodiments, since they substantially correspond to the method embodiments, reference may be made to the partial description of the method embodiments for relevant points. The above-described embodiments of the apparatus are merely illustrative, wherein the modules described as separate parts may or may not be physically separate, and the parts displayed as modules may or may not be physical modules, may be located in one place, or may be distributed on a plurality of network modules. Some or all of the modules can be selected according to actual needs to achieve the purpose of the scheme of the application. One of ordinary skill in the art can understand and implement it without inventive effort.
Other embodiments of the present application will be apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the invention following, in general, the principles of the application and including such departures from the present disclosure as come within known or customary practice within the art to which the invention pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the application being indicated by the following claims.
It will be understood that the present application is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the application is limited only by the appended claims.
Claims (8)
1. A data pushing method based on cloud computing is applied to an electronic device, and comprises the following steps:
acquiring user behavior data generated by processing an operation instruction input by a user through the intelligent terminal from the intelligent terminal; the operation instruction comprises a touch instruction, a voice instruction or a facial expression instruction; the user behavior data includes the following four categories of behavior data: behavior parameters, generation time information corresponding to the behavior parameters, parameter category information corresponding to the behavior parameters and data format information corresponding to the intelligent terminal;
extracting a first feature vector corresponding to the user behavior data and second feature vectors corresponding to the behavior parameters, the generation time information, the parameter category information and the data format information respectively; the first feature vector is an N-dimensional vector, the second feature vector is an N + 1-dimensional vector, N is a positive integer, and a last one-dimensional feature value of the second feature vector is used for representing a behavior data category corresponding to the second feature vector;
determining a first vector network corresponding to the first feature vector and a second vector network of each second feature vector; the first vector network and each second vector network comprise at least a plurality of network nodes with different network response weights, each network node corresponds to a vector value, the network response weights are convergence influence coefficients of the network nodes on the vector networks, the convergence influence coefficients are used for representing dimension expansion values of characteristic vectors corresponding to the vector networks, and the number of the network nodes in the first vector network and the number of the network nodes in each second vector network are both N;
determining an initial network node of any network response weight of the first feature vector in the first vector network, and determining a network node with the largest network response weight in each second vector network as a target network node; acquiring node information similarity between the initial network node and each target network node, wherein the node information similarity is used for representing the matching degree of the initial network node and each target network node on the vector value level;
determining a dimension expansion coefficient of each second feature vector relative to the first feature vector based on the similarity of each node information; carrying out dimension expansion processing on the first feature vector according to all the determined dimension expansion coefficients to obtain a target feature vector with dimensions of N + M, wherein M is a positive integer;
determining at least one expected data from a preset database according to the target characteristic vector, and pushing the expected data to the intelligent terminal;
wherein the method further comprises:
and acquiring the latest data from the outside and storing the latest data in the database.
2. The data pushing method according to claim 1, wherein the extracting a first feature vector corresponding to the user behavior data includes:
sequentially listing a plurality of data fields of the user behavior data according to a sequence of field lengths from large to small to obtain a data field sequence corresponding to the user behavior data; determining a field length difference value between two adjacent data fields in the data field sequence aiming at the data field sequence; determining the sequence distribution characteristics of the data field sequence according to the determined length difference values of all the fields; the data field is divided according to time intervals, and the sequence distribution characteristics are used for representing the field length distribution condition of the data field sequence;
extracting a sequence distribution value used for representing field distribution discrete degree of the data field in the sequence distribution characteristics, and generating a data field track corresponding to the data field sequence based on the sequence distribution value, wherein the data field track comprises a plurality of track nodes, the track nodes correspond to the data field one by one, each track node is connected with at least one track node except the track node in the data field track, a relevance weight exists between two mutually connected track nodes, the relevance weight has a priority from large to small, and the priority is used for representing an influence factor between the two mutually connected track nodes;
listing every two track nodes with interconnection relation according to the sequence of the priority from low to high to obtain a track node sequence, and removing repeated track nodes in the track node sequence to obtain a target track node sequence;
for each target track node in the target track node sequence, mapping field information in a data field corresponding to the target track node to a preset coordinate plane according to a preset mapping relation to obtain a mapping coordinate value, and determining a characteristic value corresponding to each data field based on the mapping coordinate value; and sequencing the characteristic values according to the data field sequence to obtain a first characteristic vector corresponding to the user behavior data.
3. The data pushing method according to claim 2, wherein the extracting a second feature vector corresponding to the behavior parameter includes:
determining at least a plurality of execution functions corresponding to the behavior parameters, wherein the execution functions are used for executing the operation instruction of the user and outputting the behavior parameters corresponding to the operation instruction;
determining input information of each execution function according to the operation instruction and determining output information of each execution function according to the behavior parameters; determining execution logic information corresponding to each execution function based on each input information and the corresponding output information;
for each piece of execution logic information, judging whether a call log exists in the piece of execution logic information, wherein the call log is an operation log of the electronic equipment for reflecting and calling a hook function according to a first function type of the execution function so as to process input information corresponding to the execution function through the hook function to obtain corresponding output information;
when the call log exists in the execution logic information, determining a second function type of the hook function corresponding to the logic execution information; determining a feature extraction list of the behavior parameters according to at least part of the first function types and at least part of the second function types corresponding to the behavior parameters; and extracting the features of the behavior parameters based on the feature extraction list, and obtaining second feature vectors corresponding to the behavior parameters based on the behavior data categories corresponding to the behavior parameters.
4. The data pushing method according to any one of claims 1 to 3, wherein the obtaining of the node information similarity between the initial network node and each target network node comprises:
determining a first information list corresponding to the node information of the initial network node and a second information list corresponding to the node information of each target network node; the first information list and the second information list have the same row number and column number, the list units of the first information list and the second information list have the same number, the first list units in the first information list have different first unit weights, and the second list units in the second information list have different second unit weights;
comparing each first list unit in the first information list with a corresponding second list unit in the second information list one by one to obtain a comparison result; when the comparison result represents that the information in the first list unit is the same as the information in the second list unit, determining the comparison similarity of the comparison result as a first set numerical value; when the comparison result represents that the information in the first list unit is different from the information in the second list unit, weighting a second set numerical value based on a first unit weight corresponding to the first list unit and a second unit weight corresponding to the second list unit to obtain a comparison similarity corresponding to the comparison result; and determining the mean value of the comparison similarity as the node information similarity between the initial network and each target network node.
5. The data pushing method according to claim 1, wherein the determining a dimension expansion coefficient of each second feature vector relative to the first feature vector based on each node information similarity comprises:
determining a similarity difference value between every two adjacent node information similarities and a median of the node information similarities;
determining an accumulated value of similarity difference values falling into a set interval which takes the median as an interval midpoint and takes the similarity average value of the node information similarity as an interval length;
judging whether the accumulated value reaches a set value; if so, determining the dimension expansion coefficient of the second feature vector corresponding to the node information similarity relative to the first feature vector according to the difference value of the node information similarity and the median; and if not, determining the dimension expansion coefficient of the second feature vector corresponding to the node information similarity relative to the first feature vector according to the difference value of the similarity of each node information similarity and the similarity average value.
6. The data pushing method according to claim 5, wherein the performing the dimension expansion processing on the first feature vector according to all the determined dimension expansion coefficients to obtain an N + M-dimensional target feature vector comprises:
determining a dimension reference value for performing dimension expansion processing on the first feature vector according to all the dimension expansion coefficients;
correcting the dimension reference value according to the vector dimension and the vector confidence of the first feature vector to obtain a target dimension value;
and carrying out dimension expansion processing on the first feature vector according to the target dimension value M to obtain an N + M-dimensional target feature vector.
7. An electronic device, comprising:
a processor, and
a memory and a network interface connected with the processor;
the network interface is connected with a nonvolatile memory in the access right verification device;
the processor, when running, retrieves a computer program from the non-volatile memory via the network interface and runs the computer program via the memory to perform the method of any of claims 1-6 above.
8. The data pushing system based on cloud computing is characterized by comprising electronic equipment and an intelligent terminal which are communicated with each other;
the intelligent terminal is used for processing an operation instruction input by a user to generate user behavior data; the operation instruction comprises a touch instruction, a voice instruction or a facial expression instruction; the user behavior data includes the following four categories of behavior data: behavior parameters, generation time information corresponding to the behavior parameters, parameter category information corresponding to the behavior parameters and data format information corresponding to the intelligent terminal;
the electronic equipment is used for collecting the user behavior data from the intelligent terminal;
the electronic device is configured to extract a first feature vector corresponding to the user behavior data and second feature vectors corresponding to the behavior parameter, the generation time information, the parameter category information, and the data format information, respectively; the first feature vector is an N-dimensional vector, the second feature vector is an N + 1-dimensional vector, N is a positive integer, and a last one-dimensional feature value of the second feature vector is used for representing a behavior data category corresponding to the second feature vector;
the electronic device is configured to determine a first vector network corresponding to the first feature vector and a second vector network for each second feature vector; the first vector network and each second vector network comprise at least a plurality of network nodes with different network response weights, each network node corresponds to a vector value, the network response weights are convergence influence coefficients of the network nodes on the vector networks, the convergence influence coefficients are used for representing dimension expansion values of characteristic vectors corresponding to the vector networks, and the number of the network nodes in the first vector network and the number of the network nodes in each second vector network are both N;
the electronic device is configured to determine an initial network node of any network response weight of the first feature vector in the first vector network, and determine a network node having a largest network response weight in each second vector network as a target network node; acquiring node information similarity between the initial network node and each target network node, wherein the node information similarity is used for representing the matching degree of the initial network node and each target network node on the vector value level;
the electronic equipment is used for determining the dimension expansion coefficient of each second feature vector relative to the first feature vector based on the information similarity of each node; carrying out dimension expansion processing on the first feature vector according to all the determined dimension expansion coefficients to obtain a target feature vector with dimensions of N + M, wherein M is a positive integer;
the electronic equipment is used for determining at least one expected data from a preset database according to the target characteristic vector and pushing the expected data to the intelligent terminal;
wherein the electronic device is further configured to:
and acquiring the latest data from the outside and storing the latest data in the database.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011182629.4A CN112286977A (en) | 2020-03-27 | 2020-03-27 | Data pushing method, electronic equipment and system based on cloud computing |
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010226279.0A CN111432003B (en) | 2020-03-27 | 2020-03-27 | Data pushing method and device applied to cloud computing, electronic equipment and system |
| CN202011182629.4A CN112286977A (en) | 2020-03-27 | 2020-03-27 | Data pushing method, electronic equipment and system based on cloud computing |
Related Parent Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010226279.0A Division CN111432003B (en) | 2020-03-27 | 2020-03-27 | Data pushing method and device applied to cloud computing, electronic equipment and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112286977A true CN112286977A (en) | 2021-01-29 |
Family
ID=71548888
Family Applications (3)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011182630.7A Withdrawn CN112286978A (en) | 2020-03-27 | 2020-03-27 | Data pushing method based on cloud computing and electronic equipment |
| CN202011182629.4A Withdrawn CN112286977A (en) | 2020-03-27 | 2020-03-27 | Data pushing method, electronic equipment and system based on cloud computing |
| CN202010226279.0A Expired - Fee Related CN111432003B (en) | 2020-03-27 | 2020-03-27 | Data pushing method and device applied to cloud computing, electronic equipment and system |
Family Applications Before (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011182630.7A Withdrawn CN112286978A (en) | 2020-03-27 | 2020-03-27 | Data pushing method based on cloud computing and electronic equipment |
Family Applications After (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010226279.0A Expired - Fee Related CN111432003B (en) | 2020-03-27 | 2020-03-27 | Data pushing method and device applied to cloud computing, electronic equipment and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (3) | CN112286978A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115860836A (en) * | 2022-12-07 | 2023-03-28 | 广东南粤分享汇控股有限公司 | E-commerce service pushing method and system based on user behavior big data analysis |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112488247A (en) * | 2020-08-06 | 2021-03-12 | 蔡淦祺 | Information processing method combining live webcasting and online e-commerce delivery and cloud server |
| CN111949720B (en) * | 2020-08-24 | 2021-07-06 | 南京硅基智能科技有限公司 | Data analysis method based on big data and artificial intelligence and cloud data server |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102929928B (en) * | 2012-09-21 | 2015-04-22 | 北京格致璞科技有限公司 | Multidimensional-similarity-based personalized news recommendation method |
| US11250347B2 (en) * | 2018-06-27 | 2022-02-15 | Microsoft Technology Licensing, Llc | Personalization enhanced recommendation models |
| CN109241440A (en) * | 2018-09-29 | 2019-01-18 | 北京工业大学 | It is a kind of based on deep learning towards implicit feedback recommended method |
| CN109582903B (en) * | 2018-11-29 | 2023-10-20 | 广州市百果园信息技术有限公司 | Information display method, device, equipment and storage medium |
| CN110399563A (en) * | 2019-08-05 | 2019-11-01 | 中国人民解放军军事科学院评估论证研究中心 | A kind of preference pattern and customized information active recommended method based on user behavior analysis |
-
2020
- 2020-03-27 CN CN202011182630.7A patent/CN112286978A/en not_active Withdrawn
- 2020-03-27 CN CN202011182629.4A patent/CN112286977A/en not_active Withdrawn
- 2020-03-27 CN CN202010226279.0A patent/CN111432003B/en not_active Expired - Fee Related
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115860836A (en) * | 2022-12-07 | 2023-03-28 | 广东南粤分享汇控股有限公司 | E-commerce service pushing method and system based on user behavior big data analysis |
| CN115860836B (en) * | 2022-12-07 | 2023-09-26 | 广东南粤分享汇控股有限公司 | E-commerce service pushing method and system based on user behavior big data analysis |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111432003B (en) | 2021-01-08 |
| CN112286978A (en) | 2021-01-29 |
| CN111432003A (en) | 2020-07-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111432003B (en) | Data pushing method and device applied to cloud computing, electronic equipment and system | |
| CN109634698B (en) | Menu display method and device, computer equipment and storage medium | |
| CN102214302A (en) | Recognition device, recognition method, and program | |
| CN109117854B (en) | Key point matching method and device, electronic equipment and storage medium | |
| CN109918498B (en) | Problem warehousing method and device | |
| CN104679818A (en) | Video keyframe extracting method and video keyframe extracting system | |
| CN111338581B (en) | Data storage method and device based on cloud computing, cloud server and system | |
| CN110728313A (en) | Classification model training method and device for intention classification recognition | |
| CN114781688A (en) | Method, device, equipment and storage medium for identifying abnormal data of business expansion project | |
| CN116361567B (en) | Data processing method and system applied to cloud office | |
| CN112257332B (en) | Simulation model evaluation method and device | |
| CN111949720B (en) | Data analysis method based on big data and artificial intelligence and cloud data server | |
| CN110377721B (en) | Automatic question answering method, device, storage medium and electronic equipment | |
| CN112765362A (en) | Knowledge graph entity alignment method based on improved self-encoder and related equipment | |
| CN112069269B (en) | Big data and multidimensional feature-based data tracing method and big data cloud server | |
| CN112784165A (en) | Training method of incidence relation estimation model and method for estimating file popularity | |
| CN113869398A (en) | Unbalanced text classification method, device, equipment and storage medium | |
| CN114268625A (en) | Feature selection method, device, equipment and storage medium | |
| CN111090859A (en) | Malicious software detection method based on graph editing distance | |
| CN117235137B (en) | Professional information query method and device based on vector database | |
| CN109905340B (en) | Feature optimization function selection method and device and electronic equipment | |
| CN112837040B (en) | Power data management method and system applied to smart grid | |
| KR20230019288A (en) | Spatial information analysis system based on artificial intelligence | |
| CN117371509A (en) | Classification method and device | |
| CN116932580A (en) | Radix estimation method and device |
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 | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20210129 |