CN115293364A - Method and device for measuring stability of machine learning characteristic factor - Google Patents
Method and device for measuring stability of machine learning characteristic factor Download PDFInfo
- Publication number
- CN115293364A CN115293364A CN202210987135.6A CN202210987135A CN115293364A CN 115293364 A CN115293364 A CN 115293364A CN 202210987135 A CN202210987135 A CN 202210987135A CN 115293364 A CN115293364 A CN 115293364A
- Authority
- CN
- China
- Prior art keywords
- machine learning
- characteristic factor
- characteristic
- model
- determining
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0202—Market predictions or forecasting for commercial activities
Abstract
The embodiment of the application provides a method and a device for measuring stability of a machine learning characteristic factor, which relate to the field of machine learning, and the method comprises the following steps: taking historical market data with a set time period as model training data; performing model training on a preset machine learning prediction model according to the model training data and a set rolling mode to obtain a characteristic factor combination sequence; determining an optimal characteristic factor combination according to the occurrence frequency and the occurrence time of each characteristic factor in the characteristic factor combination sequence; the prediction effect of the machine learning model can be effectively improved.
Description
Technical Field
The application relates to the field of machine learning and can also be used in the field of finance, in particular to a method and a device for measuring stability of a machine learning characteristic factor.
Background
In the field of financial markets, predicting market fluctuations using machine learning algorithms is a popular topic. The selection of a long-term effective feature factor combination is the central importance of the machine learning algorithm, and the prediction effect of the model is directly determined.
In a statistical sense, stock prices are generally considered to be subject to a markov process, which is colloquially understood to mean that historical stock data does not contain future stock trend information, so that earlier market data is not necessarily beneficial to model training, and may even interfere with the model to the opposite effect. But for the machine learning algorithm, historical data has to be used as a training sample of the model. The trade-off is therefore to select historical market data that is relatively new for as long a period of time as possible. In addition, market conditions are varied, a group of characteristic factor combinations which are suitable for the model and excellent in performance are difficult to select, and a group of good characteristic factor combinations can enable the model to be stable in long-term performance, so that the model has high reliability, and more excess profits are obtained while withdrawal risks are controllable.
Disclosure of Invention
Aiming at the problems in the prior art, the application provides a method and a device for measuring the stability of the machine learning characteristic factors, which can effectively improve the prediction effect of a machine learning model.
In order to solve at least one of the above problems, the present application provides the following technical solutions:
in a first aspect, the present application provides a method for measuring stability of a machine learning feature factor, including:
taking historical market data with a set time period as model training data;
performing model training on a preset machine learning prediction model according to the model training data and a set rolling mode to obtain a characteristic factor combination sequence;
and determining the optimal characteristic factor combination according to the occurrence frequency and the occurrence time of each characteristic factor in the characteristic factor combination sequence.
Further, the performing model training on a preset machine learning prediction model according to the model training data and a set rolling mode to obtain a characteristic factor combination sequence includes:
determining market fluctuation labels of the training data of the models according to the numerical comparison result between the training data of different models;
and performing model training on a preset machine learning prediction model according to the combination of different model training data to determine a characteristic factor combination sequence.
Further, the determining an optimal characteristic factor combination according to the occurrence frequency and the occurrence time of each characteristic factor in the characteristic factor combination sequence includes:
constructing a characteristic subset according to different characteristic factor combinations in the characteristic factor combination sequence;
and determining the occurrence frequency of each characteristic factor combination in all the characteristic subsets, and determining the optimal characteristic factor combination according to the occurrence frequency and the corresponding occurrence time.
Further, after determining an optimal feature factor combination according to the occurrence frequency and the occurrence time of each feature factor in the feature factor combination sequence, the method includes:
combining the optimal characteristic factors to serve as model parameters of a preset machine learning prediction model;
and performing model training again on the preset machine learning prediction model to obtain the machine learning prediction model after model training.
In a second aspect, the present application provides a device for measuring stability of machine learning feature factors, including:
the data accuracy module is used for taking historical market data with a set time period as model training data;
the rolling training module is used for carrying out model training on a preset machine learning prediction model according to the model training data and a set rolling mode to obtain a characteristic factor combination sequence;
and the optimal calculation module is used for determining an optimal characteristic factor combination according to the occurrence frequency and the occurrence time of each characteristic factor in the characteristic factor combination sequence.
Further, the rolling training module comprises:
the label determining unit is used for determining market fluctuation labels of the training data of the models according to the numerical comparison result between the training data of different models;
and the sequence determining unit is used for carrying out model training on the preset machine learning prediction model according to the combination of different model training data and determining a characteristic factor combination sequence.
Further, the optimal calculation module comprises:
the subset construction unit is used for constructing a feature subset according to different feature factor combinations in the feature factor combination sequence;
and the frequency calculation unit is used for determining the occurrence frequency of each characteristic factor combination in all the characteristic subsets and determining the optimal characteristic factor combination according to the occurrence frequency and the corresponding occurrence time.
Further, the method also comprises the following steps:
the parameter determining unit is used for combining the optimal characteristic factors to serve as model parameters of a preset machine learning prediction model;
and the model retraining unit is used for performing model retraining on the preset machine learning prediction model to obtain the machine learning prediction model after model training.
In a third aspect, the present application provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor implements the steps of the method for machine learning feature factor stability metric when executing the program.
In a fourth aspect, the present application provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of the method for machine learning feature factor stability metric.
In a fifth aspect, the present application provides a computer program product comprising computer program/instructions which, when executed by a processor, implement the steps of the method for machine learning feature factor stability metric.
According to the technical scheme, the method and the device for measuring the stability of the machine learning characteristic factors are characterized in that historical market data with a set time period are used as model training data; performing model training on a preset machine learning prediction model according to the model training data and a set rolling mode to obtain a characteristic factor combination sequence; and determining the optimal characteristic factor combination according to the occurrence frequency and the occurrence time of each characteristic factor in the characteristic factor combination sequence, thereby effectively improving the prediction effect of the machine learning model.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
Fig. 1 is a flowchart illustrating a method for measuring stability of a machine learning feature factor according to an embodiment of the present disclosure;
FIG. 2 is a second flowchart illustrating a method for measuring stability of machine learning feature factors according to an embodiment of the present application;
fig. 3 is a third flowchart illustrating a method for measuring stability of a machine learning feature factor according to an embodiment of the present application;
FIG. 4 is a fourth flowchart illustrating a method for measuring stability of machine learning feature factors according to an embodiment of the present disclosure;
FIG. 5 is a block diagram of a machine learning feature factor stability measurement apparatus according to an embodiment of the present application;
FIG. 6 is a second block diagram of a device for measuring stability of machine-learned feature factors in an embodiment of the present application;
FIG. 7 is a third block diagram of a machine learning feature factor stability measurement apparatus according to an embodiment of the present application;
FIG. 8 is a fourth block diagram of a device for measuring stability of machine learning feature factors in an embodiment of the present application;
fig. 9 is a schematic structural diagram of an electronic device in an embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments in the present application without making any creative effort belong to the protection scope of the present application.
According to the technical scheme, the data acquisition, storage, use, processing and the like meet the relevant regulations of national laws and regulations.
In consideration of the problems in the prior art, the application provides a method and a device for measuring the stability of a machine learning characteristic factor, wherein historical market data with a set time period are used as model training data; performing model training on a preset machine learning prediction model according to the model training data and a set rolling mode to obtain a characteristic factor combination sequence; and determining the optimal characteristic factor combination according to the occurrence frequency and the occurrence time of each characteristic factor in the characteristic factor combination sequence, thereby effectively improving the prediction effect of the machine learning model.
In order to effectively improve the prediction effect of the machine learning model, the present application provides an embodiment of a method for measuring stability of a machine learning feature factor, and referring to fig. 1, the method for measuring stability of a machine learning feature factor specifically includes the following contents:
step S101: and taking historical market data of a set time period as model training data.
Optionally, for the predictive market trend machine learning model, the present application may prepare data factors that may be related to the trend, including but not limited to, target exchange rate daily market, inflation rate, consumer Price Index (CPI), producer Price Index (PPI), with a time period that is selected as long as possible over the time period, such as preparing data for the last decade.
Step S102: and performing model training on a preset machine learning prediction model according to the model training data and a set rolling mode to obtain a characteristic factor combination sequence.
Optionally, the model training may be performed in a rolling mode.
For example, assume that the training data size used is 3 samples and the prediction period is a data length corresponding to 2 samples. Then, in order to prepare the training data, if the market fluctuation label is to be marked for the first sample, the data of the first sample needs to be compared with the data of the third sample, so that the data of the third sample needs to be used. In the 3 training sample data, different training models can be obtained by taking different characteristic factor combinations, and the optimal characteristic factor combination in the current sample data can be determined according to the effect on the training set. After the step is finished, the step can be rolled, new training samples are formed by the second sample, the third sample and the fourth sample, and the characteristic factor combination of the three samples can be obtained by adopting the same method. By analogy, a characteristic factor combination sequence can be finally obtained: Ω = { ω 1 ,ω 2 ,Λ,ω n-3 }。
Step S103: and determining the optimal characteristic factor combination according to the occurrence frequency and the occurrence time of each characteristic factor in the characteristic factor combination sequence.
Optionally, in the factor combination sequence obtained in the previous step, a final factor combination may be selected through a specified algorithm to predict the fx market fluctuation at the current time point. The algorithm may be the frequency of occurrence of each feature, i.e. the statistical rule is: count (feature) i ) I ∈ Ω, then n next to the former according to the selection frequency, where n = Mean ([ Count (ω) 1 ),Count(ω 2 ),Λ,Count(ω n-3 )]) The n features are used as stable feature factor combinations used in the prediction.
In another embodiment, the selection of the feature factors needs to consider the time sequence and the clustering property because some feature factors have long-term trends and some features may fluctuate periodically with seasons and the like. A scheme for identifying combinations of characteristic factors is proposed.
Combining omega = { omega ] based on said factor sequence 1 ,ω 2 ,Λ,ω n-3 And (6) establishing a time series model.
Specifically, m (m) is sequentially used<n) feature combinations are used for constructing a subset, then m-n +1 subsets can be obtained according to the existing factor sequence combinations, for each subset, the frequency of occurrence of a certain feature factor in each subset can be counted, and the frequency of the feature factor f in the ith subset is assumed to be p i Then we can get the data sequence combination { (i, p) i ) f I =1,2 Λ, m-n +1, f ∈ pool of characteristic factors }, and for each characteristic f, the sequence data combination is fitted according to the principle of least squares, and if a linear model is selected, the formula p = ki + const can be obtained, where k is the slope of the fitted line and const is a constant. Then, in all the characteristic factor pools, the application selects a plurality of factors with larger k values. The richness of the sample data can effectively improve the stability of the selection scheme.
As can be seen from the above description, the method for measuring the stability of the machine learning feature factor provided by the embodiment of the present application can use historical market data of a set time period as model training data; performing model training on a preset machine learning prediction model according to the model training data and a set rolling mode to obtain a characteristic factor combination sequence; and determining the optimal characteristic factor combination according to the occurrence frequency and the occurrence time of each characteristic factor in the characteristic factor combination sequence, thereby effectively improving the prediction effect of the machine learning model.
In an embodiment of the method for measuring stability of a machine learning feature factor according to the present application, referring to fig. 2, the following may be further included:
step S201: and determining the market fluctuation labels of the training data of the models according to the numerical comparison result between the training data of different models.
Step S202: and performing model training on a preset machine learning prediction model according to the combination of different model training data, and determining a characteristic factor combination sequence.
For example, assume that the training data size is 3 samples and the prediction period is a data length corresponding to 2 samples. Then, in order to prepare the training data, if the market fluctuation label is to be marked for the first sample, the data of the first sample and the data of the third sample need to be obtained by comparison, so that the data of the third sample is needed. In the 3 training sample data, different training models can be obtained by taking different characteristic factor combinations, and the optimal characteristic factor combination in the current sample data can be determined according to the effect on the training set. After the step is finished, rolling is carried out for one step, new training samples are formed by the second sample, the third sample and the fourth sample, and the characteristic factor combination of the three samples can be obtained by adopting the same method. By analogy, a characteristic factor combination sequence can be finally obtained.
In an embodiment of the method for measuring stability of a machine learning feature factor according to the present application, referring to fig. 3, the following may be further included:
step S301: and constructing a feature subset according to different feature factor combinations in the feature factor combination sequence.
Step S302: and determining the occurrence frequency of each characteristic factor combination in all the characteristic subsets, and determining the optimal characteristic factor combination according to the occurrence frequency and the corresponding occurrence time.
Optionally, in the factor combination sequence obtained in the previous step, a final factor combination may be selected through a specified algorithm to predict the fx market fluctuation at the current time point. The algorithm may be the frequency of occurrence of the respective feature.
In another embodiment, the selection of the feature factors needs to consider the time sequence and the clustering property because some feature factors have long-term tendency, and some features may have periodic fluctuation with seasons and the like. A scheme for identifying combinations of characteristic factors is proposed.
And establishing a time series model based on the factor series combination.
Specifically, m (m < n) feature combinations are sequentially used for constructing a subset, then m-n +1 subsets can be obtained according to the existing factor sequence combinations, for each subset, the frequency of occurrence of a certain feature factor in each subset can be counted, and if the frequency times of the feature factor f in the ith subset are assumed, then a data sequence combination can be obtained. Then, in all the characteristic factor pools, the application selects a plurality of factors with larger k values. The richness of the sample data can effectively improve the stability of the selection scheme.
In an embodiment of the method for measuring stability of the machine learning feature factor according to the present application, referring to fig. 4, the following may be further included:
step S401: and combining the optimal characteristic factors to be used as model parameters of a preset machine learning prediction model.
Step S402: and performing model training again on the preset machine learning prediction model to obtain the machine learning prediction model after model training.
Optionally, the feature factor combination, the training sample size, the training model and the prediction period determined in the previous step can be selected and described in the first step, after model training is performed, the latest data is taken for prediction, and the model is ended.
In order to effectively improve the prediction effect of the machine learning model, the present application provides an embodiment of a machine learning feature factor stability measuring apparatus for implementing all or part of the contents of the machine learning feature factor stability measuring method, and referring to fig. 5, the machine learning feature factor stability measuring apparatus specifically includes the following contents:
and the data accuracy module 10 is used for taking the historical market data of a set time period as model training data.
And the rolling training module 20 is configured to perform model training on a preset machine learning prediction model according to the model training data and a set rolling mode, so as to obtain a feature factor combination sequence.
And an optimal calculation module 30, configured to determine an optimal characteristic factor combination according to the occurrence frequency and the occurrence time of each characteristic factor in the characteristic factor combination sequence.
As can be seen from the above description, the device for measuring the stability of the machine learning feature factor according to the embodiment of the present application can use historical market data of a set time period as model training data; performing model training on a preset machine learning prediction model according to the model training data and a set rolling mode to obtain a characteristic factor combination sequence; and determining the optimal characteristic factor combination according to the occurrence frequency and the occurrence time of each characteristic factor in the characteristic factor combination sequence, thereby effectively improving the prediction effect of the machine learning model.
In an embodiment of the device for measuring stability of machine-learned feature factors of the present application, referring to fig. 6, the rolling training module 20 includes:
and the label determining unit 21 is configured to determine a market fluctuation label of each model training data according to a value comparison result between different model training data.
And the sequence determining unit 22 is configured to perform model training on the preset machine learning prediction model according to a combination of different model training data, and determine a characteristic factor combination sequence.
In an embodiment of the device for measuring stability of machine learning feature factor of the present application, referring to fig. 7, the optimal calculation module 30 includes:
a subset construction unit 31, configured to construct a feature subset according to different feature factor combinations in the feature factor combination sequence.
And the frequency calculation unit 32 is configured to determine occurrence frequencies of the feature factor combinations in all feature subsets, and determine an optimal feature factor combination according to the occurrence frequencies and corresponding occurrence times.
In an embodiment of the device for measuring stability of machine-learned feature factors of the present application, referring to fig. 8, the following contents are also specifically included:
and a parameter determining unit 41, configured to use the optimal feature factor combination as a model parameter of a preset machine learning prediction model.
And the model retraining unit 42 is configured to perform model retraining on the preset machine learning prediction model again to obtain the machine learning prediction model after model training.
In terms of hardware, in order to effectively improve the prediction effect of the machine learning model, the present application provides an embodiment of an electronic device for implementing all or part of the contents in the method for measuring the stability of the machine learning feature factor, where the electronic device specifically includes the following contents:
a processor (processor), a memory (memory), a communication Interface (Communications Interface), and a bus; the processor, the memory and the communication interface complete mutual communication through the bus; the communication interface is used for realizing information transmission between the machine learning characteristic factor stability measuring device and relevant equipment such as a core service system, a user terminal, a relevant database and the like; the logic controller may be a desktop computer, a tablet computer, a mobile terminal, and the like, but the embodiment is not limited thereto. In this embodiment, the logic controller may be implemented with reference to the embodiment of the method for measuring stability of machine learning characteristic factors and the embodiment of the device for measuring stability of machine learning characteristic factors in the embodiments, and the contents thereof are incorporated herein, and repeated details are not repeated.
It is understood that the user terminal may include a smart phone, a tablet electronic device, a network set-top box, a portable computer, a desktop computer, a Personal Digital Assistant (PDA), a vehicle-mounted device, a smart wearable device, and the like. Wherein, intelligence wearing equipment can include intelligent glasses, intelligent wrist-watch, intelligent bracelet etc..
In practical applications, part of the method for machine learning feature factor stability measurement may be performed on the electronic device side as described above, or all operations may be performed in the client device. The selection may be specifically performed according to the processing capability of the client device, the limitation of the user usage scenario, and the like. This is not a limitation of the present application. The client device may further include a processor if all operations are performed in the client device.
The client device may have a communication module (i.e., a communication unit), and may be communicatively connected to a remote server to implement data transmission with the server. The server may include a server on the task scheduling center side, and in other implementation scenarios, the server may also include a server on an intermediate platform, for example, a server on a third-party server platform that is communicatively linked to the task scheduling center server. The server may include a single computer device, or may include a server cluster formed by a plurality of servers, or a server structure of a distributed apparatus.
Fig. 9 is a schematic block diagram of a system configuration of an electronic device 9600 according to an embodiment of the present application. As shown in fig. 9, the electronic device 9600 can include a central processor 9100 and a memory 9140; the memory 9140 is coupled to the central processor 9100. Notably, this fig. 9 is exemplary; other types of structures may also be used in addition to or in place of the structures to implement telecommunications or other functions.
In one embodiment, the machine learning feature factor stability metric method function may be integrated into the central processor 9100. The central processor 9100 can be configured to perform the following control:
step S101: and taking historical market data of a set time period as model training data.
Step S102: and performing model training on a preset machine learning prediction model according to the model training data and a set rolling mode to obtain a characteristic factor combination sequence.
Step S103: and determining the optimal characteristic factor combination according to the occurrence frequency and the occurrence time of each characteristic factor in the characteristic factor combination sequence.
As can be seen from the above description, the electronic device provided in the embodiment of the present application uses historical market data of a set time period as model training data; performing model training on a preset machine learning prediction model according to the model training data and a set rolling mode to obtain a characteristic factor combination sequence; and determining the optimal characteristic factor combination according to the occurrence frequency and the occurrence time of each characteristic factor in the characteristic factor combination sequence, thereby effectively improving the prediction effect of the machine learning model.
In another embodiment, the machine-learned feature factor stability measuring device may be configured separately from the central processor 9100, for example, the machine-learned feature factor stability measuring device may be configured as a chip connected to the central processor 9100, and the function of the machine-learned feature factor stability measuring method may be implemented by the control of the central processor.
As shown in fig. 9, the electronic device 9600 may further include: a communication module 9110, an input unit 9120, an audio processor 9130, a display 9160, and a power supply 9170. It is noted that the electronic device 9600 also does not necessarily include all of the components shown in fig. 9; in addition, the electronic device 9600 may further include components not shown in fig. 9, which may be referred to in the prior art.
As shown in fig. 9, a central processor 9100, sometimes referred to as a controller or operational control, can include a microprocessor or other processor device and/or logic device, which central processor 9100 receives input and controls the operation of the various components of the electronic device 9600.
The memory 9140 can be, for example, one or more of a buffer, a flash memory, a hard drive, a removable media, a volatile memory, a non-volatile memory, or other suitable device. The information relating to the failure may be stored, and a program for executing the information may be stored. And the central processing unit 9100 can execute the program stored in the memory 9140 to realize information storage or processing, or the like.
The input unit 9120 provides input to the central processor 9100. The input unit 9120 is, for example, a key or a touch input device. The power supply 9170 is used to provide power to the electronic device 9600. The display 9160 is used for displaying display objects such as images and characters. The display may be, for example, an LCD display, but is not limited thereto.
The memory 9140 may be a solid-state memory, e.g., read Only Memory (ROM), random Access Memory (RAM), a SIM card, or the like. There may also be a memory that holds information even when power is off, can be selectively erased, and is provided with more data, an example of which is sometimes referred to as an EPROM or the like. The memory 9140 could also be some other type of device. The memory 9140 includes a buffer memory 9141 (sometimes referred to as a buffer). The memory 9140 may include an application/function storage part 9142, the application/function storage part 9142 being used to store application programs and function programs or a flow for executing the operation of the electronic device 9600 by the central processing unit 9100.
The memory 9140 can also include a data store 9143, the data store 9143 being used to store data, such as contacts, digital data, pictures, sounds, and/or any other data used by an electronic device. The driver storage portion 9144 of the memory 9140 may include various drivers for the electronic device for communication functions and/or for performing other functions of the electronic device (e.g., messaging applications, contact book applications, etc.).
The communication module 9110 is a transmitter/receiver 9110 that transmits and receives signals via an antenna 9111. The communication module (transmitter/receiver) 9110 is coupled to the central processor 9100 to provide input signals and receive output signals, which may be the same as in the case of a conventional mobile communication terminal.
Based on different communication technologies, a plurality of communication modules 9110, such as a cellular network module, a bluetooth module, and/or a wireless local area network module, may be provided in the same electronic device. The communication module (transmitter/receiver) 9110 is also coupled to a speaker 9131 and a microphone 9132 via an audio processor 9130 to provide audio output via the speaker 9131 and receive audio input from the microphone 9132 to implement general telecommunications functions. The audio processor 9130 may include any suitable buffers, decoders, amplifiers and so forth. In addition, the audio processor 9130 is also coupled to the central processor 9100, thereby enabling recording locally through the microphone 9132 and enabling locally stored sounds to be played through the speaker 9131.
An embodiment of the present application further provides a computer-readable storage medium capable of implementing all the steps in the method for measuring stability of a machine-learned feature factor, where the execution subject of the method is a server or a client in the foregoing embodiments, and the computer-readable storage medium stores a computer program, where the computer program, when executed by a processor, implements all the steps in the method for measuring stability of a machine-learned feature factor, where the execution subject of the method for measuring stability of a machine-learned feature factor is a server or a client in the foregoing embodiments, for example, the processor implements the following steps when executing the computer program:
step S101: and taking historical market data of a set time period as model training data.
Step S102: and performing model training on a preset machine learning prediction model according to the model training data and a set rolling mode to obtain a characteristic factor combination sequence.
Step S103: and determining the optimal characteristic factor combination according to the occurrence frequency and the occurrence time of each characteristic factor in the characteristic factor combination sequence.
As can be seen from the above description, the computer-readable storage medium provided in the embodiments of the present application trains data by using historical market data for a set time period as model training data; performing model training on a preset machine learning prediction model according to the model training data and a set rolling mode to obtain a characteristic factor combination sequence; and determining the optimal characteristic factor combination according to the occurrence frequency and the occurrence time of each characteristic factor in the characteristic factor combination sequence, thereby effectively improving the prediction effect of the machine learning model.
Embodiments of the present application further provide a computer program product capable of implementing all steps of the method for measuring stability of a machine learning feature factor, where an execution subject of the method is a server or a client in the foregoing embodiments, and when being executed by a processor, the computer program/instruction implements the steps of the method for measuring stability of a machine learning feature factor, for example, the computer program/instruction implements the following steps:
step S101: and taking historical market data of a set time period as model training data.
Step S102: and performing model training on a preset machine learning prediction model according to the model training data and a set rolling mode to obtain a characteristic factor combination sequence.
Step S103: and determining the optimal characteristic factor combination according to the occurrence frequency and the occurrence time of each characteristic factor in the characteristic factor combination sequence.
As can be seen from the above description, the computer program product provided in the embodiments of the present application uses historical market data of a set time period as model training data; performing model training on a preset machine learning prediction model according to the model training data and a set rolling mode to obtain a characteristic factor combination sequence; and determining the optimal characteristic factor combination according to the occurrence frequency and the occurrence time of each characteristic factor in the characteristic factor combination sequence, thereby effectively improving the prediction effect of the machine learning model.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, apparatus, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention has been described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (devices), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The principle and the implementation mode of the invention are explained by applying specific embodiments in the invention, and the description of the embodiments is only used for helping to understand the method and the core idea of the invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present invention.
Claims (11)
1. A method for machine learning feature factor stability metric, the method comprising:
taking historical market data with a set time period as model training data;
performing model training on a preset machine learning prediction model according to the model training data and a set rolling mode to obtain a characteristic factor combination sequence;
and determining the optimal characteristic factor combination according to the occurrence frequency and the occurrence time of each characteristic factor in the characteristic factor combination sequence.
2. The method for measuring stability of a machine learning feature factor according to claim 1, wherein the performing model training on a preset machine learning prediction model according to the model training data and a set rolling mode to obtain a feature factor combination sequence comprises:
determining market fluctuation labels of the training data of the models according to the numerical comparison result between the training data of different models;
and performing model training on a preset machine learning prediction model according to the combination of different model training data, and determining a characteristic factor combination sequence.
3. The method for machine-learning the characteristic factor stability metric according to claim 1, wherein the determining an optimal characteristic factor combination according to the occurrence frequency and the occurrence time of each characteristic factor in the characteristic factor combination sequence comprises:
constructing a characteristic subset according to different characteristic factor combinations in the characteristic factor combination sequence;
and determining the occurrence frequency of each characteristic factor combination in all the characteristic subsets, and determining the optimal characteristic factor combination according to the occurrence frequency and the corresponding occurrence time.
4. The method for machine-learning the characteristic factor stability metric according to claim 1, wherein after determining the optimal characteristic factor combination according to the occurrence frequency and the occurrence time of each characteristic factor in the characteristic factor combination sequence, the method comprises:
combining the optimal characteristic factors to serve as model parameters of a preset machine learning prediction model;
and performing model training again on the preset machine learning prediction model to obtain the machine learning prediction model after model training.
5. A machine-learned feature factor stability metric apparatus, comprising:
the data accuracy module is used for taking historical market data of a set time period as model training data;
the rolling training module is used for carrying out model training on a preset machine learning prediction model according to the model training data and a set rolling mode to obtain a characteristic factor combination sequence;
and the optimal calculation module is used for determining an optimal characteristic factor combination according to the occurrence frequency and the occurrence time of each characteristic factor in the characteristic factor combination sequence.
6. The machine-learned feature factor stability metric apparatus of claim 5, wherein the rolling training module comprises:
the label determining unit is used for determining market fluctuation labels of the training data of the models according to the numerical comparison result between the training data of different models;
and the sequence determining unit is used for carrying out model training on the preset machine learning prediction model according to the combination of different model training data and determining a characteristic factor combination sequence.
7. The machine-learned feature factor stability metric apparatus of claim 5, wherein the optimal computation module comprises:
the subset construction unit is used for constructing a feature subset according to different feature factor combinations in the feature factor combination sequence;
and the frequency calculation unit is used for determining the occurrence frequency of each characteristic factor combination in all the characteristic subsets and determining the optimal characteristic factor combination according to the occurrence frequency and the corresponding occurrence time.
8. The machine-learned feature factor stability metric apparatus of claim 5, further comprising:
the parameter determining unit is used for combining the optimal characteristic factors to serve as model parameters of a preset machine learning prediction model;
and the model retraining unit is used for performing model retraining on the preset machine learning prediction model to obtain the machine learning prediction model after model training.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the steps of the method of machine learning feature factor stability metric of any of claims 1 to 4 when executing the program.
10. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method of machine learning feature factor stability metric of any one of claims 1 to 4.
11. A computer program product comprising computer program/instructions, characterized in that the computer program/instructions, when executed by a processor, implement the steps of the machine learning feature factor stability metric method of any of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210987135.6A CN115293364A (en) | 2022-08-17 | 2022-08-17 | Method and device for measuring stability of machine learning characteristic factor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210987135.6A CN115293364A (en) | 2022-08-17 | 2022-08-17 | Method and device for measuring stability of machine learning characteristic factor |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115293364A true CN115293364A (en) | 2022-11-04 |
Family
ID=83829452
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210987135.6A Pending CN115293364A (en) | 2022-08-17 | 2022-08-17 | Method and device for measuring stability of machine learning characteristic factor |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115293364A (en) |
-
2022
- 2022-08-17 CN CN202210987135.6A patent/CN115293364A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP3486771B1 (en) | Prediction of applications to be preloaded based on observed user behaviour and the order of starting the applications | |
| CN104704442A (en) | Information navigation on electronic devices | |
| CN110727868A (en) | Object recommendation method, device and computer-readable storage medium | |
| CN111369247A (en) | Cross-bank transaction data processing method and device | |
| CN110471735A (en) | Operational view interface method for pushing and device | |
| CN111736772A (en) | Storage space data processing method and device of distributed file system | |
| CN110992190A (en) | Asset configuration method and device based on user portrait | |
| CN113537510A (en) | Machine learning model data processing method and device based on unbalanced data set | |
| CN111046156B (en) | Method, device and server for determining rewarding data | |
| CN115293364A (en) | Method and device for measuring stability of machine learning characteristic factor | |
| CN115495519A (en) | Report data processing method and device | |
| CN114782716A (en) | Image matching method and device | |
| CN112116402A (en) | Credit risk customer-level default rate determination method and device | |
| CN112766698B (en) | Application service pressure determining method and device | |
| CN117611059B (en) | Inventory management method and system for dynamic adjustment of storage structure | |
| CN115409602A (en) | Method and device for predicting cash delivery amount of network points | |
| CN111523802B (en) | Method, device, equipment and medium for sending time response | |
| CN115545162A (en) | Atomic service combination optimization method and device | |
| CN112035324A (en) | Batch job execution condition monitoring method and device | |
| US20230133354A1 (en) | Predictive and customizable round up platform | |
| CN115545244A (en) | Commodity reservation method and device | |
| CN113902554A (en) | User classification method, device and system based on machine learning | |
| CN113487091A (en) | Network resource dynamic optimization method and device | |
| CN116738294A (en) | Customer classification method and device | |
| CN114595311A (en) | Audit question text data processing method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |