CN112949824A - Neural network-based multi-output multi-task feature evaluation method and device and electronic equipment - Google Patents
Neural network-based multi-output multi-task feature evaluation method and device and electronic equipment Download PDFInfo
- Publication number
- CN112949824A CN112949824A CN202110167037.3A CN202110167037A CN112949824A CN 112949824 A CN112949824 A CN 112949824A CN 202110167037 A CN202110167037 A CN 202110167037A CN 112949824 A CN112949824 A CN 112949824A
- Authority
- CN
- China
- Prior art keywords
- information
- user
- feature
- model
- service
- 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
- 238000011156 evaluation Methods 0.000 title claims abstract description 68
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 25
- 238000000034 method Methods 0.000 claims abstract description 52
- 238000003860 storage Methods 0.000 claims description 25
- 239000013598 vector Substances 0.000 claims description 16
- 238000005457 optimization Methods 0.000 claims description 6
- 230000008569 process Effects 0.000 abstract description 4
- 238000012545 processing Methods 0.000 description 19
- 238000010586 diagram Methods 0.000 description 12
- 238000004140 cleaning Methods 0.000 description 10
- 230000000694 effects Effects 0.000 description 6
- 238000010606 normalization Methods 0.000 description 5
- 230000006399 behavior Effects 0.000 description 4
- 238000004891 communication Methods 0.000 description 4
- 238000004590 computer program Methods 0.000 description 4
- 230000003287 optical effect Effects 0.000 description 4
- 238000012549 training Methods 0.000 description 4
- 238000010801 machine learning Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 239000013307 optical fiber Substances 0.000 description 2
- 230000000644 propagated effect Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 238000005111 flow chemistry technique Methods 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 230000003252 repetitive effect Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
Images
Classifications
-
- 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
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/03—Credit; Loans; Processing thereof
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Accounting & Taxation (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- Evolutionary Computation (AREA)
- Finance (AREA)
- Development Economics (AREA)
- Technology Law (AREA)
- Economics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Quality & Reliability (AREA)
- Strategic Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a neural network-based multi-output and multi-task characteristic evaluation method, which is characterized by comprising the following steps of: acquiring historical user information, wherein the historical user information comprises historical user basic information and service performance information; constructing a characteristic evaluation frame model based on historical user information, wherein the frame model comprises a plurality of submodels which are connected with each other and can respectively and independently output corresponding service characteristic evaluation results; acquiring current user basic information; and inputting the basic information of the current user into a feature evaluation framework model to obtain the feature scores of a plurality of different types of services of the current user. The invention can optimize a plurality of groups and various types of labels simultaneously to obtain a model which is more excellent than the traditional method for optimizing a single target value, and meanwhile, the invention can greatly save human resources and improve the efficiency of the whole process in the aspect of online deployment.
Description
Technical Field
The invention relates to the field of computer information processing, in particular to a neural network-based multi-output and multi-task characteristic evaluation method and device, electronic equipment and a computer readable medium.
Background
With the advancement of emerging technologies and the development of economic society, it is becoming more and more common for individual users or enterprise users to receive or provide interconnected financial services, and a feature control method, which is one of the cores for developing financial businesses, is also evolving over the world following the advancement of technologies. In the prior art, a financial service organization respectively models single business target labels according to no need, and then performs combined segmentation according to scores of the models to make a strategy, so that a better effect is achieved in the aspect of wind control.
However, the method also brings some problems, firstly, the multiple service target labels are respectively modeled, the required number is increased, more personnel are required to participate, and the operation cost is not reduced; secondly, a single model can only output the value of a single label generally, and the service performance of the user cannot be seen from the time sequence; thirdly, the strategy is mainly divided manually after the analysis of a single strategy during the strategy making, and the accuracy is still deficient.
The above information disclosed in this background section is only for enhancement of understanding of the background of the disclosure and therefore it may contain information that does not constitute prior art that is already known to a person of ordinary skill in the art.
Disclosure of Invention
In view of this, the present disclosure provides a method, an apparatus, an electronic device, and a computer readable medium for multi-output and multi-task feature evaluation based on a neural network, which use a neural network training model, can optimize multiple groups and multiple types of labels simultaneously, obtain a model more excellent than the traditional optimization of a single target value, and at the same time, because there is only one model, human resources can be greatly saved in the aspect of online deployment, and the efficiency of the whole process is improved.
Additional features and advantages of the disclosure will be set forth in the detailed description which follows, or in part will be obvious from the description, or may be learned by practice of the disclosure.
According to one aspect of the present disclosure, a neural network-based multi-output and multi-task feature evaluation method is provided, the method including: acquiring historical user information, wherein the historical user information comprises historical user basic information and service performance information; constructing a characteristic evaluation frame model based on the historical user information, wherein the frame model comprises a plurality of submodels which are connected with each other and can respectively and independently output corresponding service characteristic evaluation results, the submodels are connected in series, and the output result of the previous submodel is the input information of all the following submodels; acquiring current user basic information; and inputting the basic information of the current user into the feature evaluation framework model to obtain the feature scores of a plurality of different types of services of the current user.
Optionally, the service performance information further includes different types of service performance information and/or a plurality of service performance information of the same service in time sequence.
Optionally, the multiple pieces of service performance information of the same service with time as a sequence are vectors with time as a sequence formed by selecting performance values of the same service at different times.
Optionally, the feature evaluation framework model further comprises: the input information of any submodel comprises the user basic information and the output result information of all submodels before the current submodel.
Optionally, the method further comprises: the characteristic evaluation frame model is a neural network frame built by utilizing tensorflow.
Optionally, the method further comprises: the characteristic evaluation frame model adopts a random gradient descent optimization method with momentum.
Optionally, the method further comprises: and formulating a strategy corresponding to the service according to the output result of the characteristic evaluation framework model.
Optionally, the method further comprises: and cleaning the historical user information and the current user information after acquiring the historical user information and the current user information.
In one aspect of the present disclosure, a multi-output and multi-task feature evaluation apparatus based on a neural network is provided, including: the historical information module is used for acquiring historical user information, and the historical user information comprises historical user basic information and business performance information; the model module is used for constructing a characteristic evaluation framework model based on the historical user information, the framework model comprises a plurality of submodels which are connected with each other and can respectively and independently output corresponding service characteristic evaluation results, the submodels are connected in series, and the output result of the previous submodel is the input information of all the following submodels; the current information module is used for acquiring current user basic information; and the scoring module is used for inputting the basic information of the current user into the feature evaluation framework model to obtain the feature scores of a plurality of different types of services of the current user.
Optionally, the service performance information includes different types of service performance information and/or a plurality of service performance information of the same service in a time sequence.
Optionally, the multiple pieces of service performance information of the same service with time as a sequence are vectors with time as a sequence formed by selecting performance values of the same service at different times.
Optionally, the feature evaluation framework model further comprises: the input information of any submodel comprises the user basic information and the output result information of all submodels before the current submodel.
Optionally, the feature evaluation framework model is a neural network framework built with tensorflow.
Optionally, the feature evaluation framework model adopts a stochastic gradient descent optimization method with momentum.
Optionally, the method further comprises: and the strategy module is used for making a strategy corresponding to the service according to the output result of the characteristic evaluation framework model.
Optionally, the method further comprises: and the cleaning module is used for cleaning the historical user information and the current user information after acquiring the historical user information and the current user information.
According to an aspect of the present disclosure, an electronic device is provided, the electronic device including: one or more processors; storage means for storing one or more programs; when executed by one or more processors, cause the one or more processors to implement a method as above.
According to an aspect of the disclosure, a computer-readable medium is proposed, on which a computer program is stored, which program, when being executed by a processor, carries out the method as above.
According to the user feature identification method, the user feature identification device, the electronic equipment and the computer readable medium, historical user information is obtained, wherein the historical user information comprises historical user basic information and business performance information; constructing a characteristic evaluation frame model based on the historical user information, wherein the frame model comprises a plurality of submodels which are connected with each other and can respectively and independently output corresponding service characteristic evaluation results; acquiring current user basic information; the basic information of the current user is input into the feature evaluation framework model to obtain the feature scores of a plurality of different types of services of the current user, a neural network training model is used, a plurality of groups and a plurality of types of labels can be optimized simultaneously, a model which is more excellent than the traditional method for optimizing a single target value is obtained, and meanwhile, because only one model is provided, the human resources can be greatly saved in the aspect of online deployment, and the efficiency of the whole process is improved.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.
Drawings
In order to make the technical problems solved by the present invention, the technical means adopted and the technical effects obtained more clear, the following will describe in detail the embodiments of the present invention with reference to the accompanying drawings. It should be noted, however, that the drawings described below are only illustrations of exemplary embodiments of the invention, from which other embodiments can be derived by those skilled in the art without inventive faculty.
Fig. 1 is a system block diagram illustrating a neural network-based multi-output multi-task feature evaluation method and apparatus according to an exemplary embodiment.
FIG. 2 is a flow diagram illustrating a neural network-based multi-output multi-tasking feature evaluation method in accordance with an exemplary embodiment.
FIG. 3 is a block diagram illustrating a flow structure of a feature evaluation framework model according to an exemplary embodiment.
Fig. 4 is a block diagram illustrating a neural network-based multi-output multi-tasking feature evaluation apparatus according to an exemplary embodiment.
FIG. 5 is a block diagram illustrating an electronic device in accordance with an example embodiment.
FIG. 6 is a block diagram illustrating a computer-readable medium in accordance with an example embodiment.
Detailed Description
Exemplary embodiments of the present invention will now be described more fully with reference to the accompanying drawings. The exemplary embodiments, however, may be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these exemplary embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the concept of the invention to those skilled in the art. The same reference numerals denote the same or similar elements, components, or parts in the drawings, and thus their repetitive description will be omitted.
Features, structures, characteristics or other details described in a particular embodiment do not preclude the fact that the features, structures, characteristics or other details may be combined in a suitable manner in one or more other embodiments in accordance with the technical idea of the invention.
In describing particular embodiments, the present invention has been described with reference to features, structures, characteristics or other details that are within the purview of one skilled in the art to provide a thorough understanding of the embodiments. One skilled in the relevant art will recognize, however, that the invention may be practiced without one or more of the specific features, structures, characteristics, or other details.
The flow charts shown in the drawings are merely illustrative and do not necessarily include all of the contents and operations/steps, nor do they necessarily have to be performed in the order described. For example, some operations/steps may be decomposed, and some operations/steps may be combined or partially combined, so that the actual execution sequence may be changed according to the actual situation.
The block diagrams shown in the figures are functional entities only and do not necessarily correspond to physically separate entities. I.e. these functional entities may be implemented in the form of software, or in one or more hardware modules or integrated circuits, or in different networks and/or processor means and/or microcontroller means.
It will be understood that, although the terms first, second, third, etc. may be used herein to describe various elements, components, or sections, these terms should not be construed as limiting. These phrases are used to distinguish one from another. For example, a first device may also be referred to as a second device without departing from the spirit of the present invention.
The term "and/or" and/or "includes any and all combinations of one or more of the associated listed items.
Fig. 1 is a system block diagram illustrating a neural network-based multi-output multi-task feature evaluation method and apparatus according to an exemplary embodiment.
As shown in fig. 1, the system architecture 10 may include terminal devices 101, 102, 103, a network 104, and a server 105. The network 104 serves as a medium for providing communication links between the terminal devices 101, 102, 103 and the server 105. Network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, to name a few.
The user may use the terminal devices 101, 102, 103 to interact with the server 105 via the network 104 to receive or send messages or the like. The terminal devices 101, 102, 103 may have various communication client applications installed thereon, such as a financial services application, a shopping application, a web browser application, an instant messaging tool, a mailbox client, social platform software, and the like.
The terminal devices 101, 102, 103 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop portable computers, desktop computers, and the like.
The server 105 may be a server that provides various services, such as a background management server that supports financial services websites browsed by the user using the terminal apparatuses 101, 102, and 103. The background management server may analyze and perform other processing on the received user data, and feed back a processing result (e.g., a trained user feature model or feature scoring performed on the user by the user feature model) to an administrator of the financial service website.
The server 105 may, for example, obtain historical user information, which includes historical user base information and business performance information; the server 105 may construct a feature evaluation framework model based on the historical user information, for example, where the framework model includes a plurality of sub-models that are connected to each other and are capable of outputting corresponding service feature evaluation results independently; the server 105 may, for example, obtain current user base information; the server 105 may, for example, input the current user basic information into the feature evaluation framework model to obtain feature scores of a plurality of different types of services of the current user.
It should be noted that the user feature identification method provided by the embodiment of the present disclosure may be executed by the server 105, and accordingly, the user feature identification device may be disposed in the server 105. And the web page side or application side provided for the user to browse the financial service platform is generally located in the terminal equipment 101, 102, 103.
FIG. 2 is a flow diagram illustrating a neural network-based multi-output multi-tasking feature evaluation method in accordance with an exemplary embodiment. The multi-output and multi-task feature evaluation method based on the neural network at least comprises the steps S201 to S206.
As shown in fig. 2, in step S201, historical user information is acquired, which includes historical user basic information and business performance information.
More specifically, the historical user basic information includes, but is not limited to, user-related information necessary for performing a business, such as user identity information, behavior information, financial information, etc. required in performing user feature management.
The business performance data includes but is not limited to information embodied by the user when the user actually completes the business, such as overdue condition information, movable and supported condition information, repayment condition information, financing condition information, profit condition under the credit line and the like of the user in financial activities.
In step S202, data cleaning and processing are performed. And cleaning and processing the information acquired in the step S201, wherein the user basic information is characteristic data, and the user service performance information is target data.
More specifically, the cleaning process for the feature data mainly comprises missing value padding and enumerated value encoding, wherein the enumerated value encoding is preferably realized by using an imbedding method.
The user service performance information comprises performance information of different service types of the user and/or performance information of the same service without using time. When the target value is processed, the same service performance information that changes with time can be vectorized in time series, for example, y1, y2, y3 and y4 respectively represent the first-stage service performance information, the second-stage service performance information, the third-stage service performance information and the fourth-stage service performance information of the a service user, and can be represented by a vector [ y1, y2, y3 and y4 ]. A plurality of time series based vectors may be used to tailor a plurality of business performance information of a user.
In step S203, a feature evaluation framework model is constructed based on the historical user information, where the framework model includes a plurality of sub-models that are connected to each other and are capable of independently outputting corresponding service feature evaluation results.
Specifically, a neural network frame built by tensiorflow is utilized, a characteristic evaluation frame model adopts a random gradient descent optimization method with momentum, a plurality of submodels capable of independently outputting corresponding service characteristic evaluation results are connected in series, and the output result of the former submodel is the input information of all the following submodels; the input information of any submodel comprises the user basic information and the output result information of all submodels before the current submodel.
FIG. 3 is a block diagram illustrating a flow structure of a feature evaluation framework model according to an exemplary embodiment.
As shown in fig. 3, the feature evaluation framework model is composed of a plurality of service submodels, and the service submodels are used for outputting feature scores of corresponding services. The feature evaluation framework may output the feature score values of all the sub-models, including both the sequence of the feature score values of the same type of service and the feature score values of a plurality of different services. The feature scores include, but are not limited to, risk features, activity features, revenue features.
Each service sub-model comprises a plurality of times of Full Connection (FC) and batch normalization operation (BN) on the acquired data, and a Dropout algorithm (hereinafter referred to as an operation layer) is added during each operation. The output value obtained by each service submodel is used as the input value of all the following models, and the output value of all the submodels before the service submodel is used as the input value of the model.
More specifically, each business sub-model is trained by adopting a neural network method, and comprises a plurality of operation layers, each operation layer performs full connection operation and batch normalization operation, and meanwhile, a Dropout method is adopted to simplify the model and improve the generalization force of the model. The intermediate vector formed after the operation of each operation layer is transmitted to all the operation layers behind the operation layer, wherein the intermediate vector comprises the post operation layer of the same sub-model and the operation layers of other sub-models behind the current sub-model. Therefore, a merge layer is provided before each operation layer to merge the acquired intermediate vectors formed by the previous operation layer as an input of the current operation layer.
As shown in fig. 3, for a single dimension, a full join layer connection extension is used, then after merging these intermediate vectors, full join layer and batch normalization operations are performed, and dropout is added in the middle. The front middle vector is used as an input of the rear vector, and a plurality of output nodes are arranged from top to bottom, namely the rear node can learn the information of the front target value.
In step S204, the current user basic information is acquired.
More specifically, the current user basic information includes, but is not limited to, user-related information necessary for performing a service, such as user identity information, behavior information, financial information, etc. required for performing user feature management.
Preferably, the acquired current user basic information is cleaned and processed, the cleaning processing mainly comprises missing value padding and enumerated value coding, and the enumerated value coding is preferably realized by an embedding method.
In step S205, feature evaluation is performed
And inputting the basic information of the current user acquired in the step S204 into a feature evaluation framework model to acquire the feature scores of the current user under different services. In the financial service, for example, the expected overdue probability, the dynamic branch probability, the profit margin of the user under different quota, etc. are obtained.
In step S206, a policy is made.
Based on the feature scores of the users under different services acquired in step S205, a corresponding policy is formulated. More specifically, different strategies may be preset, and the optimal strategy may be selected by a worker's judgment according to the feature score or determined by using a machine learning model based on the feature score and the strategy.
More specifically, when the determination is performed using the machine learning model, the policy model may be obtained using basic information of a historical user, feature information of different services, and performance information under policy as training data. The most suitable strategy scheme is predicted by inputting the basic information of the user and the characteristic information of different services.
It is particularly emphasized that when using machine learning models, the training method thereof is performed by methods commonly used in the art, and is not dependent on a specific method.
Those skilled in the art will appreciate that all or part of the steps to implement the above-described embodiments are implemented as programs (computer programs) executed by a computer data processing apparatus. When the computer program is executed, the method provided by the invention can be realized. Furthermore, the computer program may be stored in a computer readable storage medium, which may be a readable storage medium such as a magnetic disk, an optical disk, a ROM, a RAM, or a storage array composed of a plurality of storage media, such as a magnetic disk or a magnetic tape storage array. The storage medium is not limited to centralized storage, but may be distributed storage, such as cloud storage based on cloud computing.
Embodiments of the apparatus of the present invention are described below, which may be used to perform method embodiments of the present invention. The details described in the device embodiments of the invention should be regarded as complementary to the above-described method embodiments; reference is made to the above-described method embodiments for details not disclosed in the apparatus embodiments of the invention.
Fig. 4 is a block diagram illustrating a neural network-based multi-output multi-tasking feature evaluation apparatus according to an exemplary embodiment. As shown in FIG. 4, the multi-output multi-tasking feature evaluation device 40 includes a historical information module 401, a model module 402, a current information module 403, a scoring module 404, a policy module 405, and a cleansing module 406.
The historical information module 401 is configured to obtain historical user information, where the historical user information includes historical user basic information and service performance information.
More specifically, the historical user basic information includes, but is not limited to, user-related information necessary for performing a business, such as user identity information, behavior information, financial information, etc. required in performing user feature management.
The business performance data includes but is not limited to information embodied by the user when the user actually completes the business, such as overdue condition information, movable and supported condition information, repayment condition information, financing condition information, profit condition under the credit line and the like of the user in financial activities.
A model module 402, configured to construct a feature evaluation framework model based on the historical user information, where the framework model includes a plurality of submodels that are connected to each other and are capable of independently outputting corresponding service feature evaluation results.
Specifically, the model module 402 uses a neural network framework built by tenserflow, and the feature evaluation framework model adopts a random gradient descent optimization method with momentum, and connects a plurality of submodels capable of independently outputting corresponding service feature evaluation results in series, and the output result of the former submodel is the input information of all the following submodels; the input information of any submodel comprises the user basic information and the output result information of all submodels before the current submodel.
The feature evaluation framework model built by the model module 402 is composed of a plurality of service submodels, and the service submodels are used for outputting feature scores of corresponding services. The feature evaluation framework may output the feature score values of all the sub-models, including both the sequence of the feature score values of the same type of service and the feature score values of a plurality of different services. The feature scores include, but are not limited to, risk features, activity features, revenue features.
Each service sub-model comprises a plurality of times of Full Connection (FC) and batch normalization operation (BN) on the acquired data, and a Dropout algorithm (hereinafter referred to as an operation layer) is added during each operation. The output value obtained by each service submodel is used as the input value of all the following models, and the output value of all the submodels before the service submodel is used as the input value of the model.
More specifically, each business sub-model is trained by adopting a neural network method, and comprises a plurality of operation layers, each operation layer performs full connection operation and batch normalization operation, and meanwhile, a Dropout method is adopted to simplify the model and improve the generalization force of the model. The intermediate vector formed after the operation of each operation layer is transmitted to all the operation layers behind the operation layer, wherein the intermediate vector comprises the post operation layer of the same sub-model and the operation layers of other sub-models behind the current sub-model. Therefore, a merge layer is provided before each operation layer to merge the acquired intermediate vectors formed by the previous operation layer as an input of the current operation layer.
A current information module 403, configured to obtain current user basic information.
More specifically, the current user basic information includes, but is not limited to, user-related information necessary for performing a service, such as user identity information, behavior information, financial information, etc. required for performing user feature management.
A scoring module 404, configured to input the basic information of the current user into the feature evaluation framework model to obtain feature scores of multiple different types of services of the current user.
The basic information of the current user obtained from the current information module 403 is input to the feature evaluation framework model to obtain the feature scores of the current user under different services. In the financial service, for example, the expected overdue probability, the dynamic branch probability, the profit margin of the user under different quota, etc. are obtained.
And a policy module 405, configured to formulate a policy for the corresponding service according to an output result of the feature evaluation framework model.
And a cleaning module 406, configured to clean the historical user information and the current user information after acquiring the historical user information and the current user information.
The cleaning processing of the feature data by the cleaning module 406 mainly includes missing value padding and enumerated value encoding, wherein the enumerated value encoding is preferably implemented by using an imbedding method.
The user service performance information comprises performance information of different service types of the user and/or performance information of the same service without using time. The cleansing module 406 may time-sequentially vectorize the same service performance information that changes with time when processing the target value, for example, y1, y2, y3, and y4 may represent the first-stage service performance information, the second-stage service performance information, the third-stage service performance information, and the fourth-stage service performance information of the a service user using vectors [ y1, y2, y3, y4 ]. A plurality of time series based vectors may be used to tailor a plurality of business performance information of a user.
Those skilled in the art will appreciate that the modules in the above-described embodiments of the apparatus may be distributed as described in the apparatus, and may be correspondingly modified and distributed in one or more apparatuses other than the above-described embodiments. The modules of the above embodiments may be combined into one module, or further split into multiple sub-modules.
In the following, embodiments of the electronic device of the present invention are described, which may be regarded as specific physical implementations for the above-described embodiments of the method and apparatus of the present invention. Details described in the embodiments of the electronic device of the invention should be considered supplementary to the embodiments of the method or apparatus described above; for details which are not disclosed in embodiments of the electronic device of the invention, reference may be made to the above-described embodiments of the method or the apparatus.
FIG. 5 is a block diagram illustrating an electronic device in accordance with an example embodiment.
An electronic device 500 according to this embodiment of the disclosure is described below with reference to fig. 5. The electronic device 500 shown in fig. 5 is only an example and should not bring any limitation to the functions and the scope of use of the embodiments of the present disclosure.
As shown in fig. 5, the electronic device 500 is embodied in the form of a general purpose computing device. The components of the electronic device 500 may include, but are not limited to: at least one processing unit 510, at least one memory unit 520, a bus 530 that couples various system components including the memory unit 520 and the processing unit 510, a display unit 540, and the like.
Wherein the storage unit stores program code executable by the processing unit 510 to cause the processing unit 510 to perform the steps according to various exemplary embodiments of the present disclosure described in the above-mentioned electronic prescription flow processing method section of the present specification. For example, the processing unit 510 may perform the steps as shown in fig. 2, fig. 3.
The memory unit 520 may include a readable medium in the form of a volatile memory unit, such as a random access memory unit (RAM) 5201 and/or a cache memory unit 5202, and may further include a read only memory unit (ROM) 5203.
The memory unit 520 may also include a program/utility 5204 having a set (at least one) of program modules 5205, such program modules 5205 including, but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which, or some combination thereof, may comprise an implementation of a network environment.
The electronic device 500 may also communicate with one or more external devices 500' (e.g., keyboard, pointing device, bluetooth device, etc.), with one or more devices that enable a user to interact with the electronic device 500, and/or with any devices (e.g., router, modem, etc.) that enable the electronic device 500 to communicate with one or more other computing devices. Such communication may occur via input/output (I/O) interfaces 550. Also, the electronic device 500 may communicate with one or more networks (e.g., a Local Area Network (LAN), a Wide Area Network (WAN), and/or a public network, such as the internet) via the network adapter 560. The network adapter 560 may communicate with other modules of the electronic device 500 via the bus 530. It should be appreciated that although not shown in the figures, other hardware and/or software modules may be used in conjunction with the electronic device 500, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, among others.
Through the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein may be implemented by software, or by software in combination with necessary hardware. Therefore, as shown in fig. 6, the technical solution according to the embodiment of the present disclosure may be embodied in the form of a software product, which may be stored in a non-volatile storage medium (which may be a CD-ROM, a usb disk, a removable hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which may be a personal computer, a server, or a network device, etc.) to execute the above method according to the embodiment of the present disclosure.
The software product may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium include: an electrical connection having one or more wires, a portable disk, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
The computer readable storage medium may include a propagated data signal with readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A readable storage medium may also be any readable medium that is not a readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a readable storage medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.
Program code for carrying out operations for the present disclosure may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device and partly on a remote computing device, or entirely on the remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computing device (e.g., through the internet using an internet service provider).
The computer readable medium carries one or more programs which, when executed by a device, cause the computer readable medium to perform the functions of: acquiring a first table in a first database and a second table in a second database; comparing the data structures of the first table and the second table through a database statement to generate a comparison result; generating a first automatic processing instruction and a second automatic processing instruction according to the comparison result; and determining to execute the first automatic processing instruction or the second automatic processing instruction according to a preset strategy.
Those skilled in the art will appreciate that the modules described above may be distributed in the apparatus according to the description of the embodiments, or may be modified accordingly in one or more apparatuses unique from the embodiments. The modules of the above embodiments may be combined into one module, or further split into multiple sub-modules.
Through the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein may be implemented by software, or by software in combination with necessary hardware. Therefore, the technical solution according to the embodiments of the present disclosure may be embodied in the form of a software product, which may be stored in a non-volatile storage medium (which may be a CD-ROM, a usb disk, a removable hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which may be a personal computer, a server, a mobile terminal, or a network device, etc.) to execute the method according to the embodiments of the present disclosure.
Exemplary embodiments of the present disclosure are specifically illustrated and described above. It is to be understood that the present disclosure is not limited to the precise arrangements, instrumentalities, or instrumentalities described herein; on the contrary, the disclosure is intended to cover various modifications and equivalent arrangements included within the spirit and scope of the appended claims.
Claims (9)
1. A multi-output and multi-task feature evaluation method based on a neural network is characterized by comprising the following steps:
acquiring historical user information, wherein the historical user information comprises historical user basic information and service performance information;
constructing a characteristic evaluation frame model based on the historical user information, wherein the frame model comprises a plurality of submodels which are connected with each other and can respectively and independently output corresponding service characteristic evaluation results, the submodels are connected in series, and the output result of the previous submodel is the input information of all the following submodels;
acquiring current user basic information;
and inputting the basic information of the current user into the feature evaluation framework model to obtain the feature scores of a plurality of different types of services of the current user.
2. The method of claim 1, wherein: the service performance information comprises different types of service performance information and/or a plurality of service performance information of the same service in a time sequence.
3. The method of claim 2, wherein: the multiple service performance information of the same service with time as a sequence is a vector with time as a sequence formed by selecting performance values of the same service at different times.
4. The method of claim 1, wherein the feature evaluation framework model further comprises: the input information of any submodel comprises the user basic information and the output result information of all submodels before the current submodel.
5. The method of claim 4, further comprising: the characteristic evaluation frame model is a neural network frame built by utilizing tensorflow.
6. The method of claim 5, further comprising: the characteristic evaluation frame model adopts a random gradient descent optimization method with momentum.
7. A neural network-based multi-output multi-tasking feature evaluation apparatus, comprising:
the historical information module is used for acquiring historical user information, and the historical user information comprises historical user basic information and business performance information;
the model module is used for constructing a characteristic evaluation framework model based on the historical user information, the framework model comprises a plurality of submodels which are connected with each other and can respectively and independently output corresponding service characteristic evaluation results, the submodels are connected in series, and the output result of the previous submodel is the input information of all the following submodels;
the current information module is used for acquiring current user basic information;
and the scoring module is used for inputting the basic information of the current user into the feature evaluation framework model to obtain the feature scores of a plurality of different types of services of the current user.
8. An electronic device, wherein the electronic device comprises:
a processor; and the number of the first and second groups,
a memory storing computer-executable instructions that, when executed, cause the processor to perform the method of any of claims 1-6.
9. A computer readable storage medium, wherein the computer readable storage medium stores one or more programs which, when executed by a processor, implement the method of any of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110167037.3A CN112949824A (en) | 2021-02-07 | 2021-02-07 | Neural network-based multi-output multi-task feature evaluation method and device and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110167037.3A CN112949824A (en) | 2021-02-07 | 2021-02-07 | Neural network-based multi-output multi-task feature evaluation method and device and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112949824A true CN112949824A (en) | 2021-06-11 |
Family
ID=76243045
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110167037.3A Pending CN112949824A (en) | 2021-02-07 | 2021-02-07 | Neural network-based multi-output multi-task feature evaluation method and device and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112949824A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113487078A (en) * | 2021-06-30 | 2021-10-08 | 上海淇馥信息技术有限公司 | Method and device for executing new generation task and electronic equipment |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108846520A (en) * | 2018-06-22 | 2018-11-20 | 北京京东金融科技控股有限公司 | Overdue loan prediction technique, device and computer readable storage medium |
| US20180336796A1 (en) * | 2017-05-18 | 2018-11-22 | Pearson Education, Inc. | Multi-level executive functioning tasks |
| US20190219994A1 (en) * | 2018-01-18 | 2019-07-18 | General Electric Company | Feature extractions to model large-scale complex control systems |
| WO2019167884A1 (en) * | 2018-02-28 | 2019-09-06 | 富士フイルム株式会社 | Machine learning method and device, program, learned model, and identification device |
| WO2020005049A1 (en) * | 2018-06-25 | 2020-01-02 | 주식회사 수아랩 | Learning method for artificial neural network |
| CN111123232A (en) * | 2019-11-11 | 2020-05-08 | 浙江大学 | Radar individual identification system with task adaptability |
| CN111553759A (en) * | 2020-03-25 | 2020-08-18 | 平安科技（深圳）有限公司 | Product information pushing method, device, equipment and storage medium |
| CN111553466A (en) * | 2020-04-30 | 2020-08-18 | 上海商汤智能科技有限公司 | Information processing method, device and equipment |
| CN111582694A (en) * | 2020-04-29 | 2020-08-25 | 腾讯科技（深圳）有限公司 | Learning evaluation method and device |
| CN112258622A (en) * | 2020-10-26 | 2021-01-22 | 北京字跳网络技术有限公司 | Image processing method, image processing device, readable medium and electronic equipment |
-
2021
- 2021-02-07 CN CN202110167037.3A patent/CN112949824A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180336796A1 (en) * | 2017-05-18 | 2018-11-22 | Pearson Education, Inc. | Multi-level executive functioning tasks |
| US20190219994A1 (en) * | 2018-01-18 | 2019-07-18 | General Electric Company | Feature extractions to model large-scale complex control systems |
| WO2019167884A1 (en) * | 2018-02-28 | 2019-09-06 | 富士フイルム株式会社 | Machine learning method and device, program, learned model, and identification device |
| CN108846520A (en) * | 2018-06-22 | 2018-11-20 | 北京京东金融科技控股有限公司 | Overdue loan prediction technique, device and computer readable storage medium |
| WO2020005049A1 (en) * | 2018-06-25 | 2020-01-02 | 주식회사 수아랩 | Learning method for artificial neural network |
| CN111123232A (en) * | 2019-11-11 | 2020-05-08 | 浙江大学 | Radar individual identification system with task adaptability |
| CN111553759A (en) * | 2020-03-25 | 2020-08-18 | 平安科技（深圳）有限公司 | Product information pushing method, device, equipment and storage medium |
| CN111582694A (en) * | 2020-04-29 | 2020-08-25 | 腾讯科技（深圳）有限公司 | Learning evaluation method and device |
| CN111553466A (en) * | 2020-04-30 | 2020-08-18 | 上海商汤智能科技有限公司 | Information processing method, device and equipment |
| CN112258622A (en) * | 2020-10-26 | 2021-01-22 | 北京字跳网络技术有限公司 | Image processing method, image processing device, readable medium and electronic equipment |
Non-Patent Citations (7)
| Title |
|---|
| EDWIN LUGHOFER: "Robust Generalized Fuzzy Systems Training From High-Dimensional Time-Series Data Using Local Structure Preserving PLS", 《 IEEE TRANSACTIONS ON FUZZY SYSTEMS 》, vol. 28, no. 11, pages 2930 - 2943, XP011818759, DOI: 10.1109/TFUZZ.2019.2945535 * |
| JIE SUN 等: "Cross-Model Deep Feature Fusion for Face Detection", 《 IEEE SENSORS LETTERS》, vol. 4, no. 9, 30 September 2020 (2020-09-30), pages 1 - 4, XP011808222, DOI: 10.1109/LSENS.2020.3018594 * |
| MENGPING YU 等: "Recursive identification for Hammerstein systems with hybrid static nonlinearities and asymmetric ARX", 《IEEE JOURNAL ON MULTISCALE AND MULTIPHYSICS COMPUTATIONAL TECHNIQUES》, vol. 3, no. 1, pages 66 - 73 * |
| 刘宇瀚 等: "结合字形特征与迭代学习的金融领域命名实体识别", 《中文信息学报》, vol. 34, no. 11, pages 74 - 83 * |
| 张健: "基于智能集成粒子群算法的时间序列数据挖掘研究", 《四川理工学院学报(自然科学版)》, vol. 28, no. 04, pages 27 - 32 * |
| 杨云 等: "层聚合的对抗机器翻译模型的方法研究", 《陕西科技大学学报》, vol. 38, no. 01, 14 January 2020 (2020-01-14), pages 164 - 170 * |
| 陆鋆: "面向图像检索的深度哈希方法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》, no. 2020, pages 138 - 413 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113487078A (en) * | 2021-06-30 | 2021-10-08 | 上海淇馥信息技术有限公司 | Method and device for executing new generation task and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112348660B (en) | Method and device for generating risk warning information and electronic equipment | |
| CN110349009B (en) | Multi-head lending default prediction method and device and electronic equipment | |
| CN110111198A (en) | User's financial risks predictor method, device, electronic equipment and readable medium | |
| CN110111139B (en) | Behavior prediction model generation method and device, electronic equipment and readable medium | |
| CN107729937A (en) | For determining the method and device of user interest label | |
| CN111210335B (en) | User risk identification method and device and electronic equipment | |
| CN109492104B (en) | Training method, classification method, system, device and medium of intention classification model | |
| CN109684627A (en) | A kind of file classification method and device | |
| JP2019079302A (en) | Sales activity support system, sales activity support method and sales activity support program | |
| CN111145009A (en) | Method and device for evaluating risk after user loan and electronic equipment | |
| CN112988963B (en) | User intention prediction method, device, equipment and medium based on multi-flow nodes | |
| JP2020087023A (en) | Order acceptance prediction model generating method, order acceptance prediction model, order acceptance predicting apparatus, order acceptance predicting method, and order acceptance predicting program | |
| CN112348321A (en) | Risk user identification method and device and electronic equipment | |
| CN111611514B (en) | Page display method and device based on user login information and electronic equipment | |
| CN112015562A (en) | Resource allocation method and device based on transfer learning and electronic equipment | |
| CN117608650B (en) | Business flow chart generation method, processing device and storage medium | |
| US20230032739A1 (en) | Propensity modeling process for customer targeting | |
| CN112017062B (en) | Resource quota distribution method and device based on guest group subdivision and electronic equipment | |
| CN113051480A (en) | Resource pushing method and device, electronic equipment and storage medium | |
| CN111190967B (en) | User multidimensional data processing method and device and electronic equipment | |
| CN111210332A (en) | Method and device for generating post-loan management strategy and electronic equipment | |
| CN110348892A (en) | Customized information generation method, device and electronic equipment based on user characteristics | |
| CN114663155A (en) | Advertisement putting and selecting method and device, equipment, medium and product thereof | |
| CN111178701B (en) | Risk control method and device based on feature derivation technology and electronic equipment | |
| CN112949824A (en) | Neural network-based multi-output multi-task feature evaluation method and device and electronic equipment |
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 |