CN117370748A - Virtual resource data prediction method, device, computer equipment and storage medium - Google Patents

Virtual resource data prediction method, device, computer equipment and storage medium Download PDF

Info

Publication number
CN117370748A
CN117370748A CN202311153024.6A CN202311153024A CN117370748A CN 117370748 A CN117370748 A CN 117370748A CN 202311153024 A CN202311153024 A CN 202311153024A CN 117370748 A CN117370748 A CN 117370748A
Authority
CN
China
Prior art keywords
data
sequence
residual
prediction
data sequence
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
Application number
CN202311153024.6A
Other languages
Chinese (zh)
Inventor
师慧
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Bank of China Ltd
Original Assignee
Bank of China Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Bank of China Ltd filed Critical Bank of China Ltd
Priority to CN202311153024.6A priority Critical patent/CN117370748A/en
Publication of CN117370748A publication Critical patent/CN117370748A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/10—Pre-processing; Data cleansing
    • G06F18/15—Statistical pre-processing, e.g. techniques for normalisation or restoring missing data
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • 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
    • G06N3/0464—Convolutional networks [CNN, ConvNet]
    • 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
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2123/00—Data types
    • G06F2123/02—Data types in the time domain, e.g. time-series data
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Evolutionary Computation (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Molecular Biology (AREA)
  • Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • General Health & Medical Sciences (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Probability & Statistics with Applications (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The present application relates to a computer technology method, apparatus, computer device, storage medium and computer program product. Can be used in the financial field or other fields; the method can be used for predicting virtual resource data in the financial field and is used for improving prediction accuracy. The method comprises the following steps: acquiring a historical data sequence, wherein the historical data sequence is a data sequence obtained by sequencing virtual resource data according to data acquisition time; inputting the historical data sequence into a time sequence analysis model to predict virtual resource data, so as to obtain a predicted data sequence; determining a residual sequence based on the historical data sequence and the predicted data sequence; inputting the residual sequence into a neural network model for residual prediction to obtain a residual prediction sequence, wherein the neural network model is a model obtained by training data with a nonlinear rule; and obtaining a virtual resource data prediction result based on the residual prediction sequence and the prediction data sequence.

Description

Virtual resource data prediction method, device, computer equipment and storage medium
Technical Field
The present invention relates to the field of computer technologies, and in particular, to a virtual resource data prediction method, apparatus, computer device, storage medium, and computer program product.
Background
Conventional time series analysis models include an Auto Regression model (AR), a Moving Average Model (MA), a self-Moving model (AutoRegression Moving Average, ARMA), or an Auto Regression integral sliding Average model (Autoregressive Integrated Moving Average, ARIMA), which can be used to process data sequences formed over time, and after the model is built, future values can be predicted from historical data sequences. The idea of the model is essentially to learn a time-varying pattern from historical data, use the learned knowledge to predict future values, and time series analysis models benefit from the effectiveness of linear fitting to data sequences and the accuracy of short-term predictions, which have been widely used in recent years in various fields.
However, the data sequences in many existing fields generally have both a conventional linear rule and a complex and variable nonlinear rule, and if a time sequence analysis model is used for predicting future data, the prediction accuracy is low.
Disclosure of Invention
In view of the foregoing, it is desirable to provide a method, an apparatus, a computer device, a computer-readable storage medium, and a computer program product capable of predicting virtual resource data.
In a first aspect, the present application provides a virtual resource data prediction method. The method comprises the following steps:
acquiring a historical data sequence, wherein the historical data sequence is a data sequence obtained by sequencing virtual resource data according to data acquisition time;
inputting the historical data sequence into a time sequence analysis model to predict virtual resource data, so as to obtain a predicted data sequence;
determining a residual sequence based on the historical data sequence and the predicted data sequence;
inputting the residual sequence into a neural network model for residual prediction to obtain a residual prediction sequence, wherein the neural network model is a model obtained by training data with a nonlinear rule;
and obtaining a virtual resource data prediction result based on the residual prediction sequence and the prediction data sequence.
In one embodiment, the time series analysis model is a model trained from data having a linear law.
In one embodiment, the inputting the historical data sequence into a time sequence analysis model to predict virtual resource data, to obtain a predicted data sequence, includes:
inputting the historical data sequence into the time sequence analysis model;
and performing first-order differential processing on the historical data sequence through the time sequence analysis model, and performing virtual resource data prediction on the processed historical data sequence to obtain the predicted data sequence.
In one embodiment, the training process of the neural network model includes:
obtaining a plurality of residual samples, wherein each residual sample has a corresponding residual prediction calibration value;
normalizing the residual error sample;
inputting the normalized residual samples into an initial model for residual prediction, and performing inverse normalization processing on the prediction result to obtain residual prediction values corresponding to the residual samples;
and carrying out parameter adjustment on the initial model based on the residual prediction value and the residual prediction calibration value corresponding to each residual sample until the errors of the residual prediction value and the residual prediction calibration value are within a preset error range, so as to obtain the neural network model.
In one embodiment, the determining a residual sequence based on the historical data sequence and the predicted data sequence includes:
determining a difference between first data in the historical data sequence and second data at a corresponding position in the predicted data sequence;
and obtaining the residual data sequence based on the difference value.
In one embodiment, each first data in the historical data sequence includes sub-data of multiple dimensions; the determining a difference between the first data in the historical data sequence and the second data at a corresponding position in the predicted data sequence comprises:
for each first data, determining second data of a corresponding position from the predicted data sequence,
for each corresponding first data and second data, determining a difference value between the sub data of each dimension in the first data and the sub data of the corresponding dimension in the data, and taking the difference value as the difference value between the first data and the second data.
In a second aspect, the present application further provides a virtual resource data prediction apparatus. The device comprises:
the acquisition module is used for acquiring a historical data sequence, wherein the historical data sequence is a data sequence obtained by sequencing virtual resource data according to data acquisition time;
the first data prediction module is used for inputting the historical data sequence into a time sequence analysis model to predict virtual resource data so as to obtain a predicted data sequence;
a residual determination module, configured to determine a residual sequence based on the historical data sequence and the predicted data sequence;
the second data prediction module is used for inputting the residual sequence into a neural network model for residual prediction to obtain a residual prediction sequence, wherein the neural network model is a model obtained by training data with a nonlinear rule;
and the third data prediction module is used for obtaining a virtual resource data prediction result based on the residual prediction sequence and the prediction data sequence.
In a third aspect, the present application also provides a computer device. The computer device comprises a memory storing a computer program and a processor implementing the steps of virtual resource data prediction described above when the processor executes the computer program.
In a fourth aspect, the present application also provides a computer-readable storage medium. The computer readable storage medium has stored thereon a computer program which, when executed by a processor, performs the steps of virtual resource data prediction described above.
In a fifth aspect, the present application also provides a computer program product. The computer program product comprises a computer program which, when executed by a processor, implements the steps of virtual resource data prediction described above.
According to the virtual resource data prediction method, the device, the computer equipment, the storage medium and the computer program product, the historical data sequence is input into the time sequence analysis model to conduct data prediction to obtain the predicted data sequence, the time sequence analysis model can conduct accurate prediction on data with a linear rule, then a residual sequence is determined based on the historical data sequence and the predicted data sequence, the residual can measure errors of the model, the residual sequence can represent the nonlinear rule of the historical data sequence due to the fact that the data has the nonlinear rule, the residual sequence is input into the neural network model to conduct residual prediction to obtain the residual prediction sequence, the neural network model is obtained through training of the data with the nonlinear rule, accurate prediction can be conducted on the data with the nonlinear rule through the neural network model, and a final virtual resource data prediction result is obtained based on the residual prediction sequence and the predicted data sequence. According to the method and the device, the time sequence analysis model and the neural network model are combined to conduct data prediction, prediction advantages of the time sequence analysis model and the neural network model in two aspects of data with a linear rule and data with a nonlinear rule can be integrated, accurate prediction of the data with the linear rule and the data with the nonlinear rule is achieved, and prediction accuracy is improved.
Drawings
FIG. 1 is an application environment diagram of a virtual resource data prediction method in one embodiment;
FIG. 2 is a flowchart of a method for predicting virtual resource data according to another embodiment;
FIG. 3 is a block diagram of a virtual resource data prediction apparatus in one embodiment;
fig. 4 is an internal structural diagram of a computer device in one embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application will be further described in detail with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the present application.
The method can be used in the technical field of computers, and the method and the device for predicting the virtual resource data can be used for predicting deposit data in the financial field, can also be used in any field except the financial field, and are not limited in application field. Carrying out
In one embodiment, as shown in fig. 1, a virtual resource data prediction method is provided, where this embodiment is applied to a terminal to illustrate the method, it is understood that the method may also be applied to a server, and may also be applied to a system including a terminal and a server, and implemented through interaction between the terminal and the server. In this embodiment, the method includes the steps of:
step S101, acquiring a historical data sequence, wherein the historical data sequence is a data sequence obtained by sequencing virtual resource data according to data acquisition time;
the terminal acquires a historical data sequence, each virtual resource data in the historical data sequence has a corresponding data acquisition time, the virtual resource data are sequenced according to the data acquisition time to form a data sequence, and the data acquisition time refers to a time before the current time, so that the formed data sequence is called a historical data sequence;
the historical data sequence is a data sequence with both linear rules and nonlinear rules.
Step S102, inputting a historical data sequence into a time sequence analysis model to predict virtual resource data, so as to obtain a predicted data sequence;
the time sequence analysis model can learn a model of the change of historical data along with time, the learned knowledge is utilized to predict future values, the historical data sequence is input into the time sequence analysis model, virtual resource data at the future moment is predicted, a predicted data sequence is obtained, the predicted data sequence is virtual resource data of a predicted future period, and the predicted data sequence is equivalent to the linear rule of the historical data sequence learned through the time sequence analysis model, so that the prediction of the virtual resource data is carried out.
The time series analysis model may be an AR model, MA model, ARMA model, or ARIMA model, which can accurately predict data with linear regularity.
Since the time series analysis model can only accurately predict data with a linear rule, a predicted data sequence obtained by prediction is not accurate in practice.
Step S103, determining a residual sequence based on the historical data sequence and the predicted data sequence;
the terminal obtains a residual sequence by subtracting the historical data sequence from the predicted data sequence.
Step S104, inputting the residual sequence into a neural network model for residual prediction to obtain a residual prediction sequence, wherein the neural network model is a model obtained by training data with a nonlinear rule;
the neural network model is a model which is obtained by training the virtual resource data with nonlinear rules, so that the neural network model has higher prediction accuracy on the virtual resource data with nonlinear rules,
the residual error can measure the error of the time sequence analysis model, and the error is caused by the fact that the data has a nonlinear rule, so that the residual error sequence can represent the nonlinear rule of the historical data sequence, the residual error sequence is input into the neural network model for residual error prediction, the residual error prediction sequence can be obtained, and the method is equivalent to learning the nonlinear rule of the historical data sequence through the neural network model, and accordingly prediction of virtual resource data is carried out.
In this embodiment, the neural network model refers to a recurrent neural network (LSTM, long Short Term Memory) model, and the LSTM model can accurately predict data having a nonlinear rule.
And step 105, obtaining a virtual resource data prediction result based on the residual prediction sequence and the prediction data sequence.
And the terminal performs summation processing on the residual prediction sequence and the prediction data sequence to obtain a virtual resource data prediction result, wherein the virtual resource data prediction result is virtual resource data which is predicted to be obtained in future time.
In this embodiment, a historical data sequence is input into a time sequence analysis model to perform data prediction to obtain a predicted data sequence, the time sequence analysis model can accurately predict data with a linear rule, then a residual sequence is determined based on the historical data sequence and the predicted data sequence, and a residual error can measure errors of the model. According to the method and the device, the time sequence analysis model and the neural network model are combined to conduct data prediction, prediction advantages of the time sequence analysis model and the neural network model in two aspects of data with a linear rule and data with a nonlinear rule can be integrated, accurate prediction of the data with the linear rule and the data with the nonlinear rule is achieved, and prediction accuracy is improved.
In one embodiment, the time series analysis model is a model trained from data having a linear law.
The time sequence analysis model is obtained by training virtual resource data with a linear rule, so that the virtual resource data with the linear rule can be accurately predicted.
In this embodiment, for the existing case that the LSTM model alone predicts the historical data sequence, if the historical data sequence has both a linear rule and a nonlinear rule, the prediction accuracy of a single LSTM model is still low.
In one embodiment, inputting the historical data sequence into a time sequence analysis model to predict virtual resource data to obtain a predicted data sequence includes:
inputting the historical data sequence into a time sequence analysis model;
and performing first-order differential processing on the historical data sequence through a time sequence analysis model, and performing virtual resource data prediction on the processed historical data sequence to obtain a predicted data sequence.
In a model training stage, a time sequence analysis model is established, whether a sequence is stable or not is firstly required to be determined, a first-order difference is carried out on an original sequence to eliminate non-stationarity, and after the sequence is stable, the model recognition stage is carried out, so that the time sequence analysis model is established;
in the model use stage, the historical data sequence is subjected to first-order differential processing to eliminate non-stationarity, and then the processed historical data sequence is subjected to virtual resource data prediction.
In this embodiment, the data is subjected to first-order difference to eliminate non-stationarity, so that the prediction accuracy of the time series analysis model can be improved.
In one embodiment, the training process of the neural network model includes:
obtaining a plurality of residual samples, wherein each residual sample has a corresponding residual prediction calibration value;
normalizing the residual error sample;
inputting the normalized residual samples into an initial model for residual prediction, and performing inverse normalization processing on the prediction result to obtain residual prediction values corresponding to the residual samples;
and carrying out parameter adjustment on the initial model based on the residual prediction value and the residual prediction calibration value corresponding to each residual sample until the errors of the residual prediction value and the residual prediction calibration value are within a preset error range, so as to obtain the neural network model.
In order to find out the optimal parameter configuration of the highest precision returned by the model, 80% of all residual samples can be set as a training set and 20% as a test set in a model training stage, firstly, normalization processing is carried out on the residual samples, 80% of the training set in the normalized residual samples is input into an initial model for residual prediction, inverse normalization processing is carried out on a prediction result to obtain residual prediction values corresponding to all the residual samples, and then parameter adjustment is carried out on the initial model based on the residual prediction values and the residual prediction calibration values corresponding to all the residual samples until errors of the residual prediction values and the residual prediction calibration values are within a preset error range, so that a neural network model is obtained;
and further testing the neural network model based on 20% of the test set, and adjusting parameters to obtain the optimal parameter configuration of the model.
The predicted value of the LSTM model of 50 units each of two layers may be selected as the residual predicted value.
In this embodiment, the neural network model is normalized, so that the model solving speed can be increased.
In one embodiment, determining the residual sequence based on the historical data sequence and the predicted data sequence includes:
determining a difference between the first data in the historical data sequence and the second data at a corresponding position in the predicted data sequence;
and obtaining a residual data sequence based on the difference value.
The number of virtual resource data in the historical data sequence is the same as that of the predicted data sequence, each first data in the historical data sequence has second data in a corresponding position in the predicted data sequence, the terminal can subtract each first data from the corresponding second data to obtain a difference value, and then the residual data sequence is determined based on the obtained plurality of difference values.
In one embodiment, each first data in the historical data sequence includes sub-data of multiple dimensions; determining a difference between the first data in the historical data sequence and the second data at a corresponding location in the predicted data sequence, comprising:
for each first data, determining second data of the corresponding position from the predicted data sequence,
for each corresponding first data and second data, determining the difference value between the sub data of each dimension in the first data and the sub data of the corresponding dimension in the data, and taking the difference value as the difference value between the first data and the second data.
Each first data in the historical data sequence comprises a plurality of dimension sub-data, for example, for the financial data sequence, each first data comprises three dimension sub-data of deposit, loan and withdrawal; in the predicted data sequence predicted by the historical data sequence, each second data of the predicted data sequence also comprises a plurality of sub-data with the same dimension, for example, each second data also comprises sub-data with three dimensions of deposit, loan, withdrawal and the like;
part of the data in the financial industry is regular, and basically does not float every month, such as payroll, loan return, baby growth funds, personal pension payment and the like, but the personal deposit, loan and withdrawal rules are not so clear and stable, and have linear rules and nonlinear rules.
For each corresponding first data and second data: the terminal determines the difference between the sub-data in the first data and the sub-data in the second data in the same dimension, for example, calculates the difference between the deposit data in the first data and the deposit data in the second data, calculates the difference between the loan data in the first data and the loan data in the second data, calculates the difference between the withdrawal data in the first data and the withdrawal data in the second data, and takes the calculated difference as the difference between the first data and the corresponding second data;
it can be seen that, since each first data in the historical data sequence includes sub-data with multiple dimensions, each residual value in the residual data sequence obtained in the application also includes sub-residual values with multiple dimensions, for example, including sub-residual values with three dimensions of deposit, loan, and withdrawal.
In this embodiment, each first data in the historical data sequence includes virtual resource data with multiple dimensions, and virtual resource data prediction efficiency is improved by simultaneously predicting the virtual resource data with multiple dimensions.
The method and the device are mainly applied to virtual resource data prediction in the financial field, and the main functions of a financial institution deposit balance financial institution are financing and collecting capital, wherein the financial institution deposit balance refers to deposit amount of the financial institution in a period of time. With the increase of innovative deposit products and the enhancement of public financial consciousness, the heat of deposit is continuously increased, the development trend of deposit balances of financial institutions is researched, and the method has important practical significance for making macroscopic economic strategy and financial currency policy, guiding financial departments to operate and manage, keeping national economy and health development and the like.
As shown in fig. 2, the scheme of the present application is specifically summarized as follows:
the historical data sequence Yt in the financial field has both a linear rule and a nonlinear rule;
firstly, learning a linear rule of a historical data sequence Yt through an ARIMA model, predicting to obtain L't, wherein L't is a predicted data sequence, a residual sequence Nt of the historical data sequence Yt and the predicted data sequence L't can represent a nonlinear rule of the historical data sequence, and the residual sequence Nt=Yt-L't;
and predicting the residual sequence Nt through an LSTM model to obtain a residual prediction sequence N't, and then predicting a result Y' t=L 't+N't by virtual resource data.
The history data sequence Yt may be a deposit data sequence including a plurality of deposit data, and the resulting virtual resource data prediction result Y't is a prediction result of the deposit data.
Each data in the history data sequence Yt may also include sub-data of multiple dimensions of deposit data, loan data, withdrawal data, etc., for example yt= [ data 1, data 2, data 3], data 1 including deposit data, loan data, and withdrawal data, and the same data 2 and data 3 also including deposit data, loan data, and withdrawal data. The corresponding predicted data sequence L 't= [ data 4, data 5, data 6], data 4 (or data 5 or data 6) also includes deposit data, loan data, and withdrawal data, the calculated residual sequence nt= [ residual 1, residual 2, residual 3], residual 1 (or residual 2 or residual 3) also includes deposit data residual, loan data residual, and withdrawal data residual, the same residual predicted sequence N' t= [ residual 4, residual 5, residual 6], residual 4 (or residual 5 or residual 6) also includes deposit data residual, loan data residual, and withdrawal data residual;
and finally predicting the obtained virtual resource data sequence Y' t= [ data 4+ residual error 4, data 5+ residual error 5 and data 6+ residual error 6].
The method provided by the invention predicts the monthly deposit balance of the bank by utilizing the linear advantage of the ARIMA model on time sequence prediction and the mining capability of the LSTM model on nonlinear data. Firstly, establishing a model for historical deposit data by using an ARIMA model to obtain a linear predicted value and a residual error value, then modeling and predicting a deposit balance residual error data set by using an LSTM neural network to obtain a fitting value of the deposit balance residual error value, and finally, combining the linear predicted value and the residual error fitting value to obtain the predicted value of the deposit balance.
Finance is the core of modern economy and is the blood vessel of physical economy. The steady increase in bank deposit balance shows an increase in the bank's ability to aggregate funds. The trend change research for judging the data according to the principle and the method of the financial time series analysis has a certain practical significance. As for the prediction of bank deposit balance, the ARIMA and LSTM combined model can integrate the advantages of the ARIMA and LSTM combined model in the aspects of linearity and nonlinearity, fully utilizes the strong data feature extraction capability and learning capability of the ARIMA and LSTM combined model, and avoids the defect of a single model.
It should be understood that, although the steps in the flowcharts related to the above embodiments are sequentially shown as indicated by arrows, these steps are not necessarily sequentially performed in the order indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in the flowcharts described in the above embodiments may include a plurality of steps or a plurality of stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of the steps or stages is not necessarily performed sequentially, but may be performed alternately or alternately with at least some of the other steps or stages.
Based on the same inventive concept, the embodiment of the application also provides a virtual resource data prediction device for realizing the above-mentioned virtual resource data prediction method. The implementation of the solution provided by the apparatus is similar to the implementation described in the above method, so the specific limitation in the embodiments of one or more virtual resource data prediction apparatus provided below may be referred to the limitation of the virtual resource data prediction method hereinabove, and will not be described herein.
In one embodiment, as shown in FIG. 3, there is provided a virtual resource data prediction apparatus 300 comprising:
the acquiring module 301 is configured to acquire a historical data sequence, where the historical data sequence is a data sequence obtained by ordering virtual resource data according to a data acquisition time;
the first data prediction module 302 is configured to input the historical data sequence into the time sequence analysis model to perform virtual resource data prediction, so as to obtain a predicted data sequence;
a residual determining module 303, configured to determine a residual sequence based on the historical data sequence and the predicted data sequence;
the second data prediction module 304 is configured to input the residual sequence into a neural network model to perform residual prediction, so as to obtain a residual prediction sequence, where the neural network model is a model obtained by training data with a nonlinear rule;
and a third data prediction module 305, configured to obtain a virtual resource data prediction result based on the residual prediction sequence and the predicted data sequence.
In one embodiment, the time series analysis model is a model trained from data having a linear law.
In one embodiment, the first data prediction module 302 is specifically configured to:
inputting the historical data sequence into a time sequence analysis model;
and performing first-order differential processing on the historical data sequence through a time sequence analysis model, and performing virtual resource data prediction on the processed historical data sequence to obtain a predicted data sequence.
In one embodiment, the training process of the neural network model includes:
obtaining a plurality of residual samples, wherein each residual sample has a corresponding residual prediction calibration value;
normalizing the residual error sample;
inputting the normalized residual samples into an initial model for residual prediction, and performing inverse normalization processing on the prediction result to obtain residual prediction values corresponding to the residual samples;
and carrying out parameter adjustment on the initial model based on the residual prediction value and the residual prediction calibration value corresponding to each residual sample until the errors of the residual prediction value and the residual prediction calibration value are within a preset error range, so as to obtain the neural network model.
In one embodiment, the residual determination module 303 is specifically configured to:
determining a difference between the first data in the historical data sequence and the second data at a corresponding position in the predicted data sequence;
and obtaining a residual data sequence based on the difference value.
In one embodiment, each first data in the historical data sequence includes sub-data of multiple dimensions; the residual determination module 303 is specifically configured to, when determining a difference between the first data in the historical data sequence and the second data at a corresponding position in the predicted data sequence:
for each first data, determining second data of the corresponding position from the predicted data sequence,
for each corresponding first data and second data, determining the difference value between the sub data of each dimension in the first data and the sub data of the corresponding dimension in the data, and taking the difference value as the difference value between the first data and the second data.
The above-described respective modules in the virtual resource data predicting apparatus may be implemented in whole or in part by software, hardware, and a combination thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.
In one embodiment, a computer device is provided, which may be a terminal, and the internal structure of which may be as shown in fig. 4. The computer device includes a processor, a memory, an input/output interface, a communication interface, a display unit, and an input means. The processor, the memory and the input/output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input/output interface. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The input/output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used for carrying out wired or wireless communication with an external terminal, and the wireless mode can be realized through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. The computer program when executed by a processor implements a virtual resource data prediction method. The display unit of the computer device is used for forming a visual picture, and can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, can also be a key, a track ball or a touch pad arranged on the shell of the computer equipment, and can also be an external keyboard, a touch pad or a mouse and the like.
Those skilled in the art will appreciate that the structures shown in FIG. 4 are block diagrams only and do not constitute a limitation of the computer device on which the present aspects apply, and that a particular computer device may include more or less components than those shown, or may combine some of the components, or have a different arrangement of components.
In one embodiment, a computer device is provided that includes a memory having a computer program stored therein and a processor that when executing the computer program performs the steps of virtual resource data prediction described above.
In one embodiment, a computer readable storage medium is provided having a computer program stored thereon which, when executed by a processor, performs the steps of virtual resource data prediction described above.
In one embodiment, a computer program product is provided comprising a computer program which, when executed by a processor, performs the steps of virtual resource data prediction described above.
It should be noted that, the user information (including, but not limited to, user equipment information, user personal information, etc.) and the data (including, but not limited to, data for analysis, stored data, presented data, etc.) referred to in the present application are information and data authorized by the user or sufficiently authorized by each party, and the collection, use and processing of the related data are required to comply with the related laws and regulations and standards of the related countries and regions.
Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, database, or other medium used in the various embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, high density embedded nonvolatile Memory, resistive random access Memory (ReRAM), magnetic random access Memory (Magnetoresistive Random Access Memory, MRAM), ferroelectric Memory (Ferroelectric Random Access Memory, FRAM), phase change Memory (Phase Change Memory, PCM), graphene Memory, and the like. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory, and the like. By way of illustration, and not limitation, RAM can be in the form of a variety of forms, such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM), and the like. The databases referred to in the various embodiments provided herein may include at least one of relational databases and non-relational databases. The non-relational database may include, but is not limited to, a blockchain-based distributed database, and the like. The processors referred to in the embodiments provided herein may be general purpose processors, central processing units, graphics processors, digital signal processors, programmable logic units, quantum computing-based data processing logic units, etc., without being limited thereto.
The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The above examples only represent a few embodiments of the present application, which are described in more detail and are not to be construed as limiting the scope of the present application. It should be noted that it would be apparent to those skilled in the art that various modifications and improvements could be made without departing from the spirit of the present application, which would be within the scope of the present application. Accordingly, the scope of protection of the present application shall be subject to the appended claims.

Claims (10)

1. A virtual resource data prediction method, comprising:
acquiring a historical data sequence, wherein the historical data sequence is a data sequence obtained by sequencing virtual resource data according to data acquisition time;
inputting the historical data sequence into a time sequence analysis model to predict virtual resource data, so as to obtain a predicted data sequence;
determining a residual sequence based on the historical data sequence and the predicted data sequence;
inputting the residual sequence into a neural network model for residual prediction to obtain a residual prediction sequence, wherein the neural network model is a model obtained by training data with a nonlinear rule;
and obtaining a virtual resource data prediction result based on the residual prediction sequence and the prediction data sequence.
2. The method of claim 1, wherein the time series analysis model is a model trained from data having a linear law.
3. The method of claim 1, wherein inputting the historical data sequence into a time series analysis model for virtual resource data prediction to obtain a predicted data sequence comprises:
inputting the historical data sequence into the time sequence analysis model;
and performing first-order differential processing on the historical data sequence through the time sequence analysis model, and performing virtual resource data prediction on the processed historical data sequence to obtain the predicted data sequence.
4. The method of claim 1, wherein the training process of the neural network model comprises:
obtaining a plurality of residual samples, wherein each residual sample has a corresponding residual prediction calibration value;
normalizing the residual error sample;
inputting the normalized residual samples into an initial model for residual prediction, and performing inverse normalization processing on the prediction result to obtain residual prediction values corresponding to the residual samples;
and carrying out parameter adjustment on the initial model based on the residual prediction value and the residual prediction calibration value corresponding to each residual sample until the errors of the residual prediction value and the residual prediction calibration value are within a preset error range, so as to obtain the neural network model.
5. The method of claim 1, wherein the determining a residual sequence based on the historical data sequence and a predicted data sequence comprises:
determining a difference between first data in the historical data sequence and second data at a corresponding position in the predicted data sequence;
and obtaining the residual data sequence based on the difference value.
6. The method of claim 5, wherein each first data in the historical data sequence comprises sub-data of multiple dimensions; the determining a difference between the first data in the historical data sequence and the second data at a corresponding position in the predicted data sequence comprises:
for each first data, determining second data of a corresponding position from the predicted data sequence,
for each corresponding first data and second data, determining a difference value between the sub data of each dimension in the first data and the sub data of the corresponding dimension in the data, and taking the difference value as the difference value between the first data and the second data.
7. A virtual resource data prediction apparatus, comprising:
the acquisition module is used for acquiring a historical data sequence, wherein the historical data sequence is a data sequence obtained by sequencing virtual resource data according to data acquisition time;
the first data prediction module is used for inputting the historical data sequence into a time sequence analysis model to predict virtual resource data so as to obtain a predicted data sequence;
a residual determination module, configured to determine a residual sequence based on the historical data sequence and the predicted data sequence;
the second data prediction module is used for inputting the residual sequence into a neural network model for residual prediction to obtain a residual prediction sequence, wherein the neural network model is a model obtained by training data with a nonlinear rule;
and the third data prediction module is used for obtaining a virtual resource data prediction result based on the residual prediction sequence and the prediction data sequence.
8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any of claims 1 to 6 when the computer program is executed.
9. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 6.
10. A computer program product comprising a computer program, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 6.
CN202311153024.6A 2023-09-07 2023-09-07 Virtual resource data prediction method, device, computer equipment and storage medium Pending CN117370748A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202311153024.6A CN117370748A (en) 2023-09-07 2023-09-07 Virtual resource data prediction method, device, computer equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202311153024.6A CN117370748A (en) 2023-09-07 2023-09-07 Virtual resource data prediction method, device, computer equipment and storage medium

Publications (1)

Publication Number Publication Date
CN117370748A true CN117370748A (en) 2024-01-09

Family

ID=89403027

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202311153024.6A Pending CN117370748A (en) 2023-09-07 2023-09-07 Virtual resource data prediction method, device, computer equipment and storage medium

Country Status (1)

Country Link
CN (1) CN117370748A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN118013217A (en) * 2024-04-10 2024-05-10 南昌理工学院 Internet of things communication data missing processing method and system

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN118013217A (en) * 2024-04-10 2024-05-10 南昌理工学院 Internet of things communication data missing processing method and system

Similar Documents

Publication Publication Date Title
CN110717671B (en) Method and device for determining contribution degree of participants
CN112116245A (en) Credit risk assessment method, credit risk assessment device, computer equipment and storage medium
CN117370748A (en) Virtual resource data prediction method, device, computer equipment and storage medium
CN117035980A (en) Resource borrowing evaluation method, device, computer equipment and storage medium
CN115358330A (en) Client user loss prediction method, device, equipment and storage medium
CN114240598A (en) Credit line model generation method, credit line determination method and device
CN114170000A (en) Credit card user risk category identification method, device, computer equipment and medium
Zhang et al. Research on data mining algorithm of logistics time series based on intelligent integrated network structure
Krishna et al. Accuracy of forecasting models using Bootstrap test and Friedman’s test
CN112308295B (en) Method and device for predicting default probability
CN116977049A (en) Method, device, computer equipment and storage medium for processing resource application service
CN116703216A (en) Service scene scoring method, device, computer equipment and storage medium
CN117454320A (en) Target object identification method, device, computer equipment and storage medium
CN118313625A (en) Financial service resource allocation method, device, equipment and storage medium
CN115879608A (en) Resource information prediction method, resource information prediction device, computer equipment and storage medium
CN118154300A (en) Mortgage parameter processing method, mortgage parameter processing device, computer equipment and storage medium
CN114693431A (en) Credit evaluation method, apparatus, device, medium, and program product
CN117036040A (en) Asset development trend determination method and device and computer equipment
CN115719185A (en) Business data acquisition method and device, computer equipment and storage medium
CN118195769A (en) Product overdue state prediction method, device, computer equipment, storage medium and computer program product
CN118132091A (en) Service model processing method, device, computer equipment and storage medium
CN118520288A (en) Sample data generating method, device, computer equipment and storage medium
CN117575772A (en) Abnormal user detection method and device, computer equipment and storage medium
CN115757958A (en) Product recommendation method and device, computer equipment and storage medium
CN116781536A (en) Traffic prediction method, traffic prediction device, traffic prediction apparatus, traffic prediction device, traffic prediction program product, and traffic prediction program product

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