CN114463117A - User behavior prediction method, system and device - Google Patents
User behavior prediction method, system and device Download PDFInfo
- Publication number
- CN114463117A CN114463117A CN202210126495.7A CN202210126495A CN114463117A CN 114463117 A CN114463117 A CN 114463117A CN 202210126495 A CN202210126495 A CN 202210126495A CN 114463117 A CN114463117 A CN 114463117A
- Authority
- CN
- China
- Prior art keywords
- user
- behavior
- target
- data
- label
- 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.)
- Pending
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
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/03—Credit; Loans; Processing thereof
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Business, Economics & Management (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Software Systems (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Marketing (AREA)
- Economics (AREA)
- General Business, Economics & Management (AREA)
- Technology Law (AREA)
- Strategic Management (AREA)
- Medical Informatics (AREA)
- Development Economics (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The embodiment of the application discloses a user behavior prediction method, a user behavior prediction system and a user behavior prediction device, which are used for monitoring and predicting user behaviors and have high accuracy. The method in the embodiment of the application comprises the following steps: obtaining a target prediction model, wherein the target prediction model is obtained by performing machine learning training on user tag data and user historical behavior data, a first corresponding relation between a user tag and user behavior characteristics is stored in the target prediction model, and the user behavior characteristics are obtained by performing machine learning on the user historical behavior data; acquiring target user behavior data; acquiring a target user label; and inputting the target user behavior data and the target user label into the target prediction model to obtain user predicted behavior data output by the target prediction model according to the first corresponding relation and the user behavior characteristics.
Description
Technical Field
The embodiment of the application relates to the field of machine learning algorithms, in particular to a user behavior prediction method, a user behavior prediction system and a user behavior prediction device.
Background
With the generation of credit business and the development and popularization of personal small and micro loan business, various financial institutions issue diversified credit products at a dispute, the number of users purchasing the credit products is also increased sharply, and monitoring and risk prevention of user behaviors become a vital part in the credit product service link.
Currently, a traditional data monitoring method is generally adopted for monitoring user behavior data, various data indexes are set for the user behavior data, threshold values are set for the various indexes, whether the user behavior data meet the preset index threshold value or not is judged according to the user behavior data monitored in real time, and whether abnormal behavior data exist or not is judged, so that whether abnormal user behavior exists or not is confirmed.
However, the conventional data monitoring method generally cannot distinguish the user types, for example, the data difference of the same index of different types of users is large, and if the threshold is not set according to the user type, but the thresholds of the same index of all the user types are considered in a short time, the misjudgment is easily caused. If different threshold values are set according to different types of users, the classification is complicated due to various types and indexes of the users, or the situation that the classification rule is defined artificially is inaccurate and unreasonable exists.
Disclosure of Invention
The embodiment of the application provides a user behavior prediction method, a user behavior prediction system and a user behavior prediction device, which are used for monitoring and predicting user behaviors and have high accuracy.
The user behavior prediction method provided by the embodiment of the application comprises the following steps:
obtaining a target prediction model, wherein the target prediction model is obtained by performing machine learning training on user tag data and user historical behavior data, a first corresponding relation between a user tag and user behavior characteristics is stored in the target prediction model, and the user behavior characteristics are obtained by performing machine learning on the user historical behavior data;
acquiring target user behavior data;
acquiring a target user label;
and inputting the target user behavior data and the target user label into the target prediction model to obtain user predicted behavior data output by the target prediction model according to the first corresponding relation and the user behavior characteristics.
Optionally, before the obtaining of the target user tag, the method further includes:
acquiring user characteristic data;
clustering users according to the user characteristic data to obtain a plurality of user clusters with similar user characteristics, and setting the same user label for the users in the same user cluster to obtain user label data;
the acquiring of the target user tag comprises:
and confirming the user label of the target user of the user behavior to be predicted as the target user label.
Optionally, the clustering users according to the user feature data to obtain a plurality of user clusters with similar user features, and setting the same user tag for users in the same user cluster, and obtaining the user tag data includes:
clustering users according to the user characteristic data by adopting a clustering algorithm to obtain a plurality of user clusters with similar user characteristics;
and carrying out streaming processing on the user characteristic data so as to update the types of the user clusters and the user label data in real time.
Optionally, before the obtaining of the target user tag, the method further includes:
obtaining a user clustering model, wherein the user clustering model is obtained by performing machine learning training on the user characteristic data and the user tag data, and a second corresponding relation between the user characteristic data and the user tag data is stored in the user clustering model;
the acquiring of the target user tag comprises:
and inputting the target user characteristic data of the target user of the user behavior to be predicted into the user clustering model to obtain a target user label output by the user clustering model according to the second corresponding relation.
Optionally, before the obtaining the target prediction model, the method further includes:
obtaining user historical behavior data in a streaming mode;
acquiring user tag data;
and inputting the user historical behavior data and the user label data into an initial prediction model as training samples, performing machine learning training on the initial prediction model by using the training samples to obtain a target prediction model, wherein a first corresponding relation between a user label and a user behavior characteristic is stored in the target prediction model, and the user behavior characteristic is obtained by performing machine learning on the user historical behavior data.
Optionally, after obtaining the predicted user behavior data output by the target prediction model according to the first corresponding relationship and the user behavior feature, the method further includes:
and monitoring the user behavior in real time, and confirming that the user actual behavior is abnormal when the deviation of the user actual behavior and the user predicted behavior data does not meet a user behavior threshold value.
Optionally, after confirming that there is an abnormality in the actual user behavior when the deviation between the actual user behavior and the predicted user behavior data does not satisfy the user behavior threshold, the method further includes:
when the user actual behavior of the target user is abnormal, the user actual behavior is marked as the user abnormal behavior and is stored in the user behavior data of the target user.
The embodiment of the application provides a user behavior prediction system, which comprises:
the target prediction model is obtained by performing machine learning training on user label data and user historical behavior data, a first corresponding relation between a user label and user behavior characteristics is stored in the target prediction model, and the user behavior characteristics are obtained by performing machine learning on the user historical behavior data;
the acquisition unit is also used for acquiring target user behavior data;
the acquisition unit is also used for acquiring a target user label;
and the output unit is used for inputting the target user behavior data and the target user label into the target prediction model to obtain user predicted behavior data output by the target prediction model according to the first corresponding relation and the user behavior characteristics.
Optionally, the obtaining unit is specifically configured to obtain user feature data;
the user behavior prediction system further comprises:
the clustering unit is used for clustering users according to the user characteristic data to obtain a plurality of user clusters with similar user characteristics, and setting the same user label for the users in the same user cluster to obtain user label data;
the obtaining unit is specifically configured to determine a user tag of a target user of a user behavior to be predicted as a target user tag.
Optionally, the clustering unit is further configured to cluster the users according to the user feature data by using a clustering algorithm to obtain a plurality of user clusters with similar user features;
and carrying out streaming processing on the user characteristic data so as to update the types of the user clusters and the user label data in real time.
Optionally, the obtaining unit is specifically configured to obtain a user clustering model, where the user clustering model is obtained by performing machine learning training on the user feature data and the user tag data, and a second corresponding relationship between the user feature data and the user tag data is stored in the user clustering model;
the acquiring of the target user tag comprises:
and inputting the target user characteristic data of the target user of the user behavior to be predicted into the user clustering model to obtain a target user label output by the user clustering model according to the second corresponding relation.
Optionally, the obtaining unit is specifically configured to obtain the historical behavior data of the user in a streaming manner;
acquiring user tag data;
and inputting the user historical behavior data and the user label data into an initial prediction model as training samples, performing machine learning training on the initial prediction model by using the training samples to obtain a target prediction model, wherein a first corresponding relation between a user label and a user behavior characteristic is stored in the target prediction model, and the user behavior characteristic is obtained by performing machine learning on the user historical behavior data.
Optionally, the user behavior prediction system further includes:
and the monitoring unit is used for monitoring the user behavior in real time, and when the deviation between the actual user behavior and the predicted user behavior data does not meet the user behavior threshold value, determining that the actual user behavior is abnormal.
Optionally, the user behavior prediction system further includes:
and the marking unit is used for marking the user actual behavior as the user abnormal behavior when the user actual behavior of the target user is abnormal, and storing the user abnormal behavior in the user behavior data of the target user.
The embodiment of the application provides a user behavior prediction device, which comprises:
the system comprises a central processing unit, a memory and an input/output interface;
the memory is a transient memory or a persistent memory;
the central processor is configured to communicate with the memory and execute the instruction operations in the memory to perform the aforementioned user behavior prediction method.
The computer-readable storage medium provided by the embodiment of the application comprises instructions, and when the instructions are executed on a computer, the instructions cause the computer to execute the user behavior prediction method.
According to the technical scheme, the embodiment of the application has the following advantages:
the user behavior is predicted by establishing the target prediction model, the user is classified by setting the user label for the user, target user behavior data obtained through real-time monitoring are input into the target prediction model, user predicted behavior data output by the target prediction model according to the first corresponding relation between the user label and the user behavior characteristic is obtained, and accuracy is high.
Drawings
Fig. 1 is a schematic diagram of an embodiment of a user behavior prediction method provided in an embodiment of the present application;
fig. 2 is a schematic diagram of another implementation of a user behavior prediction method provided in an embodiment of the present application;
fig. 3 is a schematic diagram of an embodiment of a user behavior prediction system according to an embodiment of the present application;
fig. 4 is a schematic diagram of an embodiment of a user behavior prediction apparatus according to an embodiment of the present application.
Detailed Description
The embodiment of the application provides a user behavior prediction method, a user behavior prediction system and a user behavior prediction device, which are used for monitoring and predicting user behaviors and have high accuracy.
With the generation of credit business and the development and popularization of personal mini-loan business, various financial institutions issue diversified credit products, the number of users who buy the credit products is also increased rapidly, and monitoring and risk prevention of user behaviors become a vital part in a credit product service link.
Currently, a traditional data monitoring method is generally adopted for monitoring user behavior data, various data indexes are set for the user behavior data, threshold values are set for the various indexes, whether the user behavior data meet the preset index threshold value or not is judged according to the user behavior data monitored in real time, and whether abnormal behavior data exist or not is judged, so that whether abnormal user behavior exists or not is confirmed.
However, the conventional data monitoring method generally cannot distinguish the user types, for example, the data difference of the same index of different types of users is large, and if the threshold is not set according to the user type, but the thresholds of the same index of all the user types are considered in a short time, the misjudgment is easily caused. If different threshold values are set according to different types of users, the classification is complicated due to various types and indexes of the users, or the situation that the classification rule is defined artificially is inaccurate and unreasonable exists.
Based on this, the embodiment of the present application provides a user behavior prediction method combining a machine learning algorithm and a clustering algorithm, please refer to fig. 1, and an implementation manner of the user behavior prediction method provided in the embodiment of the present application includes steps 101 to 104.
101. And obtaining a target prediction model.
The target prediction model is pre-established, specifically, the target prediction model is obtained by performing machine learning training on user tag data and user historical behavior data, a first corresponding relation between a user tag and user behavior characteristics is stored in the target prediction model, and the user behavior characteristics are obtained by performing machine learning on the user historical behavior data.
Specifically, the user tag data is obtained by clustering user feature data to obtain a plurality of user clusters having similar user features, and setting the same user tag for users of the same user cluster. For example, the user characteristic data may include: the user registration place, the user work city, the user age, the user gender, the user occupation, the user credit score, the user income, and the like, which are not limited herein. And naming the user tags of the user cluster according to the characteristics of the same user cluster to obtain user tag data. Specifically, the clustering algorithm may adopt any one of clustering algorithms such as a DBSCAN algorithm, a kmeans algorithm, a mean shift, a gaussian mixture model, and the like, and may also adopt a machine learning algorithm supporting vector clustering, which is not limited herein.
The user behavior characteristics are obtained by performing machine learning training on the historical behavior data of the user, and for the users with the same user label, the user behavior characteristics can be obtained by performing machine learning training on the historical behavior data, so that the target user behavior is predicted according to the user behavior characteristics. The target prediction model stores a first corresponding relation between the user label and the user behavior characteristic.
102. And acquiring target user behavior data.
And determining the user of the user behavior to be predicted as a target user, and acquiring the target user behavior data in real time by adopting a streaming processing mode so as to perform behavior prediction on the target user according to the target user behavior data.
103. And acquiring a target user label.
And determining a user tag of the target user as a target user tag according to the user tag data so as to perform behavior prediction on the target user according to the target user tag.
104. And inputting the target user behavior data and the target user label into the target prediction model to obtain user predicted behavior data output by the target prediction model according to the first corresponding relation and the user behavior characteristics.
And inputting the target user behavior data and the target user label into a target prediction model, determining which user behavior characteristics should be adopted by the target user for behavior prediction according to the first corresponding relation, and outputting user predicted behavior data.
In the embodiment, the user behavior is predicted by establishing the target prediction model, the user is classified by setting the user label for the user, the target user behavior data obtained by real-time monitoring is input into the target prediction model, the user predicted behavior data output by the target prediction model according to the first corresponding relation between the user label and the user behavior characteristic is obtained, and the accuracy is high.
Referring to fig. 2, an implementation manner of the user behavior prediction method provided in the embodiment of the present application includes steps 201 to 207.
201. User tag data is obtained.
The user label data is obtained by clustering the user characteristic data to obtain a plurality of user clusters with similar user characteristics and setting the same user label for the users of the same user cluster. For example, the user characteristic data may include: the user registration place, the user work city, the user age, the user gender, the user occupation, the user credit score, the user income, and the like, which are not limited herein. And naming the user tags of the user cluster according to the characteristics of the same user cluster to obtain user tag data. Specifically, the clustering algorithm may adopt any one of clustering algorithms such as a DBSCAN algorithm, a kmeans algorithm, a mean shift, a gaussian mixture model, and the like, and may also adopt a machine learning algorithm supporting vector clustering, which is not limited herein.
The method for obtaining the user tag data by obtaining the clustering model through the machine learning algorithm specifically comprises the following steps:
and a user clustering model is pre-established, the user clustering model is obtained by performing machine learning training on the user characteristic data and the user label data, and a second corresponding relation between the user characteristic data and the user label data is stored in the user clustering model. And inputting the target user characteristic data of the target user of the user behavior to be predicted into the user clustering model to obtain a target user label output by the user clustering model according to the second corresponding relation.
And, stream processing is carried out on the user characteristic data, so that the types of the user clusters and the user label data are updated in real time.
202. And establishing a target prediction model.
The target prediction model is pre-established, specifically, the target prediction model is obtained by performing machine learning training on user tag data and user historical behavior data, a first corresponding relation between a user tag and user behavior characteristics is stored in the target prediction model, and the user behavior characteristics are obtained by performing machine learning on the user historical behavior data.
The user behavior characteristics are obtained by performing machine learning training on the historical behavior data of the user. And obtaining the historical behavior data of the user in a streaming mode, and obtaining the behavior characteristics of the user by performing machine learning training on the historical behavior data of the user with the same user label, so that the target user behavior is predicted according to the behavior characteristics of the user. The target prediction model stores a first corresponding relation between the user label and the user behavior characteristic.
203. And acquiring target user behavior data.
And determining the user of the user behavior to be predicted as a target user, and acquiring the target user behavior data in real time by adopting a streaming processing mode so as to perform behavior prediction on the target user according to the target user behavior data.
204. And acquiring a target user label.
And determining a user tag of the target user as a target user tag according to the user tag data so as to perform behavior prediction on the target user according to the target user tag.
205. And inputting the target user behavior data and the target user label into the target prediction model to obtain user predicted behavior data output by the target prediction model according to the first corresponding relation and the user behavior characteristics.
And inputting the target user behavior data and the target user label into a target prediction model, determining which user behavior characteristics should be adopted by the target user for behavior prediction according to the first corresponding relation, and outputting user predicted behavior data.
206. And acquiring actual behavior data of the user, monitoring in real time, and judging whether behavior abnormity exists.
The user behavior is monitored in real time, user behavior data are mainly obtained through real-time streaming, and when the deviation between the user actual behavior data and the user predicted behavior data does not meet a user behavior threshold value, the fact that the user actual behavior is abnormal is confirmed. The user behavior threshold value is preset through user behavior characteristics obtained through machine learning and manual experience.
207. And if the behavior is abnormal, marking the actual behavior of the target user as the abnormal behavior of the user.
And when the user actual behavior of the target user is abnormal, marking the user actual behavior as the user abnormal behavior, and storing the user abnormal behavior in the user behavior data of the target user. The abnormal behavior may be marked by streaming data and stored in a database, or may be marked by alarming the real-time behavior of the user and notifying the user or a monitor.
In the embodiment, the user behavior is predicted by establishing the target prediction model, the user is classified by setting the user label for the user, the target user behavior data obtained by real-time monitoring is input into the target prediction model, the user predicted behavior data output by the target prediction model according to the first corresponding relation between the user label and the user behavior characteristic is obtained, whether the actual behavior of the user is abnormal or not can be judged according to the obtained user predicted behavior, the abnormal behavior is marked, and the accuracy is high.
Referring to fig. 3, a user behavior prediction system provided in an embodiment of the present application includes:
an obtaining unit 301, configured to obtain a target prediction model, where the target prediction model is obtained by performing machine learning training on user tag data and user historical behavior data, a first corresponding relationship between a user tag and a user behavior feature is stored in the target prediction model, and the user behavior feature is obtained by performing machine learning on the user historical behavior data;
the acquiring unit 301 is further configured to acquire target user behavior data;
the acquiring unit 301 is further configured to acquire a target user tag;
the output unit 302 is configured to input the target user behavior data and the target user label into the target prediction model, so as to obtain user predicted behavior data output by the target prediction model according to the first corresponding relationship and the user behavior characteristics.
A clustering unit 303, configured to cluster users according to the user feature data to obtain multiple user clusters with similar user features, and set the same user tag for users in the same user cluster to obtain user tag data;
the obtaining unit 301 is specifically configured to determine a user tag of a target user of a user behavior to be predicted as a target user tag.
The clustering unit 303 is further configured to cluster the users according to the user feature data by using a clustering algorithm to obtain a plurality of user clusters with similar user features;
and carrying out streaming processing on the user characteristic data so as to update the types of the user clusters and the user tag data in real time.
The obtaining unit 301 is specifically configured to obtain a user clustering model, where the user clustering model is obtained by performing machine learning training on user feature data and user tag data, and a second corresponding relationship between the user feature data and the user tag data is stored in the user clustering model;
the obtaining of the target user tag comprises:
and inputting the target user characteristic data of the target user of the user behavior to be predicted into the user clustering model to obtain a target user label output by the user clustering model according to the second corresponding relation.
The obtaining unit 301 is specifically configured to obtain user historical behavior data in a streaming manner;
acquiring user tag data;
the user historical behavior data and the user label data are used as training samples to be input into an initial prediction model, the training samples are used for conducting machine learning training on the initial prediction model to obtain a target prediction model, a first corresponding relation between a user label and user behavior characteristics is stored in the target prediction model, and the user behavior characteristics are obtained through conducting machine learning on the user historical behavior data.
The user behavior prediction system further includes:
and the monitoring unit 304 is configured to monitor the user behavior in real time, and when a deviation between the actual user behavior and the predicted user behavior data does not satisfy a user behavior threshold, determine that the actual user behavior is abnormal.
The user behavior prediction system further includes:
and a marking unit 305, configured to mark the user actual behavior as a user abnormal behavior when there is an abnormal user actual behavior of the target user, and store the user abnormal behavior in the user behavior data of the target user.
The functions and processes executed by each unit in the user behavior prediction system of this embodiment are similar to those executed by the user behavior system in fig. 1 to fig. 2, and are not repeated here.
Fig. 4 is a schematic structural diagram of a user behavior prediction apparatus provided in the present application, where the user behavior prediction apparatus 400 may include one or more Central Processing Units (CPUs) 401 and a memory 405, and the memory 405 stores one or more applications or data.
The user behavior prediction device 400 may also include one or more power supplies 402, one or more wired or wireless network interfaces 403, one or more input-output interfaces 404, and/or one or more operating systems, such as Windows ServerTM, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM, etc.
The central processing unit 401 may perform the operations performed by the user behavior prediction system in the embodiments shown in fig. 1 to fig. 2, which are not described herein again.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other manners. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is only one logical division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one position, or may be distributed on multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present application 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, and can also be realized in a form of a software functional unit.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application may be substantially implemented or contributed to by the prior art, or all or part of the technical solution may be embodied in a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a read-only memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and the like.
Claims (10)
1. A user behavior prediction method is characterized by comprising the following steps:
obtaining a target prediction model, wherein the target prediction model is obtained by performing machine learning training on user tag data and user historical behavior data, a first corresponding relation between a user tag and user behavior characteristics is stored in the target prediction model, and the user behavior characteristics are obtained by performing machine learning on the user historical behavior data;
acquiring target user behavior data;
acquiring a target user label;
and inputting the target user behavior data and the target user label into the target prediction model to obtain user predicted behavior data output by the target prediction model according to the first corresponding relation and the user behavior characteristics.
2. The method of claim 1, wherein prior to the obtaining the target user label, the method further comprises:
acquiring user characteristic data;
clustering users according to the user characteristic data to obtain a plurality of user clusters with similar user characteristics, and setting the same user label for the users in the same user cluster to obtain user label data;
the acquiring of the target user tag comprises:
and confirming the user label of the target user of the user behavior to be predicted as the target user label.
3. The method according to claim 2, wherein the clustering users according to the user characteristic data to obtain a plurality of user clusters with similar user characteristics, and setting the same user tag for users in the same user cluster, and the obtaining of the user tag data comprises:
clustering users according to the user characteristic data by adopting a clustering algorithm to obtain a plurality of user clusters with similar user characteristics;
and carrying out streaming processing on the user characteristic data so as to update the types of the user clusters and the user label data in real time.
4. The method of claim 1, wherein prior to the obtaining the target user label, the method further comprises:
obtaining a user clustering model, wherein the user clustering model is obtained by performing machine learning training on the user characteristic data and the user label data, and a second corresponding relation between the user characteristic data and the user label data is stored in the user clustering model;
the acquiring of the target user tag comprises:
and inputting the target user characteristic data of the target user of the user behavior to be predicted into the user clustering model to obtain a target user label output by the user clustering model according to the second corresponding relation.
5. The method of claim 1, wherein prior to the obtaining the objective prediction model, the method further comprises:
obtaining user historical behavior data in a streaming mode;
acquiring user tag data;
and inputting the user historical behavior data and the user label data into an initial prediction model as training samples, performing machine learning training on the initial prediction model by using the training samples to obtain a target prediction model, wherein a first corresponding relation between a user label and a user behavior characteristic is stored in the target prediction model, and the user behavior characteristic is obtained by performing machine learning on the user historical behavior data.
6. The method according to any one of claims 1 to 5, wherein after obtaining the user predicted behavior data output by the target prediction model according to the first corresponding relationship and the user behavior feature, the method further comprises:
and monitoring the user behavior in real time, and confirming that the user actual behavior is abnormal when the deviation of the user actual behavior data and the user predicted behavior data does not meet a user behavior threshold value.
7. The method of claim 6, wherein after confirming that the user actual behavior is abnormal when the deviation of the user actual behavior data from the user predicted behavior data does not satisfy the user behavior threshold, the method further comprises:
when the user actual behavior of the target user is abnormal, the user actual behavior is marked as the user abnormal behavior and is stored in the user behavior data of the target user.
8. A user behavior prediction system, comprising:
the target prediction model is obtained by performing machine learning training on user label data and user historical behavior data, a first corresponding relation between a user label and user behavior characteristics is stored in the target prediction model, and the user behavior characteristics are obtained by performing machine learning on the user historical behavior data;
the acquisition unit is also used for acquiring target user behavior data;
the acquisition unit is also used for acquiring a target user label;
and the output unit is used for inputting the target user behavior data and the target user label into the target prediction model to obtain user predicted behavior data output by the target prediction model according to the first corresponding relation and the user behavior characteristics.
9. A user behavior prediction apparatus, comprising:
the system comprises a central processing unit, a memory and an input/output interface;
the memory is a transient memory or a persistent memory;
the central processor is configured to communicate with the memory and execute the operations of the instructions in the memory to perform the method of any of claims 1 to 7.
10. A computer-readable storage medium comprising instructions that, when executed on a computer, cause the computer to perform the method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210126495.7A CN114463117A (en) | 2022-02-10 | 2022-02-10 | User behavior prediction method, system and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210126495.7A CN114463117A (en) | 2022-02-10 | 2022-02-10 | User behavior prediction method, system and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114463117A true CN114463117A (en) | 2022-05-10 |
Family
ID=81413405
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210126495.7A Pending CN114463117A (en) | 2022-02-10 | 2022-02-10 | User behavior prediction method, system and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114463117A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116205376A (en) * | 2023-04-27 | 2023-06-02 | 北京阿帕科蓝科技有限公司 | Behavior prediction method, training method and device of behavior prediction model |
-
2022
- 2022-02-10 CN CN202210126495.7A patent/CN114463117A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116205376A (en) * | 2023-04-27 | 2023-06-02 | 北京阿帕科蓝科技有限公司 | Behavior prediction method, training method and device of behavior prediction model |
| CN116205376B (en) * | 2023-04-27 | 2023-10-17 | 北京阿帕科蓝科技有限公司 | Behavior prediction method, training method and device of behavior prediction model |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107918905B (en) | Abnormal transaction identification method and device and server | |
| US20150178825A1 (en) | Methods and Apparatus for Quantitative Assessment of Behavior in Financial Entities and Transactions | |
| CN108711107A (en) | Intelligent financing services recommend method and its system | |
| CN110909165A (en) | Data processing method, device, medium and electronic equipment | |
| CN107403311B (en) | Account use identification method and device | |
| CN111090807A (en) | Knowledge graph-based user identification method and device | |
| CN114186626A (en) | Abnormity detection method and device, electronic equipment and computer readable medium | |
| CN110852761B (en) | Method and device for formulating anti-cheating strategy and electronic equipment | |
| CN114463117A (en) | User behavior prediction method, system and device | |
| CN108985755B (en) | Account state identification method and device and server | |
| US11755929B2 (en) | System and method for time series pattern recognition | |
| CN116610821B (en) | Knowledge graph-based enterprise risk analysis method, system and storage medium | |
| CN108073464A (en) | A kind of time series data abnormal point detecting method and device based on speed and acceleration | |
| US11847599B1 (en) | Computing system for automated evaluation of process workflows | |
| CN116578925A (en) | Behavior prediction method, device and storage medium based on feature images | |
| CN115719244A (en) | User behavior prediction method and device | |
| CN115204322A (en) | Behavioral link abnormity identification method and device | |
| CN111723872B (en) | Pedestrian attribute identification method and device, storage medium and electronic device | |
| CN111681090A (en) | Account grouping method and device of business system, terminal equipment and storage medium | |
| CN111882339A (en) | Prediction model training and response rate prediction method, device, equipment and storage medium | |
| CN111815442B (en) | Link prediction method and device and electronic equipment | |
| CN114283018A (en) | Comprehensive risk assessment method and device | |
| US20230306432A1 (en) | Transmitting notifications based on machine learning model outputs | |
| US20230099904A1 (en) | Machine learning model prediction of interest in an object | |
| CN112329838A (en) | Method and device for determining category label of target set |
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 |