CN113628748A - Method, device and equipment for evaluating risk bearing tendency of user and storage medium - Google Patents
Method, device and equipment for evaluating risk bearing tendency of user and storage medium Download PDFInfo
- Publication number
- CN113628748A CN113628748A CN202110937324.8A CN202110937324A CN113628748A CN 113628748 A CN113628748 A CN 113628748A CN 202110937324 A CN202110937324 A CN 202110937324A CN 113628748 A CN113628748 A CN 113628748A
- Authority
- CN
- China
- Prior art keywords
- user
- evaluation model
- behavior data
- tendency
- target
- 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
- 238000000034 method Methods 0.000 title claims abstract description 78
- 238000013210 evaluation model Methods 0.000 claims abstract description 183
- 238000012549 training Methods 0.000 claims abstract description 118
- 230000008569 process Effects 0.000 claims abstract description 34
- 230000006399 behavior Effects 0.000 claims description 204
- 238000003066 decision tree Methods 0.000 claims description 61
- 238000013145 classification model Methods 0.000 claims description 52
- 238000004590 computer program Methods 0.000 claims description 15
- 238000010606 normalization Methods 0.000 claims description 13
- 230000000694 effects Effects 0.000 claims description 12
- 238000004422 calculation algorithm Methods 0.000 claims description 9
- 238000012216 screening Methods 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 4
- 238000011156 evaluation Methods 0.000 abstract description 17
- 238000013473 artificial intelligence Methods 0.000 abstract description 7
- 238000005516 engineering process Methods 0.000 description 8
- 238000007637 random forest analysis Methods 0.000 description 8
- 230000009467 reduction Effects 0.000 description 7
- 238000012360 testing method Methods 0.000 description 6
- 230000009471 action Effects 0.000 description 4
- 238000007621 cluster analysis Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 3
- 238000010801 machine learning Methods 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 238000007781 pre-processing Methods 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 230000003542 behavioural effect Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000037213 diet Effects 0.000 description 1
- 235000005911 diet Nutrition 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 230000000391 smoking effect Effects 0.000 description 1
- 230000011273 social behavior Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment
-
- 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
- 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
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Public Health (AREA)
- Medical Informatics (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- Primary Health Care (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Pathology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Epidemiology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Databases & Information Systems (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application relates to the technical field of artificial intelligence, and provides a method, a device, equipment and a storage medium for evaluating a user risk bearing tendency, wherein the method comprises the following steps: extracting behavior data of a user who makes a query paper in a preset time range from a target platform; constructing an evaluation model for evaluating the risk bearing tendency of a user, and training the evaluation model by using behavior data extracted from a target platform to obtain a trained evaluation model; and acquiring the target behavior data of each user from the target platform, and calling the trained evaluation model to process the target behavior data of each user to obtain the risk bearing tendency of each user. According to the method and the device, the evaluation model can be used for automatically evaluating the risk bearing tendency of the user according to the target behavior data filled in the questionnaire process of the user, and the evaluation efficiency of the risk bearing tendency of the user is improved.
Description
Technical Field
The application relates to the technical field of artificial intelligence, in particular to a method, a device, equipment and a storage medium for evaluating a user risk bearing tendency.
Background
The risk tolerance tendency refers to the tendency of an individual to risk attitude and risk when the individual is in the face of uncertainty, and the risk tolerance tendency affects social behaviors such as investment, professional choice (whether or not to create business) and the like and also affects physical health related habits such as diet, smoking, exercise and the like. The difference between risk tolerance and risk tolerance is the subjective acceptance of risk by the user and the objective acceptance of risk (related to age, income, presence or absence of intrinsic assets, etc.). The two values can also change along with time, and the knowledge of the risk bearing tendency of the user is particularly important for the financial platform, so that the user can be better understood, and different standard financial products can be recommended to different users by using the platform.
However, in the existing method, the risk tolerance tendency of the user is evaluated in a manual questionnaire manner, and after the user fills in the questionnaire, statistics needs to be performed on each option of the questionnaire to obtain an evaluation result, so that the risk tolerance tendency evaluation method provided by the prior art has a problem of low risk evaluation efficiency.
Disclosure of Invention
The main purpose of the present application is to provide a method, an apparatus, a device and a storage medium for evaluating a user risk tolerance, so as to improve the efficiency of evaluating the user risk tolerance.
In order to achieve the above object, the present application provides a method for evaluating a risk tolerance tendency of a user, comprising the steps of:
extracting behavior data of a user who makes a query paper in a preset time range from a target platform;
constructing an evaluation model for evaluating the risk bearing tendency of a user, carrying out normalization pretreatment on the extracted behavior data, establishing a training set of the evaluation model, randomly extracting K training samples from the training set of the evaluation model, and training the evaluation model based on the K training samples to obtain a trained evaluation model; wherein K is a positive integer;
and acquiring target behavior data of each user from the target platform, and calling the trained evaluation model to process the target behavior data of each user to obtain the risk bearing tendency of each user.
Preferably, the evaluation model includes a decision tree classification model, and the step of training the evaluation model based on the K training samples to obtain a trained evaluation model includes:
training the K training samples based on a decision tree algorithm to obtain K first decision tree classification models;
and combining the K first decision tree classification models to obtain a trained evaluation model.
Further, after the step of obtaining the trained evaluation model, the method further includes:
extracting incremental user behavior data from the target platform at regular time;
training T second decision tree classification models by using the incremental user behavior data; wherein T is a positive integer and is less than K;
verifying the T second decision tree classification models and K first decision tree classification models contained in the currently trained evaluation model by using the incremental user behavior data to determine the classification effect of each decision tree classification model;
sorting the T second decision tree classification models and the K first decision tree classification models from high to low based on the classification effect, and screening K decision tree classification models with the classification effect arranged at the top K positions;
and combining the screened K first decision tree classification models to obtain a new evaluation model, and replacing the trained evaluation model with the new evaluation model.
Further, after the step of calling the trained evaluation model to process the target behavior data of each user to obtain the risk tolerance tendency of each user, the method further includes:
acquiring target incremental user behavior data of a plurality of users regularly pushed by the target platform;
generating the latest target behavior data of each user according to the target increment user behavior data of each user and the stock behavior data of each user;
and calling the new evaluation model to process the latest target behavior data of each user to obtain the latest risk bearing tendency of each user, and correspondingly replacing the risk bearing tendency of each user with the latest risk bearing tendency.
Preferably, the step of generating the latest target behavior data of each user according to the target incremental user behavior data of each user and the stock behavior data of each user includes:
determining behavior identifications corresponding to behavior records contained in target increment user behavior data of each user;
updating corresponding behavior records contained in the stock behavior data of each user according to the behavior identification corresponding to each behavior record to obtain the latest behavior data of each user;
and screening the latest target behavior data from the latest behavior data of each user.
Preferably, the step of calling the trained evaluation model to process the target behavior data of each user to obtain a risk tolerance tendency of each user includes:
determining a user risk bearing tendency grade of each user according to the risk bearing tendency, and calling the trained evaluation model to calculate the joint probability of the user risk bearing tendency grade of the user and the target behavior data;
respectively calculating the conditional probability of the user risk bearing tendency grade under different target behavior data;
and calculating the conditional entropy of the user risk tolerance tendency grades under different target behavior data based on the joint probability and the conditional probability, and generating the risk tolerance tendency based on the conditional entropy of the user risk tolerance tendency grades.
Further, before the step of obtaining the trained evaluation model, the method further includes:
calculating a loss value of the evaluation model after each training;
judging whether the loss value is smaller than a preset loss value or not;
when the loss value is smaller than a preset loss value, judging that the evaluation model completes training;
and when the loss value is determined to be not less than the preset loss value, adjusting parameters of the evaluation model according to the loss value, and returning to the step of executing the training of the evaluation model based on the K training samples, so as to train the evaluation model after parameter adjustment again until the loss value is less than the preset loss value.
The present application also provides an apparatus for evaluating a risk tolerance tendency of a user, comprising:
the extraction module is used for extracting behavior data of a user who makes a query paper in a preset time range from the target platform;
the training module is used for constructing an evaluation model for evaluating the risk bearing tendency of a user, carrying out normalization pretreatment on the extracted behavior data, establishing a training set of the evaluation model, randomly extracting K training samples from the training set of the evaluation model, and training the evaluation model based on the K training samples to obtain a trained evaluation model; wherein K is a positive integer;
and the calling module is used for acquiring the target behavior data of each user from the target platform, calling the trained evaluation model to process the target behavior data of each user, and obtaining the risk bearing tendency of each user.
The present application further provides a computer device comprising a memory and a processor, the memory storing a computer program, the processor implementing the steps of any of the above methods when executing the computer program.
The present application also provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of any of the methods described above.
According to the method, the device, the equipment and the storage medium for evaluating the risk bearing tendency of the user, behavior data of the user who makes a survey questionnaire in a preset time range are extracted from a target platform; constructing an evaluation model for evaluating the risk bearing tendency of a user, carrying out normalization pretreatment on the extracted behavior data, establishing a training set of the evaluation model, randomly extracting K training samples from the training set of the evaluation model, and training the evaluation model based on the K training samples to obtain a trained evaluation model; the target behavior data of each user are acquired from the target platform, the trained evaluation model is called to process the target behavior data of each user, and the risk bearing tendency of each user is obtained, so that the risk bearing tendency of the user can be evaluated automatically according to the target behavior data filled in the questionnaire process of the user by using the evaluation model, the risk bearing tendency of the user is not required to be evaluated in a traditional manual questionnaire mode, the labor cost and the time cost for manufacturing and sorting the questionnaire are reduced, and the evaluation efficiency of the risk bearing tendency of the user is improved.
Drawings
Fig. 1 is a schematic flowchart of a method for evaluating a risk tolerance tendency of a user according to an embodiment of the present application;
FIG. 2 is a block diagram schematically illustrating a structure of an apparatus for evaluating a risk tolerance tendency of a user according to an embodiment of the present disclosure;
fig. 3 is a block diagram illustrating a structure of a computer device according to an embodiment of the present application.
The implementation, functional features and advantages of the objectives of the present application will be further explained with reference to the accompanying drawings.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
The embodiment of the application can acquire and process related data based on an artificial intelligence technology. Among them, Artificial Intelligence (AI) is a theory, method, technique and application system that simulates, extends and expands human Intelligence using a digital computer or a machine controlled by a digital computer, senses the environment, acquires knowledge and uses the knowledge to obtain the best result.
The artificial intelligence infrastructure generally includes technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technologies, operation/interaction systems, mechatronics, and the like. The artificial intelligence software technology mainly comprises a computer vision technology, a robot technology, a biological recognition technology, a voice processing technology, a natural language processing technology, machine learning/deep learning and the like.
Referring to fig. 1, the present application provides a method for evaluating a user risk tolerance, which uses a server as an execution main body, where the server may be an independent server, or a cloud server that provides basic cloud computing services such as cloud service, a cloud database, cloud computing, a cloud function, cloud storage, Network service, cloud communication, middleware service, domain name service, security service, Content Delivery Network (CDN), big data, and an artificial intelligence platform.
In this application, the method for evaluating a user risk tolerance tendency is used to solve the technical problem that the evaluation efficiency of the current user risk tolerance tendency is low, referring to fig. 1, in one embodiment, the method for evaluating a user risk tolerance tendency includes the following steps:
s11, extracting behavior data of the user who makes the inquiry paper in a preset time range from the target platform;
s12, constructing an evaluation model for evaluating the risk bearing tendency of a user, carrying out normalization pretreatment on the extracted behavior data, establishing a training set of the evaluation model, randomly extracting K training samples from the training set of the evaluation model, and training the evaluation model based on the K training samples to obtain a trained evaluation model; wherein K is a positive integer;
s13, acquiring target behavior data of each user from the target platform, and calling the trained evaluation model to process the target behavior data of each user to obtain the risk bearing tendency of each user.
As described in the above step S11, the target platform may be a financial platform, and the user may purchase a financial product and operate the financial product in the platform. After the user performs the questionnaire on the target platform, the target platform may collect the user's behavioral data.
The risk tolerance tendency can be divided into a plurality of grades in advance, for example, the risk tolerance tendency can be divided into 4 grades which are 1,2,3 and 4 from low to high. And the evaluation result of the risk bearing tendency of the user, which is calculated by the evaluation model, is used for representing the grade of the risk bearing tendency of the user.
Wherein the behavior data of the user may include at least the following data:
the average webpage browsing time for a user to purchase a certain financial product for the first time;
the number of products browsed by a user before purchasing a certain financial product for the first time;
the user's warehousing (or redemption) the average web page browsing duration of a financial product (warehousing and redemption are 2 features);
the user inputs the payment password for average time consumption;
the times of annual yield intervals of the financial products when the users are added with the bins (redeemed) relative to the purchases, and the times of the money amount accounting for the specific weight of the position taken by the financial products. For example, the profit rate is … … times with 5 times of-20% -10%, the redemption amount is less than 1 time of 10%, the redemption amount is less than 2 times of 10% -20%, the profit rate is 1 time of 0-2%, the profit rate is 2 times of 2-5%, and the profit rate is 5 times of 5% -10%.
The number of 7-day yield intervals of the financing product when the user adds the warehouse (redeems the financing product), and the number of times of the position holding specific weight of the financing product in terms of money amount; these two rates of return may reflect the actual rate of return of the user and the short term fluctuations of the financial product.
As described in step S12, this step constructs an evaluation model for evaluating the risk tolerance tendency of the user, where the evaluation model is a neural network model, and N rounds of training may be performed on the evaluation model using all the behavior data extracted from the target platform until the training result meets the requirement, so as to obtain a trained evaluation model.
Specifically, the extracted behavior data may be subjected to normalization preprocessing, such as conversion into a uniform data format, or data without substantial meaning in the behavior data may be removed, and then a training set and a test set of the evaluation model may be established according to the normalized and preprocessed behavior data, for example, ninety percent of the behavior data may be used as the training set, and the remaining ten percent may be used as the test set. In addition, the ratio of the capacities between the training set and the test set of the evaluation model can be set according to the requirements of a specific scene, and preferably, the ratio of the capacities can be set to be 8: 2. And then, randomly extracting K training samples from the training set of the evaluation model, and training the evaluation model based on the extracted K training samples, so that the trained evaluation model is obtained, and the training efficiency is improved.
As described in step S13, the target behavior data may be the behavior data closest to the current time, and the target behavior data may be screened from the behavior data, for example, when the behavior data is the user 'S stock data on the target platform in one month, the target behavior data may be the current-day behavior data, and the target behavior data may be used to reflect the user' S stock habits, such as when to buy, when to sell, and the like, and then the trained evaluation model is invoked to process the target behavior data of each user, such as performing cluster analysis on each target behavior data, and determining the risk tolerance tendency of each user according to the result of the cluster analysis.
Further, in the embodiment, a Random Forest algorithm (RF for short) may be selected to construct the evaluation model. In machine learning, a random forest is a classifier that contains multiple decision trees, and the class of its output is determined by the mode of the class output by the individual trees. It is interpreted from an intuitive perspective that each decision tree is a classifier (assuming that the classification problem is now addressed), and then for an input sample, N trees have N classification results. And random forests integrate all classification voting results, and the classification with the largest voting times is designated as final output, so that the simplest Bagging idea is realized. The random forest algorithm has high prediction accuracy, has good tolerance on abnormal values and noise, and is not easy to over-fit, so that the classification problem of the risk bearing tendency of the user can be effectively solved.
According to the method for evaluating the risk bearing tendency of the user, the behavior data of the user who makes the inquiry paper in a preset time range is extracted from the target platform; constructing an evaluation model for evaluating the risk bearing tendency of a user, carrying out normalization pretreatment on the extracted behavior data, establishing a training set of the evaluation model, randomly extracting K training samples from the training set of the evaluation model, and training the evaluation model based on the K training samples to obtain a trained evaluation model; the target behavior data of each user are acquired from the target platform, the trained evaluation model is called to process the target behavior data of each user, and the risk bearing tendency of each user is obtained, so that the risk bearing tendency of the user can be evaluated automatically according to the target behavior data filled in the questionnaire process of the user by using the evaluation model, the risk bearing tendency of the user is not required to be evaluated in a traditional manual questionnaire mode, the labor cost and the time cost for manufacturing and sorting the questionnaire are reduced, and the evaluation efficiency of the risk bearing tendency of the user is improved.
In an embodiment, in step S12, the step of training the evaluation model based on the K training samples to obtain a trained evaluation model includes:
s121, training the K training samples based on a decision tree algorithm to obtain K first decision tree classification models;
and S122, combining the K first decision tree classification models to obtain a trained evaluation model.
As described in step S121, in this step, K training samples may be trained based on a decision tree algorithm to obtain K first decision tree classification models. When the K training samples are extracted, behavior data close to a time node can be used as one training sample, for example, behavior data in the same month can be used as one training sample, and when the training samples are trained based on a decision tree algorithm, each obtained decision tree classification model can well reflect the user risk bearing tendency of the time node.
As described in step S122 above, the K first decision tree classification models may be randomly combined to obtain a trained evaluation model, for example, the K first decision tree classification models may be combined into the trained evaluation model, or at least two of the K first decision tree classification models may be selected and combined into the trained evaluation model, so that the evaluation model obtained by combining the separately trained decision tree classification models has a better training result, and the evaluation effect of the risk tolerance tendency of the user is improved.
In another embodiment, weights are respectively set for each decision tree classification model, and a combined model is constructed; wherein an accumulated value of each of the weights is 1. When the evaluation error of the combined model is minimum, the optimal weight of each decision tree classification model is determined, and the combined model with the minimum evaluation error is used as a trained evaluation model, so that the condition that the evaluation result is incorrect due to a single decision tree classification model can be avoided, and the accuracy of the evaluation result can be greatly improved.
In an embodiment, after the step of obtaining the trained evaluation model, the method may further include:
extracting incremental user behavior data from the target platform at regular time;
training T second decision tree classification models by using the incremental user behavior data; wherein T is a positive integer and is less than K;
verifying the T second decision tree classification models and K first decision tree classification models contained in the currently trained evaluation model by using the incremental user behavior data to determine the classification effect of each decision tree classification model;
sorting the T second decision tree classification models and the K first decision tree classification models from high to low based on the classification effect, and screening K decision tree classification models with the classification effect arranged at the top K positions;
and combining the screened K first decision tree classification models to obtain a new evaluation model, and replacing the trained evaluation model with the new evaluation model.
In the embodiment, considering that the behavior rule of the user usually changes over time, the prediction result of the initially trained evaluation model is more and more deviated from the actual situation, in order to obtain an evaluation model with good and even excellent prediction results, the whole amount of user behavior data is extracted from a target platform in the past, and retraining the evaluation model based on the full amount of user behavior data, however, in the way of retraining the evaluation model by using the full amount of user behavior data, along with the increase of the data amount of the user behavior data, the time spent for training the evaluation model is correspondingly increased, the computing resources required for training the evaluation model are correspondingly increased, which results in the lower and lower efficiency of training the evaluation model, and in this way, corresponding adjustment cannot be made in time according to the change of the user behavior data, and related services cannot be responded in time.
Therefore, in the embodiment, incremental user behavior data (which may be behavior data of a newly added user) is regularly pulled from a target platform, and a part of newly added decision tree classification models are trained based on the incremental user behavior data, for example, a trained evaluation model is composed of 300 first decision tree classification models, so that 150 new second decision tree classification models can be trained each time updating is performed, then all decision tree classification models (that is, 450 decision tree classification models composed of 300 first decision tree classification models and 150 second decision tree classification models) are verified to obtain classification effects of each decision tree classification model, then 300 decision tree classification models are selected from the 450 decision tree classification models, and the 300 decision tree classification models are combined to obtain a new evaluation model.
Although the prediction effect of the new evaluation model is not good as that of the evaluation model trained by the full amount of user behavior data, the embodiment can update the existing evaluation model on the basis of the newly generated user behavior data more timely, so that the updated evaluation model can serve the business more timely and better meet the business requirements.
In an embodiment, after the step of calling the trained evaluation model to process the target behavior data of each user to obtain a risk tolerance tendency of each user, the method further includes:
acquiring target incremental user behavior data of a plurality of users regularly pushed by the target platform;
generating the latest target behavior data of each user according to the target increment user behavior data of each user and the stock behavior data of each user;
and calling the new evaluation model to process the latest target behavior data of each user to obtain the latest risk bearing tendency of each user, and correspondingly replacing the risk bearing tendency of each user with the latest risk bearing tendency.
In the embodiment, in order to ensure that the risk bearing tendency of the user can be updated in time to ensure the timeliness and accuracy of the evaluation, the server acquires the behavior data of the user from the target platform at regular time and calculates the latest risk bearing tendency of the user based on the acquired behavior data, and in order to reduce the data transmission amount, reduce the interface pressure of the target platform, improve the data acquisition speed and further improve the update efficiency of the risk bearing tendency, the behavior data of each user in a preset time range is maintained in the server, and then only the target incremental user behavior data of each user is acquired from the target platform.
And the target incremental user behavior data of the user pushed by the target platform comprises the behavior record of the user which is modified. For example, the action that the user takes an average time to input the payment password is originally 1 second, and after the action is modified, for example, the action is modified to 2 seconds, the target platform pushes the latest data corresponding to the action that the user takes an average time to input the payment password to the server. When the server stores the behavior data of the user, each behavior record in the behavior data of the user may be stored in a key-value form, where the key refers to a behavior identifier, and the value refers to a value corresponding to a behavior, for example, assuming that the behavior identifier of the behavior that the user takes the average time to input the payment password is m1, the key-value of "m 1, 2" indicates that the user takes the average time to input the payment password is 2 seconds.
In an embodiment, the step of generating the latest target behavior data of each user according to the target incremental user behavior data of each user and the stock behavior data of each user includes:
determining behavior identifications corresponding to behavior records contained in target increment user behavior data of each user;
updating corresponding behavior records contained in the stock behavior data of each user according to the behavior identification corresponding to each behavior record to obtain the latest behavior data of each user;
and screening the latest target behavior data from the latest behavior data of each user.
In this embodiment, it is assumed that a behavior of a behavior that a user inputs an average time taken to pay a password is identified as m1, a behavior of a user for purchasing a financial product for the first time for browsing a web page is identified as m2, incremental behavior data of the user are "m 1, 1" and "m 2, 20", and corresponding behavior records in inventory behavior data of the user (i.e., behavior records corresponding to m1 and m 2) are "m 1, 2" and "m 2, 15", and after corresponding behavior records in inventory behavior data of the user are updated according to the incremental behavior data of the user, a value corresponding to m1 is 1, and a value corresponding to m2 is 20 in the inventory behavior data of the user. The stock behavior data of the user can be updated rapidly and timely through the incremental behavior data of the user, and then the latest risk bearing tendency of the user can be calculated through the currently trained evaluation model.
In an embodiment, the step of calling the trained evaluation model to process the target behavior data of each user to obtain a risk tolerance tendency of each user includes:
determining a user risk bearing tendency grade of each user according to the risk bearing tendency, and calling the trained evaluation model to calculate the joint probability of the user risk bearing tendency grade of the user and the target behavior data;
respectively calculating the conditional probability of the user risk bearing tendency grade under different target behavior data;
and calculating the conditional entropy of the user risk tolerance tendency grades under different target behavior data based on the joint probability and the conditional probability, and generating the risk tolerance tendency based on the conditional entropy of the user risk tolerance tendency grades.
In this embodiment, the user risk bearing tendency may be classified into three grades, for example, A, B, C grades, where the grade a indicates that the user risk bearing is the strongest, and the grade C indicates that the user risk bearing is the weakest.
Then, the joint probability of the simultaneous occurrence of each user risk tolerance tendency level and the corresponding target behavior data is queried, for example, when the user risk tolerance tendency level of the user is the lowest (level C), the corresponding target behavior data may be taken as high-speed reduction, and when 90 operations of high-speed reduction and 10 operations of high-speed reduction occur in 100 investment behaviors of the user, the corresponding joint probability is 90%. And meanwhile, respectively calculating the conditional probabilities of the user risk bearing tendency levels under different target behavior data, calculating the user risk bearing tendency level conditional entropy under different target behavior data based on the joint probability and the conditional probability, wherein the risk bearing tendency level conditional entropy is a specific numerical value and is used for quantifying the risk bearing tendency of the user, and the risk bearing tendency can be generated based on the user risk bearing tendency level conditional entropy, so that the user risk bearing tendency of the user is accurately evaluated.
For example, when the target behavior data belonging to the lowest risk tolerance tendency level includes operations of high bin reduction and low bin addition, if 80 operations of high bin reduction, 10 operations of low bin addition and 10 operations of high bin addition occur in 100 investment behaviors of the user, the corresponding joint probability is 90%, the corresponding conditional probability of high bin reduction is 80%, and the corresponding conditional probability of low bin addition is 10%, and after different weights are respectively given to the joint probability and the conditional probability, the conditional entropy of the risk tolerance tendency level of the user corresponding to high bin reduction and the conditional entropy of the risk tolerance tendency level of the user corresponding to low bin addition are calculated.
In an embodiment, before the step of obtaining the trained evaluation model, the method further includes:
calculating a loss value of the evaluation model after each training;
judging whether the loss value is smaller than a preset loss value or not;
when the loss value is smaller than a preset loss value, judging that the evaluation model completes training;
and when the loss value is determined to be not less than the preset loss value, adjusting parameters of the evaluation model according to the loss value, and returning to the step of executing the training of the evaluation model based on the K training samples, so as to train the evaluation model after parameter adjustment again until the loss value is less than the preset loss value.
In this embodiment, after the evaluation model is trained each time, the loss value after the training is completed can be calculated by using the loss function, and when the loss value meets a preset threshold value or is smaller than the preset loss value, that is, meets the requirement, it is indicated that the evaluation model meets the training requirement, and the training of the evaluation model is completed, so as to improve the evaluation accuracy of the evaluation model on the user risk tolerance tendency.
When the loss value is not less than the preset loss value, forward transmission can be performed in a neural network structure of the evaluation model according to the loss value, relevant parameters of the evaluation model are adjusted, the adjusted evaluation model is retrained based on the reset relevant parameters, retraining is calculated until the loss value is less than the preset loss value, and when the loss value meets the preset requirement, parameters of the evaluation model corresponding to the loss value meeting the preset threshold value are finally obtained, so that the training of the evaluation model is finished, and the extraction of entities in the text data by the evaluation model is ensured to meet the requirement.
Referring to fig. 2, an embodiment of the present application further provides an apparatus for evaluating a risk tolerance tendency of a user, including:
the extraction module 11 is configured to extract behavior data of a user who makes a query paper within a preset time range from a target platform;
the training module 12 is configured to construct an evaluation model for evaluating a risk tolerance tendency of a user, perform normalization preprocessing on the extracted behavior data, establish a training set of the evaluation model, randomly extract K training samples from the training set of the evaluation model, and train the evaluation model based on the K training samples to obtain a trained evaluation model; wherein K is a positive integer;
and the calling module 13 is configured to obtain target behavior data of each user from the target platform, and call the trained evaluation model to process the target behavior data of each user, so as to obtain a risk tolerance tendency of each user.
The target platform may be a financial platform in which the user may purchase financial products and operate financial products. After the user performs the questionnaire on the target platform, the target platform may collect the user's behavioral data.
The risk tolerance tendency can be divided into a plurality of grades in advance, for example, the risk tolerance tendency can be divided into 4 grades which are 1,2,3 and 4 from low to high. And the evaluation result of the risk bearing tendency of the user, which is calculated by the evaluation model, is used for representing the grade of the risk bearing tendency of the user.
Wherein the behavior data of the user may include at least the following data:
the average webpage browsing time for a user to purchase a certain financial product for the first time;
the number of products browsed by a user before purchasing a certain financial product for the first time;
the user's warehousing (or redemption) the average web page browsing duration of a financial product (warehousing and redemption are 2 features);
the user inputs the payment password for average time consumption;
the times of annual yield intervals of the financial products when the users are added with the bins (redeemed) relative to the purchases, and the times of the money amount accounting for the specific weight of the position taken by the financial products. For example, the profit rate is … … times with 5 times of-20% -10%, the redemption amount is less than 1 time of 10%, the redemption amount is less than 2 times of 10% -20%, the profit rate is 1 time of 0-2%, the profit rate is 2 times of 2-5%, and the profit rate is 5 times of 5% -10%.
The number of 7-day yield intervals of the financing product when the user adds the warehouse (redeems the financing product), and the number of times of the position holding specific weight of the financing product in terms of money amount; these two rates of return may reflect the actual rate of return of the user and the short term fluctuations of the financial product.
The device also constructs an evaluation model for evaluating the risk bearing tendency of the user, wherein the evaluation model is a neural network model, and N rounds of training can be carried out on the evaluation model by using all the behavior data extracted from the target platform until the training result meets the requirement, so that the trained evaluation model is obtained.
Specifically, the extracted behavior data may be subjected to normalization preprocessing, such as conversion into a uniform data format, or data without substantial meaning in the behavior data may be removed, and then a training set and a test set of the evaluation model may be established according to the normalized and preprocessed behavior data, for example, ninety percent of the behavior data may be used as the training set, and the remaining ten percent may be used as the test set. In addition, the ratio of the capacities between the training set and the test set of the evaluation model can be set according to the requirements of a specific scene, and preferably, the ratio of the capacities can be set to be 8: 2. And then, randomly extracting K training samples from the training set of the evaluation model, and training the evaluation model based on the extracted K training samples, so that the trained evaluation model is obtained, and the training efficiency is improved.
The target behavior data may be the behavior data closest to the current time, and the target behavior data may be screened from the behavior data, for example, when the behavior data is the stock data of the user on the target platform in one month, the target behavior data may be the behavior data of the current day, and the target behavior data may be used to reflect the stock habits of the user, such as when to buy, when to sell, and the like, and then the trained evaluation model is invoked to process the target behavior data of each user, such as performing cluster analysis on each target behavior data, and determining the risk tolerance tendency of each user according to the cluster analysis result.
Further, in the embodiment, a Random Forest algorithm (RF for short) may be selected to construct the evaluation model. In machine learning, a random forest is a classifier that contains multiple decision trees, and the class of its output is determined by the mode of the class output by the individual trees. It is interpreted from an intuitive perspective that each decision tree is a classifier (assuming that the classification problem is now addressed), and then for an input sample, N trees have N classification results. And random forests integrate all classification voting results, and the classification with the largest voting times is designated as final output, so that the simplest Bagging idea is realized. The random forest algorithm has high prediction accuracy, has good tolerance on abnormal values and noise, and is not easy to over-fit, so that the classification problem of the risk bearing tendency of the user can be effectively solved.
As described above, it can be understood that each component of the device for evaluating a user risk tolerance tendency provided in the present application may implement the function of any one of the methods for evaluating a user risk tolerance tendency described above, and the specific structure is not described again.
Referring to fig. 3, an embodiment of the present application further provides a computer device, and an internal structure of the computer device may be as shown in fig. 3. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the computer designed processor is used to provide computational and control capabilities. The memory of the computer device comprises a storage medium and an internal memory. The storage medium stores an operating system, a computer program, and a database. The memory provides an environment for the operation of the operating system and computer programs in the storage medium. The database of the computer device is used for storing behavior data, risk bearing tendency and other data. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a method for assessing risk tolerance tendencies of a user.
The processor executes the method for evaluating the risk tolerance tendency of the user, and comprises the following steps:
extracting behavior data of a user who makes a query paper in a preset time range from a target platform;
constructing an evaluation model for evaluating the risk bearing tendency of a user, carrying out normalization pretreatment on the extracted behavior data, establishing a training set of the evaluation model, randomly extracting K training samples from the training set of the evaluation model, and training the evaluation model based on the K training samples to obtain a trained evaluation model; wherein K is a positive integer;
and acquiring target behavior data of each user from the target platform, and calling the trained evaluation model to process the target behavior data of each user to obtain the risk bearing tendency of each user.
An embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored, and the computer program, when executed by a processor, implements a method for evaluating a risk tolerance tendency of a user, including the steps of:
extracting behavior data of a user who makes a query paper in a preset time range from a target platform;
constructing an evaluation model for evaluating the risk bearing tendency of a user, carrying out normalization pretreatment on the extracted behavior data, establishing a training set of the evaluation model, randomly extracting K training samples from the training set of the evaluation model, and training the evaluation model based on the K training samples to obtain a trained evaluation model; wherein K is a positive integer;
and acquiring target behavior data of each user from the target platform, and calling the trained evaluation model to process the target behavior data of each user to obtain the risk bearing tendency of each user.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by a computer program, which can be stored in a computer-readable storage medium, and can include the processes of the embodiments of the methods described above when the computer program is executed. Any reference to memory, storage, database, or other medium provided herein and used in the examples may include non-volatile and/or volatile memory. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), double-rate SDRAM (SSRSDRAM), Enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and bus dynamic RAM (RDRAM).
To sum up, the most beneficial effect of this application lies in:
according to the method, the device, the equipment and the storage medium for evaluating the risk bearing tendency of the user, behavior data of the user who makes a survey questionnaire in a preset time range are extracted from a target platform; constructing an evaluation model for evaluating the risk bearing tendency of a user, carrying out normalization pretreatment on the extracted behavior data, establishing a training set of the evaluation model, randomly extracting K training samples from the training set of the evaluation model, and training the evaluation model based on the K training samples to obtain a trained evaluation model; the target behavior data of each user are acquired from the target platform, the trained evaluation model is called to process the target behavior data of each user, and the risk bearing tendency of each user is obtained, so that the risk bearing tendency of the user can be evaluated automatically according to the target behavior data filled in the questionnaire process of the user by using the evaluation model, the risk bearing tendency of the user is not required to be evaluated in a traditional manual questionnaire mode, the labor cost and the time cost for manufacturing and sorting the questionnaire are reduced, and the evaluation efficiency of the risk bearing tendency of the user is improved.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, apparatus, article, or method. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, apparatus, article, or method that includes the element.
The above description is only a preferred embodiment of the present application, and not intended to limit the scope of the present application, and all modifications of equivalent structures and equivalent processes, which are made by the contents of the specification and the drawings of the present application, or which are directly or indirectly applied to other related technical fields, are also included in the scope of the present application.
Claims (10)
1. A method for assessing risk tolerance tendencies of a user, comprising:
extracting behavior data of a user who makes a query paper in a preset time range from a target platform;
constructing an evaluation model for evaluating the risk bearing tendency of a user, carrying out normalization pretreatment on the extracted behavior data, establishing a training set of the evaluation model, randomly extracting K training samples from the training set of the evaluation model, and training the evaluation model based on the K training samples to obtain a trained evaluation model; wherein K is a positive integer;
and acquiring target behavior data of each user from the target platform, and calling the trained evaluation model to process the target behavior data of each user to obtain the risk bearing tendency of each user.
2. The method of claim 1, wherein the evaluation model comprises a decision tree classification model, and the step of training the evaluation model based on the K training samples to obtain a trained evaluation model comprises:
training the K training samples based on a decision tree algorithm to obtain K first decision tree classification models;
and combining the K first decision tree classification models to obtain a trained evaluation model.
3. The method of claim 2, wherein the step of obtaining the trained evaluation model is followed by:
extracting incremental user behavior data from the target platform at regular time;
training T second decision tree classification models by using the incremental user behavior data; wherein T is a positive integer and is less than K;
verifying the T second decision tree classification models and K first decision tree classification models contained in the currently trained evaluation model by using the incremental user behavior data to determine the classification effect of each decision tree classification model;
sorting the T second decision tree classification models and the K first decision tree classification models from high to low based on the classification effect, and screening K decision tree classification models with the classification effect arranged at the top K positions;
and combining the screened K decision tree classification models to obtain a new evaluation model, and replacing the trained evaluation model with the new evaluation model.
4. The method according to claim 3, wherein the step of invoking the trained evaluation model to process the target behavior data of each user to obtain a risk tolerance tendency of each user further comprises:
acquiring target incremental user behavior data of a plurality of users regularly pushed by the target platform;
generating the latest target behavior data of each user according to the target increment user behavior data of each user and the stock behavior data of each user;
and calling the new evaluation model to process the latest target behavior data of each user to obtain the latest risk bearing tendency of each user, and correspondingly replacing the risk bearing tendency of each user with the latest risk bearing tendency.
5. The method of claim 4, wherein the step of generating the latest target behavior data for each user based on the target incremental user behavior data for each user and the inventory behavior data for each user comprises:
determining behavior identifications corresponding to behavior records contained in target increment user behavior data of each user;
updating corresponding behavior records contained in the stock behavior data of each user according to the behavior identification corresponding to each behavior record to obtain the latest behavior data of each user;
and screening the latest target behavior data from the latest behavior data of each user.
6. The method according to claim 1, wherein the step of invoking the trained evaluation model to process the target behavior data of each user to obtain a risk tolerance tendency of each user comprises:
determining a user risk bearing tendency grade of each user according to the risk bearing tendency, and calling the trained evaluation model to calculate the joint probability of the user risk bearing tendency grade of the user and the target behavior data;
respectively calculating the conditional probability of the user risk bearing tendency grade under different target behavior data;
and calculating the conditional entropy of the user risk tolerance tendency grades under different target behavior data based on the joint probability and the conditional probability, and generating the risk tolerance tendency based on the conditional entropy of the user risk tolerance tendency grades.
7. The method of claim 1, wherein the step of obtaining the trained evaluation model is preceded by:
calculating a loss value of the evaluation model after each training;
judging whether the loss value is smaller than a preset loss value or not;
when the loss value is smaller than a preset loss value, judging that the evaluation model completes training;
and when the loss value is determined to be not less than the preset loss value, adjusting parameters of the evaluation model according to the loss value, and returning to the step of executing the training of the evaluation model based on the K training samples, so as to train the evaluation model after parameter adjustment again until the loss value is less than the preset loss value.
8. An apparatus for evaluating a risk tolerance tendency of a user, comprising:
the extraction module is used for extracting behavior data of a user who makes a query paper in a preset time range from the target platform;
the training module is used for constructing an evaluation model for evaluating the risk bearing tendency of a user, carrying out normalization pretreatment on the extracted behavior data, establishing a training set of the evaluation model, randomly extracting K training samples from the training set of the evaluation model, and training the evaluation model based on the K training samples to obtain a trained evaluation model; wherein K is a positive integer;
and the calling module is used for acquiring the target behavior data of each user from the target platform, calling the trained evaluation model to process the target behavior data of each user, and obtaining the risk bearing tendency of each user.
9. A computer device, comprising:
a processor;
a memory;
a computer program, wherein the computer program is stored in the memory and configured to be executed by the processor, the computer program being configured to perform the method of assessing a user's risk tolerance propensity according to any one of claims 1 to 7.
10. A computer-readable storage medium, wherein a computer program is stored on the computer-readable storage medium, and when executed by a processor, the computer program implements the method for assessing risk tolerance propensity of a user according to any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110937324.8A CN113628748A (en) | 2021-08-16 | 2021-08-16 | Method, device and equipment for evaluating risk bearing tendency of user and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110937324.8A CN113628748A (en) | 2021-08-16 | 2021-08-16 | Method, device and equipment for evaluating risk bearing tendency of user and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113628748A true CN113628748A (en) | 2021-11-09 |
Family
ID=78385796
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110937324.8A Pending CN113628748A (en) | 2021-08-16 | 2021-08-16 | Method, device and equipment for evaluating risk bearing tendency of user and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113628748A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114816964A (en) * | 2022-06-29 | 2022-07-29 | 深圳竹云科技股份有限公司 | Risk model construction method, risk detection device and computer equipment |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106156809A (en) * | 2015-04-24 | 2016-11-23 | 阿里巴巴集团控股有限公司 | For updating the method and device of disaggregated model |
| CN108665366A (en) * | 2018-04-27 | 2018-10-16 | 平安科技（深圳）有限公司 | Determine method, terminal device and the computer readable storage medium of consumer's risk grade |
| CN108665159A (en) * | 2018-05-09 | 2018-10-16 | 深圳壹账通智能科技有限公司 | A kind of methods of risk assessment, device, terminal device and storage medium |
| CN109359812A (en) * | 2018-09-04 | 2019-02-19 | 深圳壹账通智能科技有限公司 | Finance product recommended method, server and computer readable storage medium |
| CN109636510A (en) * | 2018-11-28 | 2019-04-16 | 阿里巴巴集团控股有限公司 | A kind of determining consumer's risk preference, the recommended method of finance product and device |
| CN109636084A (en) * | 2018-10-22 | 2019-04-16 | 中国平安人寿保险股份有限公司 | Consumer's risk type evaluation method, storage medium and equipment based on big data |
| CN109766454A (en) * | 2019-01-18 | 2019-05-17 | 国家电网有限公司 | A kind of investor's classification method, device, equipment and medium |
| CN110287408A (en) * | 2019-05-23 | 2019-09-27 | 上海拍拍贷金融信息服务有限公司 | A kind of method and system customizing push financial product |
| CN110390587A (en) * | 2018-04-18 | 2019-10-29 | 苏宁易购集团股份有限公司 | A kind of customer evaluation method and system |
| US20200090077A1 (en) * | 2017-05-26 | 2020-03-19 | Alibaba Group Holding Limited | Method for determining risk preference of user, information recommendation method, and apparatus |
| CN111833113A (en) * | 2020-07-27 | 2020-10-27 | 中国平安财产保险股份有限公司 | Product recommendation method, device and equipment based on big data and storage medium |
-
2021
- 2021-08-16 CN CN202110937324.8A patent/CN113628748A/en active Pending
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106156809A (en) * | 2015-04-24 | 2016-11-23 | 阿里巴巴集团控股有限公司 | For updating the method and device of disaggregated model |
| US20200090077A1 (en) * | 2017-05-26 | 2020-03-19 | Alibaba Group Holding Limited | Method for determining risk preference of user, information recommendation method, and apparatus |
| CN110390587A (en) * | 2018-04-18 | 2019-10-29 | 苏宁易购集团股份有限公司 | A kind of customer evaluation method and system |
| CN108665366A (en) * | 2018-04-27 | 2018-10-16 | 平安科技（深圳）有限公司 | Determine method, terminal device and the computer readable storage medium of consumer's risk grade |
| CN108665159A (en) * | 2018-05-09 | 2018-10-16 | 深圳壹账通智能科技有限公司 | A kind of methods of risk assessment, device, terminal device and storage medium |
| CN109359812A (en) * | 2018-09-04 | 2019-02-19 | 深圳壹账通智能科技有限公司 | Finance product recommended method, server and computer readable storage medium |
| CN109636084A (en) * | 2018-10-22 | 2019-04-16 | 中国平安人寿保险股份有限公司 | Consumer's risk type evaluation method, storage medium and equipment based on big data |
| CN109636510A (en) * | 2018-11-28 | 2019-04-16 | 阿里巴巴集团控股有限公司 | A kind of determining consumer's risk preference, the recommended method of finance product and device |
| CN109766454A (en) * | 2019-01-18 | 2019-05-17 | 国家电网有限公司 | A kind of investor's classification method, device, equipment and medium |
| CN110287408A (en) * | 2019-05-23 | 2019-09-27 | 上海拍拍贷金融信息服务有限公司 | A kind of method and system customizing push financial product |
| CN111833113A (en) * | 2020-07-27 | 2020-10-27 | 中国平安财产保险股份有限公司 | Product recommendation method, device and equipment based on big data and storage medium |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114816964A (en) * | 2022-06-29 | 2022-07-29 | 深圳竹云科技股份有限公司 | Risk model construction method, risk detection device and computer equipment |
| CN114816964B (en) * | 2022-06-29 | 2022-09-20 | 深圳竹云科技股份有限公司 | Risk model construction method, risk detection device and computer equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20220036461A1 (en) | Sentiment and rules-based equity analysis using customized neural networks in multi-layer, machine learning-based model | |
| CN109902753B (en) | User recommendation model training method and device, computer equipment and storage medium | |
| CN111897861A (en) | Content recommendation method and device, computer equipment and storage medium | |
| WO2022105525A1 (en) | Method and apparatus for predicting user probability, and computer device | |
| CN111340148B (en) | Training method of business classification model, business classification method and terminal | |
| CN112633962B (en) | Service recommendation method and device, computer equipment and storage medium | |
| CN110991474A (en) | Machine learning modeling platform | |
| CN110781379A (en) | Information recommendation method and device, computer equipment and storage medium | |
| CN101339563A (en) | Interest model update method facing to odd discovery recommendation | |
| CN111967971A (en) | Bank client data processing method and device | |
| CN112149352B (en) | Prediction method for marketing activity clicking by combining GBDT automatic characteristic engineering | |
| CN111797320A (en) | Data processing method, device, equipment and storage medium | |
| CN117495512B (en) | Order data management method, device, equipment and storage medium | |
| CN112819523A (en) | Marketing prediction method combining inner/outer product feature interaction and Bayesian neural network | |
| CN113628748A (en) | Method, device and equipment for evaluating risk bearing tendency of user and storage medium | |
| CN114493686A (en) | Operation content generation and pushing method and device | |
| CN111309994A (en) | User matching method and device, electronic equipment and readable storage medium | |
| CN112990989B (en) | Value prediction model input data generation method, device, equipment and medium | |
| CN113762392A (en) | Financial product recommendation method, device, equipment and medium based on artificial intelligence | |
| CN116739794B (en) | User personalized scheme recommendation method and system based on big data and machine learning | |
| CN113722098A (en) | Server capacity prediction method, device, equipment and medium based on artificial intelligence | |
| CN112819499A (en) | Information transmission method, information transmission device, server and storage medium | |
| CN113850670A (en) | Bank product recommendation method, device, equipment and storage medium | |
| CN113220947A (en) | Method and device for encoding event characteristics | |
| CN116795357B (en) | Processing method, device, equipment and storage medium of business rule |
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 |