CN112801718B - User behavior prediction method, device, equipment and medium - Google Patents
User behavior prediction method, device, equipment and medium Download PDFInfo
- Publication number
- CN112801718B CN112801718B CN202110199787.9A CN202110199787A CN112801718B CN 112801718 B CN112801718 B CN 112801718B CN 202110199787 A CN202110199787 A CN 202110199787A CN 112801718 B CN112801718 B CN 112801718B
- Authority
- CN
- China
- Prior art keywords
- data set
- behavior prediction
- user behavior
- target
- determining
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0202—Market predictions or forecasting for commercial activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0251—Targeted advertisements
- G06Q30/0255—Targeted advertisements based on user history
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Strategic Management (AREA)
- Finance (AREA)
- Development Economics (AREA)
- Accounting & Taxation (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Entrepreneurship & Innovation (AREA)
- Evolutionary Computation (AREA)
- General Business, Economics & Management (AREA)
- General Engineering & Computer Science (AREA)
- Marketing (AREA)
- Economics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Game Theory and Decision Science (AREA)
- Biophysics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computational Linguistics (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- General Health & Medical Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to the field of artificial intelligence, and provides a user behavior prediction method, a device, equipment and a medium, which can initialize the weights of a source domain data set and a target domain data set to obtain an initial weight, use a logistic regression classifier as a base model, and use a single model as a final classifier, so that the model can keep the interpretability of a linear model and is more suitable for actual services. In addition, the invention also relates to a block chain technology, and the target classifier can be stored in the block chain node.
Description
Technical Field
The invention relates to the technical field of artificial intelligence, in particular to a user behavior prediction method, device, equipment and medium.
Background
Customer purchasing behavior prediction can guide an enterprise to perform accurate marketing, for example: under the scenes of life insurance and the like, when new life insurance products, new region sales products are pushed out or projects are restarted, samples are lacked or the samples are old, at the moment, if the future purchasing behaviors of customers are required to be predicted to provide marketing support, the problem of establishing a business model from 0 to 1 is faced, however, certain data accumulation is required for establishing the model, and the lack of data is a great obstacle.
In the above situation, a common solution is to construct some strategies based on experience, or wait for data accumulation, or directly use existing models in similar scenarios to predict the purchasing behavior of customers.
However, the above method may affect the timeliness and accuracy of the business strategy, for example: when a mode of waiting for data accumulation is adopted, due to the fact that a certain time is needed for data accumulation, timeliness is affected; when the existing model under similar scenes is directly used for predicting the purchasing behavior of a customer, the accuracy of prediction is influenced because data between different scenes has certain difference.
Disclosure of Invention
In view of the above, it is necessary to provide a user behavior prediction method, device, apparatus, and medium, which can make training data more suitable for practical application scenarios by continuously iteratively adjusting the weights of a source domain data set and a target domain data set based on a migration learning method, so as to train to obtain an optimal model for prediction, thereby solving the problem of sample shortage, improving the effect of model training, and further realizing more accurate user behavior prediction.
A user behavior prediction method, comprising:
responding to a user behavior prediction instruction, and determining a source domain and a target domain according to the user behavior prediction instruction;
acquiring data of the source domain to construct a source domain data set, and acquiring data of the target domain to construct a target domain data set;
initializing the weights of the source domain data set and the target domain data set to obtain an initial weight;
performing iteration training of a preset number of rounds on a logistic regression classifier according to the source domain data set, the target domain data set and the initial weight, and storing a base classifier obtained by each round of iteration;
acquiring a test data set, calculating a KS value of each base classifier on the test data set, and determining the base classifier with the highest KS value as a target classifier;
and acquiring data to be processed, inputting the data to be processed into the target classifier, and determining a behavior prediction result according to the output of the target classifier.
According to a preferred embodiment of the present invention, the determining a source domain and a target domain according to the user behavior prediction instruction comprises:
analyzing a method body of the user behavior prediction instruction to obtain information carried by the user behavior prediction instruction;
acquiring a first preset label and a second preset label;
constructing a first regular expression according to the first preset label, and constructing a second regular expression according to the second preset label;
traversing in the information carried by the user behavior prediction instruction by using the first regular expression, and determining traversed data as a first identifier;
traversing in the information carried by the user behavior prediction instruction by using the second regular expression, and determining traversed data as a second identifier;
and determining the source domain according to the first identifier and determining the target domain according to the second identifier.
According to a preferred embodiment of the present invention, the formula of the initial weight is as follows:
wherein the content of the first and second substances,and the initial weight value of the ith sample is represented, n represents the number of samples in the source domain data set, m represents the number of samples in the target domain data set, and i is n + m.
According to a preferred embodiment of the invention, the method further comprises:
calculating a union set of the source domain data set and the target domain data set to obtain a sample set;
for each iteration training, configuring a first threshold for calculating the weight of the source domain data set and configuring a second threshold for calculating the weight of the target domain data set;
for each sample in the sample set, calculating a classification result based on a base classifier obtained by each iteration training;
acquiring an actual label of each sample, and acquiring the weight of the source domain data set and the target domain data set in each iteration training process;
updating the weights of the source domain data set and the target domain data set in each iteration training process according to the first threshold, the second threshold, the classification result and the actual label;
and executing the next round of iterative training by using the updated weight.
According to a preferred embodiment of the present invention, the formula of the first threshold is as follows:
wherein β represents the first threshold value, and N represents the preset number of rounds;
the formula of the second threshold is as follows:
wherein, betatRepresents the second threshold value, UtRepresenting the error rate produced on the target domain data set, the formula is as follows:
wherein the content of the first and second substances,represents the weight of the ith sample during the t iteration, ht(xi) Represents the classification result of the ith sample in the process of the t round iteration, c (x)i) The actual label of the ith sample is represented.
According to the preferred embodiment of the present invention, the calculating the classification result based on the base classifier obtained by each iteration training includes:
acquiring historical data and an actual label of each historical data;
identifying a label corresponding to the configuration behavior from the actual labels as a target label, and calculating the proportion of the target label in the historical data as a conversion rate;
determining a quantile point as a classification threshold value according to the conversion rate;
when the output of the base classifier is greater than or equal to the classification threshold, determining that the classification result is 1; or
When the output of the base classifier is less than the classification threshold, determining that the classification result is 0.
According to a preferred embodiment of the present invention, the formula of the updated weight is as follows:
wherein the content of the first and second substances,represents the weight of the ith sample after updating in the (t +1) th iteration process.
A user behavior prediction apparatus, the user behavior prediction apparatus comprising:
the determining unit is used for responding to a user behavior prediction instruction and determining a source domain and a target domain according to the user behavior prediction instruction;
the acquisition unit is used for acquiring data of the source domain to construct a source domain data set and acquiring data of the target domain to construct a target domain data set;
the initialization unit is used for initializing the weights of the source domain data set and the target domain data set to obtain an initial weight;
the training unit is used for performing iteration training of preset rounds on the logistic regression classifier according to the source domain data set, the target domain data set and the initial weight, and storing a base classifier obtained by each iteration;
the calculation unit is used for acquiring a test data set, calculating a KS value of each base classifier on the test data set, and determining the base classifier with the highest KS value as a target classifier;
the determining unit is further configured to acquire data to be processed, input the data to be processed to the target classifier, and determine a behavior prediction result according to an output of the target classifier.
An electronic device, the electronic device comprising:
a memory storing at least one instruction; and
a processor executing instructions stored in the memory to implement the user behavior prediction method.
A computer-readable storage medium having at least one instruction stored therein, the at least one instruction being executable by a processor in an electronic device to implement the user behavior prediction method.
According to the technical scheme, the method can respond to a user behavior prediction instruction, determine a source domain and a target domain according to the user behavior prediction instruction, acquire a data construction source domain data set of the source domain, acquire a data construction target domain data set of the target domain, initialize the weights of the source domain data set and the target domain data set to acquire an initial weight, firstly give a higher weight to the target domain data set to ensure that a classifier obtained by subsequent training has higher accuracy because the target domain data set is the same as data distribution under an actual service scene, perform iterative training of preset rounds on a logistic regression classifier according to the source domain data set, the target domain data set and the initial weight, and store a base classifier obtained by each round of iteration, which is different from a voting mode generally adopted in the prior art, the present embodiment uses a logistic regression classifier as the base model, a single model as the final classifier, so that the model can keep the interpretability of the linear model, is more suitable for actual business, acquires a test data set, calculating the KS value of each base classifier on the test data set, determining the base classifier with the highest KS value as a target classifier, acquiring data to be processed, inputting the data to be processed to the target classifier, and determining a behavior prediction result from the output of the target classifier, based on a method of transfer learning, by continuously and iteratively adjusting the weights of the source domain data set and the target domain data set, the training data is more suitable for the actual application scene, and then training to obtain an optimal model for prediction, so that the problem of sample shortage is solved, the model training effect is improved, and more accurate user behavior prediction is realized.
Drawings
FIG. 1 is a flow chart of a preferred embodiment of a user behavior prediction method of the present invention.
Fig. 2 is a functional block diagram of a user behavior prediction apparatus according to a preferred embodiment of the present invention.
Fig. 3 is a schematic structural diagram of an electronic device implementing a method for predicting user behavior according to a preferred embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be described in detail with reference to the accompanying drawings and specific embodiments.
Fig. 1 is a flow chart of a user behavior prediction method according to a preferred embodiment of the present invention. The order of the steps in the flow chart may be changed and some steps may be omitted according to different needs.
The user behavior prediction method is applied to one or more electronic devices, which are devices capable of automatically performing numerical calculation and/or information processing according to preset or stored instructions, and hardware of the electronic devices includes, but is not limited to, a microprocessor, an Application Specific Integrated Circuit (ASIC), a Programmable Gate Array (FPGA), a Digital Signal Processor (DSP), an embedded device, and the like.
The electronic device may be any electronic product capable of performing human-computer interaction with a user, for example, a Personal computer, a tablet computer, a smart phone, a Personal Digital Assistant (PDA), a game machine, an interactive Internet Protocol Television (IPTV), an intelligent wearable device, and the like.
The electronic device may also include a network device and/or a user device. The network device includes, but is not limited to, a single network server, a server group consisting of a plurality of network servers, or a Cloud Computing (Cloud Computing) based Cloud consisting of a large number of hosts or network servers.
The Network where the electronic device is located includes, but is not limited to, the internet, a wide area Network, a metropolitan area Network, a local area Network, a Virtual Private Network (VPN), and the like.
And S10, responding to the user behavior prediction instruction, and determining a source domain and a target domain according to the user behavior prediction instruction.
In this embodiment, the user behavior prediction instruction may be triggered by a designated staff, which is not limited in the present invention.
For example: when a new product is sold, due to the lack of enough samples, it is difficult to train a model for behavior prediction, at this time, a target domain corresponding to the source domain needs to be determined based on the idea of transfer learning, and the behavior of the user on the target domain needs to be predicted by combining the data of the source domain, so that the relevant responsible person (such as a salesman) can trigger the user behavior prediction instruction. Wherein the target domain refers to the new product, and the source domain refers to an existing product similar to the new product.
In at least one embodiment of the present invention, the determining a source domain and a target domain according to the user behavior prediction instruction comprises:
analyzing a method body of the user behavior prediction instruction to obtain information carried by the user behavior prediction instruction;
acquiring a first preset label and a second preset label;
constructing a first regular expression according to the first preset label, and constructing a second regular expression according to the second preset label;
traversing in the information carried by the user behavior prediction instruction by using the first regular expression, and determining traversed data as a first identifier;
traversing in the information carried by the user behavior prediction instruction by using the second regular expression, and determining traversed data as a second identifier;
and determining the source domain according to the first identifier and determining the target domain according to the second identifier.
The user behavior prediction instruction is a code, and in the user behavior prediction instruction, contents between { } are called as the methodology according to the writing principle of the code.
The first preset tag and the second preset tag may be configured by a user, for example: the first preset label may be configured as YID, and further, a first regular expression YID () is established with the first preset label, and traversal is performed with YID ().
Through the implementation mode, the identification can be rapidly determined based on the regular expression and the preset label, the source domain and the target domain are further determined according to the identification, and the data acquisition efficiency is improved.
S11, obtaining the data of the source domain to construct a source domain data set, and obtaining the data of the target domain to construct a target domain data set.
In this embodiment, the source domain data set may include, but is not limited to: the age of the user, the sex of the user, and the like, and the purchase information of the user.
Likewise, the target domain data set may also include, but is not limited to: the age of the user, the sex of the user, and the like, and the purchase information of the user.
It should be noted that the source domain data set and the data distribution in the actual service scenario are different, and the target domain data set and the data distribution in the actual service scenario are the same. The present embodiment aims to apply part of knowledge learned from the source domain data set to the target domain data set so as to predict the domain lacking data, and in order to make full use of this part of information, find useful data, filter out data with different distributions, the present embodiment automatically adjusts the weight of the source domain data set using the target domain data set with the same distribution.
S12, initializing the weights of the source domain data set and the target domain data set to obtain initial weights.
Specifically, the formula of the initial weight is as follows:
wherein the content of the first and second substances,and the initial weight value of the ith sample is represented, n represents the number of samples in the source domain data set, m represents the number of samples in the target domain data set, and i is n + m.
In the embodiment, when the initial weight is configured, since the target domain data set has the same data distribution as that in an actual service scene, a higher weight is first given to the target domain data set, so as to ensure that a classifier obtained through subsequent training has higher accuracy.
And S13, performing iteration training of preset rounds on the logistic regression classifier according to the source domain data set, the target domain data set and the initial weight, and storing the base classifier obtained in each iteration.
The preset round can be configured in a user-defined manner, and the invention is not limited.
Different from a voting mode generally adopted in the prior art, the embodiment uses a logistic regression classifier as a base model and uses a single model as a final classifier, so that the model can keep the interpretability of a linear model and is more suitable for actual services.
In addition, the embodiment is based on a transfer learning method, by using knowledge in an existing service scene for reference, the sample weights in the source domain are adjusted to be distributed close to the target domain, and the training effect of the model is further improved by adjusting the sample weights in the target domain.
Specifically, the method further comprises:
calculating a union set of the source domain data set and the target domain data set to obtain a sample set;
for each iteration training, configuring a first threshold for calculating the weight of the source domain data set and configuring a second threshold for calculating the weight of the target domain data set;
for each sample in the sample set, calculating a classification result based on a base classifier obtained by each iteration training;
acquiring an actual label of each sample, and acquiring the weight of the source domain data set and the target domain data set in each iteration training process;
updating the weights of the source domain data set and the target domain data set in each iteration training process according to the first threshold, the second threshold, the classification result and the actual label;
and executing the next round of iterative training by using the updated weight.
It will be appreciated that the present embodiment aims to adjust the sample weights of the sample set.
Specifically, through the iteration, if the sample of the target domain is wrongly classified, the sample is determined to be difficult to classify, adjustment is further performed according to the classification error rate of the sample of the target domain, the sample weight is increased, and the wrongly classified samples are more concerned in the next iteration; if the source domain samples are misclassified, it is determined that they are not the same as the target domain data distribution, and their weights are reduced in the next iteration. The specific implementation can be seen in the following formulas.
By the embodiment, the weights of the source domain data set and the target domain data set in each iteration process can be continuously adjusted in the training process, so that the weight configuration of the sample set is more adaptive to the actual application scene.
Specifically, the formula of the first threshold is as follows:
wherein β represents the first threshold value, and N represents the preset number of rounds;
the formula of the second threshold is as follows:
wherein, betatRepresents the second threshold value, UtRepresenting the error rate produced on the target domain data set, the formula is as follows:
wherein the content of the first and second substances,represents the weight of the ith sample during the t iteration, ht(xi) Represents the classification result of the ith sample in the process of the t round iteration, c (x)i) The actual label of the ith sample is represented.
The first threshold and the second threshold configured in the above embodiment can be used to adjust the weights of the sample set in an iterative process.
Further, the calculating the classification result based on the base classifier obtained by each iteration training includes:
acquiring historical data and an actual label of each historical data;
identifying a label corresponding to the configuration behavior from the actual labels as a target label, and calculating the proportion of the target label in the historical data as a conversion rate;
determining a quantile point as a classification threshold value according to the conversion rate;
when the output of the base classifier is greater than or equal to the classification threshold, determining that the classification result is 1; or
When the output of the base classifier is less than the classification threshold, determining that the classification result is 0.
Wherein the actual label can indicate the actual classification result of each historical data.
Wherein, the classification result represents whether the user behavior is possible to occur, 1 represents possible occurrence, and 0 represents impossible occurrence.
For example: when the classification result is 1, the purchase behavior is generated; or when the classification result is 0, it indicates that no purchasing behavior is generated.
Through the implementation mode, the quantiles can be determined based on the actual conversion rate in the historical data so as to meet the actual classification requirement of the scene.
Further, the formula of the updated weight is as follows:
wherein the content of the first and second substances,represents the weight of the ith sample after updating in the (t +1) th iteration process.
Through the embodiment, the weight of the next iteration can be continuously updated according to the weight of each iteration, so that the weight configuration of the sample set is continuously improved.
In this embodiment, normalization processing may be performed on the weights of the source domain data set and the target domain data set to unify the specification of data.
S14, acquiring a test data set, calculating a KS (Kolmogorov-Smirnov) value of each base classifier on the test data set, and determining the base classifier with the highest KS value as a target classifier.
The calculation method of the KS value belongs to a relatively mature technology, and is not described herein.
Through the implementation mode, the KS value of the base classifier obtained by each iteration on the test data set is recorded, the base classifier with the highest KS value is obtained and serves as a target classifier, and then the model which is best in performance on the test data set is used for classification, so that the classification accuracy is improved.
S15, acquiring data to be processed, inputting the data to be processed into the target classifier, and determining a behavior prediction result according to the output of the target classifier.
For example: when the data to be processed is information such as age and historical purchasing behavior of a user, if the output of the target classifier is 0, the possibility that the purchasing behavior does not occur is determined as the behavior prediction result, and if the output of the target classifier is 1, the possibility that the purchasing behavior occurs is determined as the behavior prediction result.
In the above embodiment, the weight of the source domain data set and the weight of the target domain data set are adjusted through continuous iteration based on the transfer learning method, so that the training data are more suitable for practical application scenarios, and then the optimal model is obtained for prediction through training, thereby solving the problem of sample shortage, improving the effect of model training, and further realizing more accurate user behavior prediction.
It should be noted that, in order to further ensure the security of the data, the target classifier may be deployed in the blockchain to avoid malicious tampering of the data.
According to the technical scheme, the method can respond to a user behavior prediction instruction, determine a source domain and a target domain according to the user behavior prediction instruction, acquire a data construction source domain data set of the source domain, acquire a data construction target domain data set of the target domain, initialize the weights of the source domain data set and the target domain data set to acquire an initial weight, firstly give a higher weight to the target domain data set to ensure that a classifier obtained by subsequent training has higher accuracy because the target domain data set is the same as data distribution under an actual service scene, perform iterative training of preset rounds on a logistic regression classifier according to the source domain data set, the target domain data set and the initial weight, and store a base classifier obtained by each round of iteration, which is different from a voting mode generally adopted in the prior art, the present embodiment uses a logistic regression classifier as the base model, a single model as the final classifier, so that the model can keep the interpretability of the linear model, is more suitable for actual business, acquires a test data set, calculating the KS value of each base classifier on the test data set, determining the base classifier with the highest KS value as a target classifier, acquiring data to be processed, inputting the data to be processed to the target classifier, and determining a behavior prediction result from the output of the target classifier, based on a method of transfer learning, by continuously and iteratively adjusting the weights of the source domain data set and the target domain data set, the training data is more suitable for the actual application scene, and then training to obtain an optimal model for prediction, so that the problem of sample shortage is solved, the model training effect is improved, and more accurate user behavior prediction is realized.
Fig. 2 is a functional block diagram of a user behavior prediction apparatus according to a preferred embodiment of the present invention. The user behavior prediction apparatus 11 includes a determination unit 110, an acquisition unit 111, an initialization unit 112, a training unit 113, and a calculation unit 114. The module/unit referred to in the present invention refers to a series of computer program segments that can be executed by the processor 13 and that can perform a fixed function, and that are stored in the memory 12. In the present embodiment, the functions of the modules/units will be described in detail in the following embodiments.
In response to the user behavior prediction instruction, the determination unit 110 determines a source domain and a target domain according to the user behavior prediction instruction.
In this embodiment, the user behavior prediction instruction may be triggered by a designated staff, which is not limited in the present invention.
For example: when a new product is sold, due to the lack of enough samples, it is difficult to train a model for behavior prediction, at this time, a target domain corresponding to the source domain needs to be determined based on the idea of transfer learning, and the behavior of the user on the target domain needs to be predicted by combining the data of the source domain, so that the relevant responsible person (such as a salesman) can trigger the user behavior prediction instruction. Wherein the target domain refers to the new product, and the source domain refers to an existing product similar to the new product.
In at least one embodiment of the present invention, the determining unit 110 determines the source domain and the target domain according to the user behavior prediction instruction includes:
analyzing a method body of the user behavior prediction instruction to obtain information carried by the user behavior prediction instruction;
acquiring a first preset label and a second preset label;
constructing a first regular expression according to the first preset label, and constructing a second regular expression according to the second preset label;
traversing in the information carried by the user behavior prediction instruction by using the first regular expression, and determining traversed data as a first identifier;
traversing in the information carried by the user behavior prediction instruction by using the second regular expression, and determining traversed data as a second identifier;
and determining the source domain according to the first identifier and determining the target domain according to the second identifier.
The user behavior prediction instruction is a code, and in the user behavior prediction instruction, contents between { } are called as the methodology according to the writing principle of the code.
The first preset tag and the second preset tag may be configured by a user, for example: the first preset label may be configured as YID, and further, a first regular expression YID () is established with the first preset label, and traversal is performed with YID ().
Through the implementation mode, the identification can be rapidly determined based on the regular expression and the preset label, the source domain and the target domain are further determined according to the identification, and the data acquisition efficiency is improved.
The obtaining unit 111 obtains data of the source domain to construct a source domain data set, and obtains data of the target domain to construct a target domain data set.
In this embodiment, the source domain data set may include, but is not limited to: the age of the user, the sex of the user, and the like, and the purchase information of the user.
Likewise, the target domain data set may also include, but is not limited to: the age of the user, the sex of the user, and the like, and the purchase information of the user.
It should be noted that the source domain data set and the data distribution in the actual service scenario are different, and the target domain data set and the data distribution in the actual service scenario are the same. The present embodiment aims to apply part of knowledge learned from the source domain data set to the target domain data set so as to predict the domain lacking data, and in order to make full use of this part of information, find useful data, filter out data with different distributions, the present embodiment automatically adjusts the weight of the source domain data set using the target domain data set with the same distribution.
The initializing unit 112 initializes the weights of the source domain data set and the target domain data set to obtain initial weights.
Specifically, the formula of the initial weight is as follows:
wherein the content of the first and second substances,and the initial weight value of the ith sample is represented, n represents the number of samples in the source domain data set, m represents the number of samples in the target domain data set, and i is n + m.
In the embodiment, when the initial weight is configured, since the target domain data set has the same data distribution as that in an actual service scene, a higher weight is first given to the target domain data set, so as to ensure that a classifier obtained through subsequent training has higher accuracy.
The training unit 113 performs iteration training of a preset number of rounds on the logistic regression classifier according to the source domain data set, the target domain data set and the initial weight, and stores a base classifier obtained by each iteration.
The preset round can be configured in a user-defined manner, and the invention is not limited.
Different from a voting mode generally adopted in the prior art, the embodiment uses a logistic regression classifier as a base model and uses a single model as a final classifier, so that the model can keep the interpretability of a linear model and is more suitable for actual services.
In addition, the embodiment is based on a transfer learning method, by using knowledge in an existing service scene for reference, the sample weights in the source domain are adjusted to be distributed close to the target domain, and the training effect of the model is further improved by adjusting the sample weights in the target domain.
Specifically, a union set of the source domain data set and the target domain data set is calculated to obtain a sample set;
for each iteration training, configuring a first threshold for calculating the weight of the source domain data set and configuring a second threshold for calculating the weight of the target domain data set;
for each sample in the sample set, calculating a classification result based on a base classifier obtained by each iteration training;
acquiring an actual label of each sample, and acquiring the weight of the source domain data set and the target domain data set in each iteration training process;
updating the weights of the source domain data set and the target domain data set in each iteration training process according to the first threshold, the second threshold, the classification result and the actual label;
and executing the next round of iterative training by using the updated weight.
It will be appreciated that the present embodiment aims to adjust the sample weights of the sample set.
Specifically, through the iteration, if the sample of the target domain is wrongly classified, the sample is determined to be difficult to classify, adjustment is further performed according to the classification error rate of the sample of the target domain, the sample weight is increased, and the wrongly classified samples are more concerned in the next iteration; if the source domain samples are misclassified, it is determined that they are not the same as the target domain data distribution, and their weights are reduced in the next iteration. The specific implementation can be seen in the following formulas.
By the embodiment, the weights of the source domain data set and the target domain data set in each iteration process can be continuously adjusted in the training process, so that the weight configuration of the sample set is more adaptive to the actual application scene.
Specifically, the formula of the first threshold is as follows:
wherein β represents the first threshold value, and N represents the preset number of rounds;
the formula of the second threshold is as follows:
wherein, betatRepresents the second threshold value, UtRepresenting the error rate produced on the target domain data set, the formula is as follows:
wherein the content of the first and second substances,represents the weight of the ith sample during the t iteration, ht(xi) Represents the classification result of the ith sample in the process of the t round iteration, c (x)i) The actual label of the ith sample is represented.
The first threshold and the second threshold configured in the above embodiment can be used to adjust the weights of the sample set in an iterative process.
Further, the calculating the classification result based on the base classifier obtained by each iteration training includes:
acquiring historical data and an actual label of each historical data;
identifying a label corresponding to the configuration behavior from the actual labels as a target label, and calculating the proportion of the target label in the historical data as a conversion rate;
determining a quantile point as a classification threshold value according to the conversion rate;
when the output of the base classifier is greater than or equal to the classification threshold, determining that the classification result is 1; or
When the output of the base classifier is less than the classification threshold, determining that the classification result is 0.
Wherein the actual label can indicate the actual classification result of each historical data.
Wherein, the classification result represents whether the user behavior is possible to occur, 1 represents possible occurrence, and 0 represents impossible occurrence.
For example: when the classification result is 1, the purchase behavior is generated; or when the classification result is 0, it indicates that no purchasing behavior is generated.
Through the implementation mode, the quantiles can be determined based on the actual conversion rate in the historical data so as to meet the actual classification requirement of the scene.
Further, the formula of the updated weight is as follows:
wherein the content of the first and second substances,represents the weight of the ith sample after updating in the (t +1) th iteration process.
Through the embodiment, the weight of the next iteration can be continuously updated according to the weight of each iteration, so that the weight configuration of the sample set is continuously improved.
In this embodiment, normalization processing may be performed on the weights of the source domain data set and the target domain data set to unify the specification of data.
The calculation unit 114 acquires a test data set, and calculates a KS (Kolmogorov-Smirnov) value of each base classifier on the test data set, and determines the base classifier having the highest KS value as a target classifier.
The calculation method of the KS value belongs to a relatively mature technology, and is not described herein.
Through the implementation mode, the KS value of the base classifier obtained by each iteration on the test data set is recorded, the base classifier with the highest KS value is obtained and serves as a target classifier, and then the model which is best in performance on the test data set is used for classification, so that the classification accuracy is improved.
The determining unit 110 obtains data to be processed, inputs the data to be processed to the target classifier, and determines a behavior prediction result according to an output of the target classifier.
For example: when the data to be processed is information such as age and historical purchasing behavior of a user, if the output of the target classifier is 0, the possibility that the purchasing behavior does not occur is determined as the behavior prediction result, and if the output of the target classifier is 1, the possibility that the purchasing behavior occurs is determined as the behavior prediction result.
In the above embodiment, the weight of the source domain data set and the weight of the target domain data set are adjusted through continuous iteration based on the transfer learning method, so that the training data are more suitable for practical application scenarios, and then the optimal model is obtained for prediction through training, thereby solving the problem of sample shortage, improving the effect of model training, and further realizing more accurate user behavior prediction.
It should be noted that, in order to further ensure the security of the data, the target classifier may be deployed in the blockchain to avoid malicious tampering of the data.
According to the technical scheme, the method can respond to a user behavior prediction instruction, determine a source domain and a target domain according to the user behavior prediction instruction, acquire a data construction source domain data set of the source domain, acquire a data construction target domain data set of the target domain, initialize the weights of the source domain data set and the target domain data set to acquire an initial weight, firstly give a higher weight to the target domain data set to ensure that a classifier obtained by subsequent training has higher accuracy because the target domain data set is the same as data distribution under an actual service scene, perform iterative training of preset rounds on a logistic regression classifier according to the source domain data set, the target domain data set and the initial weight, and store a base classifier obtained by each round of iteration, which is different from a voting mode generally adopted in the prior art, the present embodiment uses a logistic regression classifier as the base model, a single model as the final classifier, so that the model can keep the interpretability of the linear model, is more suitable for actual business, acquires a test data set, calculating the KS value of each base classifier on the test data set, determining the base classifier with the highest KS value as a target classifier, acquiring data to be processed, inputting the data to be processed to the target classifier, and determining a behavior prediction result from the output of the target classifier, based on a method of transfer learning, by continuously and iteratively adjusting the weights of the source domain data set and the target domain data set, the training data is more suitable for the actual application scene, and then training to obtain an optimal model for prediction, so that the problem of sample shortage is solved, the model training effect is improved, and more accurate user behavior prediction is realized.
Fig. 3 is a schematic structural diagram of an electronic device implementing a method for predicting user behavior according to a preferred embodiment of the present invention.
The electronic device 1 may comprise a memory 12, a processor 13 and a bus, and may further comprise a computer program, such as a user behavior prediction program, stored in the memory 12 and executable on the processor 13.
It will be understood by those skilled in the art that the schematic diagram is merely an example of the electronic device 1, and does not constitute a limitation to the electronic device 1, the electronic device 1 may have a bus-type structure or a star-type structure, the electronic device 1 may further include more or less hardware or software than those shown in the figures, or different component arrangements, for example, the electronic device 1 may further include an input and output device, a network access device, and the like.
It should be noted that the electronic device 1 is only an example, and other existing or future electronic products, such as those that can be adapted to the present invention, should also be included in the scope of the present invention, and are included herein by reference.
The memory 12 includes at least one type of readable storage medium, which includes flash memory, removable hard disks, multimedia cards, card-type memory (e.g., SD or DX memory, etc.), magnetic memory, magnetic disks, optical disks, etc. The memory 12 may in some embodiments be an internal storage unit of the electronic device 1, for example a removable hard disk of the electronic device 1. The memory 12 may also be an external storage device of the electronic device 1 in other embodiments, such as a plug-in mobile hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), and the like provided on the electronic device 1. Further, the memory 12 may also include both an internal storage unit and an external storage device of the electronic device 1. The memory 12 may be used not only to store application software installed in the electronic device 1 and various types of data, such as codes of a user behavior prediction program, etc., but also to temporarily store data that has been output or is to be output.
The processor 13 may be composed of an integrated circuit in some embodiments, for example, a single packaged integrated circuit, or may be composed of a plurality of integrated circuits packaged with the same or different functions, including one or more Central Processing Units (CPUs), microprocessors, digital Processing chips, graphics processors, and combinations of various control chips. The processor 13 is a Control Unit (Control Unit) of the electronic device 1, connects various components of the electronic device 1 by using various interfaces and lines, and executes various functions and processes data of the electronic device 1 by running or executing programs or modules (e.g., executing a user behavior prediction program, etc.) stored in the memory 12 and calling data stored in the memory 12.
The processor 13 executes an operating system of the electronic device 1 and various installed application programs. The processor 13 executes the application program to implement the steps in the above-described embodiments of the user behavior prediction method, such as the steps shown in fig. 1.
Illustratively, the computer program may be divided into one or more modules/units, which are stored in the memory 12 and executed by the processor 13 to accomplish the present invention. The one or more modules/units may be a series of computer program instruction segments capable of performing specific functions, which are used to describe the execution process of the computer program in the electronic device 1. For example, the computer program may be divided into a determination unit 110, an acquisition unit 111, an initialization unit 112, a training unit 113, a calculation unit 114.
The integrated unit implemented in the form of a software functional module may be stored in a computer-readable storage medium. The software functional module is stored in a storage medium and includes several instructions to enable a computer device (which may be a personal computer, a computer device, or a network device) or a processor (processor) to execute parts of the user behavior prediction method according to the embodiments of the present invention.
The integrated modules/units of the electronic device 1 may be stored in a computer-readable storage medium if they are implemented in the form of software functional units and sold or used as separate products. Based on such understanding, all or part of the flow of the method according to the embodiments of the present invention may be implemented by a computer program, which may be stored in a computer-readable storage medium, and when the computer program is executed by a processor, the steps of the method embodiments may be implemented.
Wherein the computer program comprises computer program code, which may be in the form of source code, object code, an executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U-disk, removable hard disk, magnetic disk, optical disk, computer Memory, Read-Only Memory (ROM), random-access Memory, or the like.
Further, the computer-readable storage medium may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function, and the like; the storage data area may store data created according to the use of the blockchain node, and the like.
The block chain is a novel application mode of computer technologies such as distributed data storage, point-to-point transmission, a consensus mechanism, an encryption algorithm and the like. A block chain (Blockchain), which is essentially a decentralized database, is a series of data blocks associated by using a cryptographic method, and each data block contains information of a batch of network transactions, so as to verify the validity (anti-counterfeiting) of the information and generate a next block. The blockchain may include a blockchain underlying platform, a platform product service layer, an application service layer, and the like.
The bus may be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one arrow is shown in FIG. 3, but this does not indicate only one bus or one type of bus. The bus is arranged to enable connection communication between the memory 12 and at least one processor 13 or the like.
Although not shown, the electronic device 1 may further include a power supply (such as a battery) for supplying power to each component, and preferably, the power supply may be logically connected to the at least one processor 13 through a power management device, so as to implement functions of charge management, discharge management, power consumption management, and the like through the power management device. The power supply may also include any component of one or more dc or ac power sources, recharging devices, power failure detection circuitry, power converters or inverters, power status indicators, and the like. The electronic device 1 may further include various sensors, a bluetooth module, a Wi-Fi module, and the like, which are not described herein again.
Further, the electronic device 1 may further include a network interface, and optionally, the network interface may include a wired interface and/or a wireless interface (such as a WI-FI interface, a bluetooth interface, etc.), which are generally used for establishing a communication connection between the electronic device 1 and other electronic devices.
Optionally, the electronic device 1 may further comprise a user interface, which may be a Display (Display), an input unit (such as a Keyboard), and optionally a standard wired interface, a wireless interface. Alternatively, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch device, or the like. The display, which may also be referred to as a display screen or display unit, is suitable for displaying information processed in the electronic device 1 and for displaying a visualized user interface, among other things.
It is to be understood that the described embodiments are for purposes of illustration only and that the scope of the appended claims is not limited to such structures.
Fig. 3 only shows the electronic device 1 with components 12-13, and it will be understood by a person skilled in the art that the structure shown in fig. 3 does not constitute a limitation of the electronic device 1, and may comprise fewer or more components than shown, or a combination of certain components, or a different arrangement of components.
With reference to fig. 1, the memory 12 of the electronic device 1 stores a plurality of instructions to implement a user behavior prediction method, and the processor 13 executes the plurality of instructions to implement:
responding to a user behavior prediction instruction, and determining a source domain and a target domain according to the user behavior prediction instruction;
acquiring data of the source domain to construct a source domain data set, and acquiring data of the target domain to construct a target domain data set;
initializing the weights of the source domain data set and the target domain data set to obtain an initial weight;
performing iteration training of a preset number of rounds on a logistic regression classifier according to the source domain data set, the target domain data set and the initial weight, and storing a base classifier obtained by each round of iteration;
acquiring a test data set, calculating a KS value of each base classifier on the test data set, and determining the base classifier with the highest KS value as a target classifier;
and acquiring data to be processed, inputting the data to be processed into the target classifier, and determining a behavior prediction result according to the output of the target classifier.
Specifically, the processor 13 may refer to the description of the relevant steps in the embodiment corresponding to fig. 1 for a specific implementation method of the instruction, which is not described herein again.
In the embodiments provided in the present invention, it should be understood that the disclosed system, apparatus and method may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the modules is only one logical functional division, and other divisions may be realized in practice.
The modules described as separate parts may or may not be physically separate, and parts displayed as modules may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment.
In addition, functional modules in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, or in a form of hardware plus a software functional module.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential attributes thereof.
The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference signs in the claims shall not be construed as limiting the claim concerned.
Furthermore, it is obvious that the word "comprising" does not exclude other elements or steps, and the singular does not exclude the plural. A plurality of units or means recited in the present invention may also be implemented by one unit or means through software or hardware. The terms first, second, etc. are used to denote names, but not any particular order.
Finally, it should be noted that the above embodiments are only for illustrating the technical solutions of the present invention and not for limiting, and although the present invention is described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications or equivalent substitutions may be made on the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.
Claims (9)
1. A user behavior prediction method, characterized by comprising:
in response to a user behavior prediction instruction, determining a source domain and a target domain according to the user behavior prediction instruction, wherein the target domain is a new product, the source domain is an existing product similar to the new product, and the determining the source domain and the target domain according to the user behavior prediction instruction comprises: analyzing a method body of the user behavior prediction instruction to obtain information carried by the user behavior prediction instruction; acquiring a first preset label and a second preset label; constructing a first regular expression according to the first preset label, and constructing a second regular expression according to the second preset label; traversing in the information carried by the user behavior prediction instruction by using the first regular expression, and determining traversed data as a first identifier; traversing in the information carried by the user behavior prediction instruction by using the second regular expression, and determining traversed data as a second identifier; determining the source domain according to the first identifier and determining the target domain according to the second identifier;
acquiring data of the source domain to construct a source domain data set, and acquiring data of the target domain to construct a target domain data set;
initializing the weights of the source domain data set and the target domain data set to obtain an initial weight;
performing iteration training of a preset number of rounds on a logistic regression classifier according to the source domain data set, the target domain data set and the initial weight, and storing a base classifier obtained by each round of iteration;
acquiring a test data set, calculating a KS value of each base classifier on the test data set, and determining the base classifier with the highest KS value as a target classifier;
and acquiring data to be processed, inputting the data to be processed into the target classifier, and determining a behavior prediction result according to the output of the target classifier.
2. The user behavior prediction method of claim 1, characterized in that the formula of the initial weight is as follows:
3. The method of user behavior prediction according to claim 2, characterized in that the method further comprises:
calculating a union set of the source domain data set and the target domain data set to obtain a sample set;
for each iteration training, configuring a first threshold for calculating the weight of the source domain data set and configuring a second threshold for calculating the weight of the target domain data set;
for each sample in the sample set, calculating a classification result based on a base classifier obtained by each iteration training;
acquiring an actual label of each sample, and acquiring the weight of the source domain data set and the target domain data set in each iteration training process;
updating the weights of the source domain data set and the target domain data set in each iteration training process according to the first threshold, the second threshold, the classification result and the actual label;
and executing the next round of iterative training by using the updated weight.
4. A method for predicting user behavior according to claim 3, wherein the formula of the first threshold value is as follows:
wherein β represents the first threshold value, and N represents the preset number of rounds;
the formula of the second threshold is as follows:
wherein, betatRepresents the second threshold value, UtRepresenting the error rate produced on the target domain data set, the formula is as follows:
5. The method of claim 3, wherein computing the classification result based on the base classifier from each iteration of training comprises:
acquiring historical data and an actual label of each historical data;
identifying a label corresponding to the configuration behavior from the actual labels as a target label, and calculating the proportion of the target label in the historical data as a conversion rate;
determining a quantile point as a classification threshold value according to the conversion rate;
when the output of the base classifier is greater than or equal to the classification threshold, determining that the classification result is 1; or
When the output of the base classifier is less than the classification threshold, determining that the classification result is 0.
7. A user behavior prediction apparatus, characterized in that the user behavior prediction apparatus comprises:
a determining unit, configured to determine, in response to a user behavior prediction instruction, a source domain and a target domain according to the user behavior prediction instruction, where the target domain is a new product, the source domain is an existing product similar to the new product, and the determining the source domain and the target domain according to the user behavior prediction instruction includes: analyzing a method body of the user behavior prediction instruction to obtain information carried by the user behavior prediction instruction; acquiring a first preset label and a second preset label; constructing a first regular expression according to the first preset label, and constructing a second regular expression according to the second preset label; traversing in the information carried by the user behavior prediction instruction by using the first regular expression, and determining traversed data as a first identifier; traversing in the information carried by the user behavior prediction instruction by using the second regular expression, and determining traversed data as a second identifier; determining the source domain according to the first identifier and determining the target domain according to the second identifier;
the acquisition unit is used for acquiring data of the source domain to construct a source domain data set and acquiring data of the target domain to construct a target domain data set;
the initialization unit is used for initializing the weights of the source domain data set and the target domain data set to obtain an initial weight;
the training unit is used for performing iteration training of preset rounds on the logistic regression classifier according to the source domain data set, the target domain data set and the initial weight, and storing a base classifier obtained by each iteration;
the calculation unit is used for acquiring a test data set, calculating a KS value of each base classifier on the test data set, and determining the base classifier with the highest KS value as a target classifier;
the determining unit is further configured to acquire data to be processed, input the data to be processed to the target classifier, and determine a behavior prediction result according to an output of the target classifier.
8. An electronic device, characterized in that the electronic device comprises:
a memory storing at least one instruction; and
a processor executing instructions stored in the memory to implement a user behavior prediction method as claimed in any one of claims 1 to 6.
9. A computer-readable storage medium characterized by: the computer-readable storage medium has stored therein at least one instruction that is executable by a processor in an electronic device to implement the user behavior prediction method of any of claims 1-6.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110199787.9A CN112801718B (en) | 2021-02-22 | 2021-02-22 | User behavior prediction method, device, equipment and medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110199787.9A CN112801718B (en) | 2021-02-22 | 2021-02-22 | User behavior prediction method, device, equipment and medium |
Publications (2)
Publication Number | Publication Date |
---|---|
CN112801718A CN112801718A (en) | 2021-05-14 |
CN112801718B true CN112801718B (en) | 2021-10-01 |
Family
ID=75815390
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110199787.9A Active CN112801718B (en) | 2021-02-22 | 2021-02-22 | User behavior prediction method, device, equipment and medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112801718B (en) |
Families Citing this family (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113406496B (en) * | 2021-05-26 | 2023-02-28 | 广州市香港科大霍英东研究院 | Battery capacity prediction method, system, device and medium based on model migration |
CN113343087A (en) * | 2021-06-09 | 2021-09-03 | 南京星云数字技术有限公司 | Method and system for acquiring marketing user |
CN113256181A (en) * | 2021-07-13 | 2021-08-13 | 平安科技(深圳)有限公司 | Risk factor prediction method, device, equipment and medium |
CN113516231B (en) * | 2021-08-10 | 2024-03-29 | 大连海事大学 | Daily behavior migration identification method based on DSN deep countermeasure migration network |
CN113657535A (en) * | 2021-08-24 | 2021-11-16 | 北京奇艺世纪科技有限公司 | Model training method and device, electronic equipment and storage medium |
CN113516205B (en) * | 2021-09-03 | 2021-12-14 | 平安科技(深圳)有限公司 | Employee stability classification method based on artificial intelligence and related equipment |
CN113743678B (en) * | 2021-09-18 | 2024-05-31 | 中国银行股份有限公司 | User credit score prediction method and related equipment |
Family Cites Families (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106599922B (en) * | 2016-12-16 | 2021-08-24 | 中国科学院计算技术研究所 | Transfer learning method and system for large-scale data calibration |
US10776693B2 (en) * | 2017-01-31 | 2020-09-15 | Xerox Corporation | Method and system for learning transferable feature representations from a source domain for a target domain |
EP3818446A4 (en) * | 2018-07-06 | 2021-09-08 | Telefonaktiebolaget Lm Ericsson (Publ) | Methods and systems for dynamic service performance prediction using transfer learning |
CN109544306B (en) * | 2018-11-30 | 2021-09-21 | 苏州大学 | Cross-domain recommendation method and device based on user behavior sequence characteristics |
CN109462610A (en) * | 2018-12-24 | 2019-03-12 | 哈尔滨工程大学 | A kind of network inbreak detection method based on Active Learning and transfer learning |
CN110400169B (en) * | 2019-07-02 | 2022-12-23 | 创新先进技术有限公司 | Information pushing method, device and equipment |
CN111027870A (en) * | 2019-12-14 | 2020-04-17 | 支付宝(杭州)信息技术有限公司 | User risk assessment method and device, electronic equipment and storage medium |
CN112115916B (en) * | 2020-09-29 | 2023-05-02 | 西安电子科技大学 | Domain adaptive Faster R-CNN semi-supervised SAR detection method |
CN112270545A (en) * | 2020-10-27 | 2021-01-26 | 上海淇馥信息技术有限公司 | Financial risk prediction method and device based on migration sample screening and electronic equipment |
CN112231586A (en) * | 2020-12-15 | 2021-01-15 | 平安科技(深圳)有限公司 | Course recommendation method, device, equipment and medium based on transfer learning |
CN112288042B (en) * | 2020-12-18 | 2021-04-02 | 蚂蚁智信(杭州)信息技术有限公司 | Updating method and device of behavior prediction system, storage medium and computing equipment |
-
2021
- 2021-02-22 CN CN202110199787.9A patent/CN112801718B/en active Active
Also Published As
Publication number | Publication date |
---|---|
CN112801718A (en) | 2021-05-14 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112801718B (en) | User behavior prediction method, device, equipment and medium | |
CN112231586A (en) | Course recommendation method, device, equipment and medium based on transfer learning | |
CN114663198A (en) | Product recommendation method, device and equipment based on user portrait and storage medium | |
CN115146865A (en) | Task optimization method based on artificial intelligence and related equipment | |
CN113626606B (en) | Information classification method, device, electronic equipment and readable storage medium | |
CN112396547B (en) | Course recommendation method, device, equipment and medium based on unsupervised learning | |
CN111985545B (en) | Target data detection method, device, equipment and medium based on artificial intelligence | |
CN115081538A (en) | Customer relationship identification method, device, equipment and medium based on machine learning | |
CN112380454A (en) | Training course recommendation method, device, equipment and medium | |
CN114612194A (en) | Product recommendation method and device, electronic equipment and storage medium | |
CN114781832A (en) | Course recommendation method and device, electronic equipment and storage medium | |
CN113313211B (en) | Text classification method, device, electronic equipment and storage medium | |
CN113658002A (en) | Decision tree-based transaction result generation method and device, electronic equipment and medium | |
CN111652282B (en) | Big data-based user preference analysis method and device and electronic equipment | |
CN112651782A (en) | Behavior prediction method, device, equipment and medium based on zoom dot product attention | |
CN112269875A (en) | Text classification method and device, electronic equipment and storage medium | |
CN111950707A (en) | Behavior prediction method, apparatus, device and medium based on behavior co-occurrence network | |
CN111429085A (en) | Contract data generation method and device, electronic equipment and storage medium | |
CN116843481A (en) | Knowledge graph analysis method, device, equipment and storage medium | |
CN116842290A (en) | Data caching method, device, equipment and computer readable storage medium | |
CN111679959A (en) | Computer performance data determination method and device, computer equipment and storage medium | |
CN115239958A (en) | Wheel hub damage detection method based on artificial intelligence and related equipment | |
CN113705684B (en) | Reverse iteration anomaly detection method and device, electronic equipment and medium | |
CN113435746B (en) | User workload scoring method and device, electronic equipment and storage medium | |
CN112330080B (en) | Factor screening method, device, equipment and medium based on connectivity graph |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |