CN111192144A - Financial data prediction method, device, equipment and storage medium - Google Patents
Financial data prediction method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN111192144A CN111192144A CN202010004452.2A CN202010004452A CN111192144A CN 111192144 A CN111192144 A CN 111192144A CN 202010004452 A CN202010004452 A CN 202010004452A CN 111192144 A CN111192144 A CN 111192144A
- Authority
- CN
- China
- Prior art keywords
- financial data
- data
- historical
- financial
- time
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/04—Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- 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 invention provides a financial data prediction method, which comprises the following steps: obtaining historical financial data and real-time financial data from one or more data sources; the historical financial data are normalized according to historical time points so that the historical data and the current real-time financial data form a corresponding relation; performing characteristic processing on the historical financial data according to financial processing standards to obtain training data; establishing a layer-by-layer deep learning prediction model according to the historical financial data, and continuously correcting the deep learning prediction model according to the training data until the model converges; and inputting the corresponding relation and the real-time financial data into the deep learning prediction model to obtain a financial data prediction result. The method and the system can automatically obtain the financial data change trend probability concerned by the user according to the financial data recording condition of each financial data website and the browsing habit information of the user, simplify complicated operation and improve user experience.
Description
Technical Field
The invention relates to the technical field of computers, in particular to a financial data prediction method, device, equipment and storage medium.
Background
In the financial field, there are a great variety of financial data, such as stocks, futures, funds, precious metals, foreign exchanges, etc., and in order to effectively obtain data information in the financial field in time and to perform a prediction operation using the financial data information to provide a policy reference for a user transaction, thereby quantifying the transaction, it is urgently required to perform a fast and accurate identification process on the financial data, and read and analyze historical financial data in real time to form a reference suggestion, thereby reducing the risk of investment transaction. For example, CN107274007A discloses a method for predicting financial data based on artificial neural network to predict future trends of various financial markets; or the patent CN105022825A relates to a financial variety price prediction method combining financial news mining and financial historical data, and predicts the future price trend of the financial market by using a combined prediction model combining multivariate linear regression and ARIMA; for another example, patent CN109711665A discloses a prediction model construction method based on financial wind control data and related equipment, which mainly solves the problem of unbalanced types of financial wind control data by preprocessing financial wind control data and optimizing and adjusting parameters of financial wind control data models.
Therefore, how to improve the accuracy and the accuracy of the financial data mining technology, reduce the data calling and processing cost and fully utilize the data prediction value, the actual problems to be dealt with urgently in the practical application of the technology still have a lot of unreported concrete solutions.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention provides a financial data prediction method, a device, equipment and a storage medium, and the specific technical scheme of the invention is as follows:
the invention provides a financial data prediction method in a first aspect, which comprises the following steps:
obtaining historical financial data and real-time financial data from one or more data sources;
the historical financial data are normalized according to historical time points so that the historical data and the current real-time financial data form a corresponding relation;
performing characteristic processing on the historical financial data according to financial processing standards to obtain training data;
establishing a layer-by-layer deep learning prediction model according to the historical financial data, and continuously correcting the deep learning prediction model according to the training data until the model converges;
and inputting the corresponding relation and the real-time financial data into the deep learning prediction model to obtain a financial data prediction result.
A second aspect of the invention provides a prediction apparatus for financial data, the apparatus comprising: the acquisition module is used for acquiring historical financial data and real-time financial data from one or more data sources; the data corresponding module is used for regulating the historical financial data according to historical time points so as to enable the historical data and the current real-time financial data to form a corresponding relation; the training module is used for performing characteristic processing on the historical financial data according to financial processing standards to obtain training data; the correction module is used for establishing a layer-by-layer deep learning prediction model according to the historical financial data and continuously correcting the deep learning prediction model according to the training data until the model converges; and the prediction module is used for inputting the corresponding relation and the real-time financial data into the deep learning prediction model to obtain a financial data prediction result.
A third aspect of the invention provides a prediction device for financial data comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the financial data prediction method when executing the computer program.
A fourth aspect of the present invention provides a computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the financial data prediction method.
In a fifth aspect, an embodiment of the present application provides a computer program product, which, when running on a prediction device, causes a terminal device to execute the client management method according to any one of the first aspect.
It is understood that the beneficial effects of the second aspect to the fifth aspect can be referred to the related description of the first aspect, and are not described herein again.
The beneficial effects obtained by the invention comprise: 1. the method is characterized in that the trend of financial data is predicted by applying a deep learning technology and combining with some technical characteristics representing the trend of financial data in the market, and finally, opinion reference is provided for participating in financial activities according to the probability of trend change of the financial data; 2. the financial data change trend probability concerned by the user can be automatically obtained according to the financial data recording condition of each financial data website and the browsing habit information of the user, so that the complicated operation is simplified, and the user experience is improved; 3. the consistency and the uniformity of the financial data are ensured, and errors in repeated entry are avoided; the data consistency is solved without complicated manual checking or designing a complex interface, and the system development and maintenance cost is saved.
Drawings
The present invention will be further understood from the following description taken in conjunction with the accompanying drawings, the emphasis instead being placed upon illustrating the principles of the embodiments.
FIG. 1 is a flow chart illustrating a method for predicting financial data according to one embodiment of the present invention;
FIG. 2 is a flow chart illustrating a method for constructing a prediction model according to one embodiment of the present invention;
FIG. 3 is a schematic diagram of a prediction apparatus for financial data according to one embodiment of the present invention;
FIG. 4 is a schematic diagram of a prediction device for financial data according to one embodiment of the present invention;
FIG. 5 is a schematic diagram of a prediction process of a financial item price prediction method in the prior art CN 105022825A;
fig. 6 is a schematic flow chart of a prediction model construction method in CN109711665A in the prior art.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail below with reference to embodiments thereof; it should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. Other systems, methods, and/or features of the present embodiments will become apparent to those skilled in the art upon review of the following detailed description. It is intended that all such additional systems, methods, features and advantages be included within this description, be within the scope of the invention, and be protected by the accompanying claims. Additional features of the disclosed embodiments are described in, and will be apparent from, the detailed description that follows.
The same or similar reference numerals in the drawings of the embodiments of the present invention correspond to the same or similar components; in the description of the present invention, it should be understood that if there is an orientation or positional relationship indicated by the terms "upper", "lower", "left", "right", etc. based on the orientation or positional relationship shown in the drawings, it is only for convenience of describing the present invention and simplifying the description, but it is not intended to indicate or imply that the device or component referred to must have a specific orientation, be constructed in a specific orientation, and be operated, and therefore, the terms describing the positional relationship in the drawings are only used for illustrative purposes and are not to be construed as limiting the present patent, and the specific meaning of the terms described above will be understood by those of ordinary skill in the art according to the specific circumstances.
The invention relates to a financial data prediction method, a device, equipment and a storage medium, which explain the following embodiments according to the attached drawings:
the first embodiment is as follows:
the embodiment provides a financial data prediction method based on deep learning in a first aspect, including:
s101, acquiring historical financial data and real-time financial data from one or more data sources;
s102, the historical financial data are normalized according to historical time points so that the historical data and the current real-time financial data form a corresponding relation;
s103, performing characteristic processing on the historical financial data according to financial processing standards to obtain training data;
s104, establishing a layer-by-layer deep learning prediction model according to the historical financial data, and continuously correcting the deep learning prediction model according to the training data until the model converges;
and S105, inputting the corresponding relation and the real-time financial data into the deep learning prediction model to obtain a financial data prediction result.
Wherein the historical financial data comprises a first primary historical financial data and a second secondary historical exogenous characteristic data, wherein the first historical financial data comprises: recording financial data source information, historical financial data characteristic information, historical time points corresponding to the historical financial data characteristic information one by one and a financial data historical time sequence; the second auxiliary historical exogenous feature data is a feature event corresponding to the historical financial data feature information.
Wherein, the regularizing the historical financial data according to the recording time point so as to form a corresponding relation between the historical data and the current real-time financial data comprises:
acquiring financial historical data in a preset time period including the historical time point according to the historical time point, and performing normalization processing according to a preset standardized data format rule to obtain normalized historical financial data in the preset time period;
acquiring current real-time financial data according to the preset time period, and performing normalization processing according to the preset standardized data format rule to obtain normalized current real-time financial data in the preset time period;
calculating to obtain financial data similarity according to the normalized financial historical data in the preset time period and the normalized current real-time financial data in the preset time period, and judging whether the financial data similarity is greater than or equal to the preset financial data similarity;
if the financial data similarity is greater than or equal to the preset financial data similarity, setting the corresponding relation between the historical financial data of the preset time period including the historical time point and the current real-time financial data of the preset time period, and storing marks;
and if the financial data similarity is smaller than the preset financial data similarity, moving the historical time point according to a historical time axis, continuously acquiring the financial historical data in a preset time period including the historical time point and the current real-time financial data in the preset time period, and calculating the financial data similarity.
And acquiring the times of forming the corresponding relation between the historical financial data and the current real-time financial data, and the minimum time interval and the maximum time interval of forming the corresponding relation according to the historical time sequence and the historical time point.
And according to the historical time sequence and the historical time points, obtaining the historical time points which form the corresponding relationship most recently from the current time point of the current real-time financial data, and calculating to obtain the time intervals which form the corresponding relationship most recently according to the current time points and the historical time points which form the corresponding relationship most recently.
And calculating a probability value of the latest formed corresponding relation according to the minimum time interval and the maximum time interval for forming the corresponding relation and the time interval for forming the latest corresponding relation, and outputting the financial data prediction result according to the probability value.
Establishing a layer-by-layer deep learning prediction model according to the historical financial data, and continuously correcting the deep learning prediction model according to the training data until the model converges, wherein the step of establishing the layer-by-layer deep learning prediction model according to the historical financial data comprises the following steps:
s201, obtaining a plurality of initial deep learning prediction models by applying a plurality of groups of initialization training data; initializing training data, wherein the initializing training data corresponds to the initial deep learning prediction model one to one;
s202, storing check points in each initial deep learning prediction model;
s203, calculating a prediction weight of the initial deep learning prediction model corresponding to the check point according to the check point;
and S204, carrying out model convergence fusion on each initial deep learning prediction model according to each prediction weight to obtain the deep learning prediction model.
A second aspect of the present embodiments provides a prediction apparatus for financial data, the apparatus comprising: the acquisition module is used for acquiring historical financial data and real-time financial data from one or more data sources; the data corresponding module is used for regulating the historical financial data according to historical time points so as to enable the historical data and the current real-time financial data to form a corresponding relation; the training module is used for performing characteristic processing on the historical financial data according to financial processing standards to obtain training data; the correction module is used for establishing a layer-by-layer deep learning prediction model according to the historical financial data and continuously correcting the deep learning prediction model according to the training data until the model converges; and the prediction module is used for inputting the corresponding relation and the real-time financial data into the deep learning prediction model to obtain a financial data prediction result.
A third aspect of the present embodiment provides a prediction apparatus for financial data, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the financial data prediction method when executing the computer program.
A fourth aspect of the present embodiment provides a computer-readable storage medium storing a computer program, wherein the computer program is executed by a processor to implement the financial data prediction method.
A fifth aspect of the present embodiment provides a computer program product, which, when run on a prediction device, causes a terminal device to execute the client management method according to any one of the first aspects.
Example two:
the embodiment provides a financial data prediction method based on deep learning in a first aspect, including:
s101, acquiring historical financial data and real-time financial data from one or more data sources; the data source data files acquire source data files from financial systems, financial web pages and the like. External financial data from external sources is provided to one or more external databases. The external database provides the external financial data to the internal database through control processing, and historical financial data and real-time financial data are acquired. The private internal financial data is also provided to the internal database. The calculation processing unit requests financial data, which may be external or internal financial data, from an internal database. The financial data is processed according to a predetermined specialized algorithm to generate further financial data which is provided to the server processing unit for display in the user interface. Specifically, a large number of data files are stored in a core database, an overseas database and other databases of the financial system, the data files can be acquired from the databases through a source data file acquisition module to serve as source data files, and the acquired source data files are only stored without any other processing. Therefore, the time for storing the source data files by the source data file acquisition module can be set to be about one month, and the source data files can be cleaned up after the time is up to leave a storage space for new source data files. Expired source data files may be stored on an extended storage medium, such as tape, for backup, and if large data technologies can be incorporated, the expired source data files may be stored using distributed inexpensive storage devices without having to be backed up to tape. Because the storage life of the data after being put in storage is not longer than two years, if the data needs to be stored for more than two years, the data does not need to be recovered from the magnetic tape any more, and the use of the historical data is more convenient. For very common data such as a client table, an account table and a transaction table, key treatment can be considered, and the online retention time can be specially designed. In addition, the financial data feature set acquired according to the source data file is backed up and stored for calling later-period historical data, the source data file in the time period does not need to be acquired again, the financial data feature selection does not need to be performed again, and the time for re-selection is consumed, so that the efficiency of financial data analysis is greatly improved. And correspondingly distributing type labels aiming at the source data and the financial data feature set one by one, storing the source data according to the type labels, and recording the corresponding relation among the source data, the type labels and the storage positions of the source data and the financial data feature set. In addition, the source data file, the financial data to be modeled and the financial data characteristic set are respectively stored in a source data storage area, a standard data storage area and a selected characteristic storage area, so that the storage and calling efficiency and the regularity are improved. After the data stored in the selected feature storage area is selected, the data required by positioning can be easily provided through searching the data registry, which is equivalent to providing a uniform data service interface, and the data requirements can be unambiguously and uniquely mapped onto bare data, so that a user can conveniently call financial data feature set data, predict and visually operate financial data, and perform feature recognition extraction processing on historical financial data through financial processing standards to obtain training data suitable for the deep learning prediction model.
S102, the historical financial data are normalized according to historical time points so that the historical data and the current real-time financial data form a corresponding relation;
s103, performing characteristic processing on the historical financial data according to financial processing standards to obtain training data; the financial data processing standard can be divided into a plurality of classes according to different standard types: (1) according to the financial business activity score, dividing financial data into banking business data, security business data, insurance business data and data in the aspects of trust, consultation and the like, wherein the banking business data also comprises data in the aspects of credit, accounting, savings, settlement, interest rate and the like; the securities business data also comprises data in aspects of quotation, entrustment, bargain, capital market supply and demand, marketing company operation state and the like; the insurance business data also comprises data on application insurance, claim settlement, investment and the like. The data reflects the characteristics, laws and operation conditions of financial activities from a certain side; (2) according to the information content, the financial data can be divided into financial system internal data and financial system external data, wherein the financial system internal data refers to data generated in various business activities of the financial institution, and the financial system external information refers to data collected and stored by the financial institution for developing various financial activities in the whole society; (3) according to the source of the acquired information, the financial data are divided into data from the inside of the financial institution, data from the market and data from the whole society, wherein the data from the inside of the financial institution refers to data generated in each business activity of the financial institution; data from the market refers to data generated during market competition and trading; data from the whole society refers to data acquired by financial institutions from governments, businesses, institutions, individuals, including data on revenue, management, credit, and the like.
S104, establishing a layer-by-layer deep learning prediction model according to the historical financial data, and continuously correcting the deep learning prediction model according to the training data until the model converges; specifically, a training input vector and a label value corresponding to the training input vector are determined according to historical financial data. In training the predictive model, the training input vector may be in the form of a three-dimensional vector. Wherein the training input vector comprises a training period parameter and a characteristic parameter
And S105, inputting the corresponding relation and the real-time financial data into the deep learning prediction model to obtain a financial data prediction result.
Wherein the historical financial data comprises a first primary historical financial data and a second secondary historical exogenous characteristic data, wherein the first historical financial data comprises: recording financial data source information, historical financial data characteristic information, historical time points corresponding to the historical financial data characteristic information one by one and a financial data historical time sequence; the second auxiliary historical exogenous feature data is a feature event corresponding to the historical financial data feature information. The deep learning prediction model provided by the invention can be used for learning and expressing exogenous characteristic data besides time characteristics, and increasing the constraint on a plurality of factors influencing the trend of the financial data, so that the prediction model closer to the actual condition of the financial data is obtained, and a more accurate prediction result of the change of the financial data is obtained. In addition, the historical time series and the exogenous feature data are converted into training input vectors and label values corresponding to the training input vectors according to preset search and conversion rules. The exogenous features are determined according to factors which have the greatest influence on the financial data change, and can be obtained by analyzing historical event data accompanying the development of the financial industry by a user or by applying a specific analysis algorithm. The historical events may be financial storms, enterprise listings, enterprise complaints, bank bankruptures, and the like. In addition, the second auxiliary historical extrinsic feature data further comprises user browsing habit information so as to provide reference opinions for participating in financial data activities according to user habits through a deep learning prediction model.
Wherein, the regularizing the historical financial data according to the recording time point so as to form a corresponding relation between the historical data and the current real-time financial data comprises:
acquiring financial historical data in a preset time period including the historical time point according to the historical time point, and performing normalization processing according to a preset standardized data format rule to obtain normalized historical financial data in the preset time period; acquiring current real-time financial data according to the preset time period, and performing normalization processing according to the preset standardized data format rule to obtain normalized current real-time financial data in the preset time period;
calculating to obtain financial data similarity according to the normalized financial historical data in the preset time period and the normalized current real-time financial data in the preset time period, and judging whether the financial data similarity is greater than or equal to the preset financial data similarity; in essence, the invention uses the best prediction model used for prediction of historical financial data with greater similarity as a candidate model for predicting the vector to be predicted, and therefore, the trend of the current real-time financial data change is known. And if the matching degree of the historical financial data with the current real-time financial data is very high and is more than or equal to a preset matching degree in a certain time period in the history, the matching degree is taken as the financial data similarity to be output.
If the financial data similarity is greater than or equal to the preset financial data similarity, setting the corresponding relation between the historical financial data of the preset time period including the historical time point and the current real-time financial data of the preset time period, and storing marks;
and if the financial data similarity is smaller than the preset financial data similarity, moving the historical time point according to a historical time axis, continuously acquiring the financial historical data in a preset time period including the historical time point and the current real-time financial data in the preset time period, and calculating the financial data similarity.
And acquiring the times of forming the corresponding relation between the historical financial data and the current real-time financial data, and the minimum time interval and the maximum time interval of forming the corresponding relation according to the historical time sequence and the historical time point.
And according to the historical time sequence and the historical time points, obtaining the historical time points which form the corresponding relationship most recently from the current time point of the current real-time financial data, and calculating to obtain the time intervals which form the corresponding relationship most recently according to the current time points and the historical time points which form the corresponding relationship most recently.
And calculating a probability value of the latest formed corresponding relation according to the minimum time interval and the maximum time interval for forming the corresponding relation and the time interval for forming the latest corresponding relation, and outputting the financial data prediction result according to the probability value. For example, after obtaining the minimum time interval and the maximum time interval, the difference between the minimum time interval and the maximum time interval is calculated. In addition, time intervals between historical financial data acquired in past times and current real-time financial data forming corresponding relations are arranged according to the size sequence, the time intervals with the recently formed corresponding relations are subjected to interval matching with the time intervals with the historical financial data acquired in past times and the current real-time financial data forming corresponding relations, if the same time intervals exist in the history in a matching mode, the difference value between the time intervals and the minimum time interval is obtained through calculation, then the difference value between the difference value and the minimum time interval and the difference value between the difference value and the maximum time interval are obtained through calculation, a first operation result is obtained, then the first operation result is subjected to subtraction operation with the number '1', a second operation result is obtained, and the second operation result is output as the probability value of the newly formed corresponding relations. If the time interval is not matched with the same time interval in the history, comparing the time interval with the minimum time interval, if the time interval is larger than the minimum time interval, calculating to obtain the difference value between the time interval and the minimum time interval, and performing the operation steps.
In addition, in order to improve the prediction accuracy of the prediction model, different initialized deep learning prediction models are obtained through different initialized training data, the prediction weight of each initial deep learning prediction model is determined according to the check points in each initial prediction model, and the final model parameters are obtained through weighted average according to the prediction weight. Specifically, the establishing a layer-by-layer deep learning prediction model according to the historical financial data, and continuously correcting the deep learning prediction model according to the training data until the model converges includes:
s201, obtaining a plurality of initial deep learning prediction models by applying a plurality of groups of initialization training data; initializing training data, wherein the initializing training data corresponds to the initial deep learning prediction model one to one;
s202, storing check points in each initial deep learning prediction model;
s203, calculating a prediction weight of the initial deep learning prediction model corresponding to the check point according to the check point;
and S204, carrying out model convergence fusion on each initial deep learning prediction model according to each prediction weight to obtain the deep learning prediction model.
A second aspect of the present embodiments provides a prediction apparatus for financial data, the apparatus comprising:
the acquisition module 100 is configured to acquire historical financial data and real-time financial data from one or more data sources.
The data corresponding module 200 is configured to normalize the historical financial data according to a historical time point, so that the historical data and the current real-time financial data form a corresponding relationship;
the training module 300 is configured to perform feature processing on the historical financial data according to a financial processing standard to obtain training data;
a correction module 400, configured to establish a layer-by-layer deep learning prediction model according to the historical financial data, and continuously correct the deep learning prediction model according to the training data until the model converges;
and the prediction module 500 is configured to input the corresponding relationship and the real-time financial data into the deep learning prediction model to obtain a financial data prediction result.
A third aspect of the present embodiment provides a prediction device 4 for financial data, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the financial data prediction method when executing the computer program.
Referring to fig. 4, an embodiment of the present application further provides a prediction apparatus 4 for financial data, including: at least one processor 40, a memory 41, and a computer program 42 stored in the memory 41 and executable on the at least one processor 42, the processor 40 implementing the steps in any of the various method embodiments described above when executing the computer program 42, such as the steps S101 to S105 described in fig. 1. Alternatively, the processor 40, when executing the computer program 42, implements the functions of the modules in the device embodiments, such as the functions of the modules 100 to 500 shown in fig. 3.
Illustratively, the computer program 42 may be partitioned into one or more modules/units that are stored in the memory 41 and executed by the processor 40 to accomplish the present application. The one or more modules/units may be a series of computer program instruction segments capable of performing specific functions that describe the execution of the computer program 42 in the prediction device 4 of the financial data.
Those skilled in the art will appreciate that fig. 4 is merely an example of a predictive device for financial data and does not constitute a limitation of a predictive device for financial data, and may include more or fewer components than shown, or some components in combination, or different components, e.g., the predictive device for financial data may also include input and output devices, buses, etc.
The Processor 40 may be a Central Processing Unit (CPU), and the Processor 40 may also be other general purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate arrays (FPGAs) or other Programmable logic devices, discrete Gate or transistor logic devices, discrete hardware components, etc. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The memory 41 may in some embodiments be an internal storage unit of the prediction device 4 of financial data, such as a hard disk or a memory of the prediction device 4 of financial data. The memory 41 may be an external storage device of the prediction device 4 for the financial data in other embodiments, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), and the like, which are provided on the prediction device 4 for the financial data. Further, the memory 41 may also include both an internal storage unit and an external storage device of the prediction device 4 of the financial data. The memory 41 is used for storing an operating system, an application program, a Boot Loader (Boot Loader), data, and other programs, such as program codes of the computer programs. The memory 41 may also be used to temporarily store data that has been output or is to be output.
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-mentioned division of each functional module is illustrated, and in practical applications, the above-mentioned functional allocation may be performed by different functional units or modules according to requirements, that is, the internal structure of the apparatus is divided into different functional units or modules, so as to perform all or part of the above-mentioned functions. Each functional module in the embodiments may be integrated in one processing unit, or each unit may exist alone physically, or two or more units are integrated in one unit, and the integrated unit may be implemented in a form of hardware, or in a form of software functional unit. In addition, specific names of the functional modules are only used for distinguishing one functional module from another, and are not used for limiting the protection scope of the application. The specific working process of the modules in the apparatus may refer to the corresponding process in the foregoing method embodiment, and is not described herein again.
The present embodiment also provides a computer-readable storage medium, which stores a computer program that, when being executed by a processor, implements the steps of the method embodiments described above.
The present embodiment provides a computer program product, which when run on a mobile terminal, enables the mobile terminal to implement the steps of the above-mentioned method embodiments when executed.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, all or part of the processes in the methods of the embodiments described above can be implemented by a computer program, which can be stored in a computer-readable storage medium and can implement the steps of the embodiments of the methods described above when the computer program is executed by a processor. Wherein the computer program comprises computer program code, which may be in the form of source code, object code, an executable file or some intermediate form, etc. The computer readable medium may include at least: any entity or device capable of carrying computer program code to a photographing apparatus/terminal apparatus, a recording medium, computer Memory, Read-Only Memory (ROM), random-access Memory (RAM), an electrical carrier signal, a telecommunications signal, and a software distribution medium. Such as a usb-disk, a removable hard disk, a magnetic or optical disk, etc. In certain jurisdictions, computer-readable media may not be an electrical carrier signal or a telecommunications signal in accordance with legislative and patent practice.
In summary, the financial data prediction method based on deep learning disclosed by the present invention has the following beneficial technical effects: 1. the method is characterized in that the trend of financial data is predicted by applying a deep learning technology and combining with some technical characteristics representing the trend of financial data in the market, and finally, opinion reference is provided for participating in financial activities according to the probability of trend change of the financial data; 2. the financial data change trend probability concerned by the user can be automatically obtained according to the financial data recording condition of each financial data website and the browsing habit information of the user, so that the complicated operation is simplified, and the user experience is improved; 3. the consistency and the uniformity of the financial data are ensured, and errors in repeated entry are avoided; the data consistency is solved without complicated manual checking or designing a complex interface, and the system development and maintenance cost is saved.
Although the invention has been described above with reference to various embodiments, it should be understood that many changes and modifications may be made without departing from the scope of the invention. That is, the methods, systems, and devices discussed above are examples, and various configurations may omit, replace, or add various processes or components as appropriate. For example, in alternative configurations, the methods may be performed in an order different than that described and/or various components may be added, omitted, and/or combined. Moreover, features described with respect to certain configurations may be combined in various other configurations, as different aspects and elements of the configurations may be combined in a similar manner. Further, elements therein may be updated as technology evolves, i.e., many of the elements are examples and do not limit the scope of the disclosure or claims.
Specific details are given in the description to provide a thorough understanding of the exemplary configurations including implementations. However, configurations may be practiced without these specific details, such as well-known circuits, processes, algorithms, structures, and techniques, which have been shown without unnecessary detail in order to avoid obscuring the configurations. This description provides example configurations only, and does not limit the scope, applicability, or configuration of the claims. Rather, the foregoing description of the configurations will provide those skilled in the art with an enabling description for implementing the described techniques. Various changes may be made in the function and arrangement of elements without departing from the spirit or scope of the disclosure.
It is intended that the foregoing detailed description be regarded as illustrative rather than limiting, and that it be understood that it is the following claims, including all equivalents, that are intended to define the spirit and scope of this invention. The above examples are to be construed as merely illustrative and not limitative of the remainder of the disclosure. After reading the description of the invention, the skilled person can make various changes or modifications to the invention, and these equivalent changes and modifications also fall into the scope of the invention defined by the claims.
Claims (10)
1. A financial data prediction method based on deep learning is characterized by comprising the following steps:
obtaining historical financial data and real-time financial data from one or more data sources;
the historical financial data are normalized according to historical time points so that the historical data and the current real-time financial data form a corresponding relation;
performing characteristic processing on the historical financial data according to financial processing standards to obtain training data;
establishing a layer-by-layer deep learning prediction model according to the historical financial data, and continuously correcting the deep learning prediction model according to the training data until the model converges;
and inputting the corresponding relation and the real-time financial data into the deep learning prediction model to obtain a financial data prediction result.
2. The deep learning-based financial data prediction method of claim 1 wherein the historical financial data comprises a first primary historical financial data, a second secondary historical exogenous feature data, wherein the first historical financial data comprises: recording financial data source information, historical financial data characteristic information, historical time points corresponding to the historical financial data characteristic information one by one and a financial data historical time sequence; the second auxiliary historical exogenous feature data is a feature event corresponding to the historical financial data feature information.
3. The deep learning-based financial data prediction method of claim 2, wherein the warping the historical financial data according to a recording time point to form a corresponding relationship between the historical data and the current real-time financial data comprises:
acquiring financial historical data in a preset time period including the historical time point according to the historical time point, and performing normalization processing according to a preset standardized data format rule to obtain normalized historical financial data in the preset time period;
acquiring current real-time financial data according to the preset time period, and performing normalization processing according to the preset standardized data format rule to obtain normalized current real-time financial data in the preset time period;
calculating to obtain financial data similarity according to the normalized financial historical data in the preset time period and the normalized current real-time financial data in the preset time period, and judging whether the financial data similarity is greater than or equal to the preset financial data similarity;
if the financial data similarity is greater than or equal to the preset financial data similarity, setting the corresponding relation between the historical financial data of the preset time period including the historical time point and the current real-time financial data of the preset time period, and storing marks;
and if the financial data similarity is smaller than the preset financial data similarity, moving the historical time point according to a historical time axis, continuously acquiring the financial historical data in a preset time period including the historical time point and the current real-time financial data in the preset time period, and calculating the financial data similarity.
4. The deep learning-based financial data prediction method according to claim 3, wherein the number of times the historical financial data forms a correspondence with the current real-time financial data, the minimum time interval and the maximum time interval at which the correspondence is formed are obtained based on the historical time series and the historical time points.
5. The deep learning-based financial data prediction method of claim 4, wherein a historical time point at which the correspondence relationship is formed most recently from the current time point of the current real-time financial data is obtained from the historical time series and the historical time points, and a time interval at which the correspondence relationship is formed most recently is calculated from the current time point and the historical time points at which the correspondence relationship is formed most recently.
6. The deep learning-based financial data prediction method of claim 5, wherein a probability value of the most recently formed correspondence is calculated according to the minimum time interval, the maximum time interval and the time interval of the most recently formed correspondence, and the financial data prediction result is output according to the probability value.
7. The method of deep learning based financial data prediction according to claim 6, wherein said building a layer-by-layer deep learning prediction model based on said historical financial data and continuously modifying said deep learning prediction model based on said training data until model convergence comprises:
obtaining a plurality of initial deep learning prediction models by applying a plurality of groups of initialization training data; initializing training data, wherein the initializing training data corresponds to the initial deep learning prediction model one to one;
saving a check point in each of the initial deep learning prediction models;
calculating a prediction weight of an initial deep learning prediction model corresponding to the check point according to the check point;
and carrying out model convergence fusion on each initial deep learning prediction model according to each prediction weight to obtain the deep learning prediction model.
8. A prediction apparatus for financial data, the apparatus comprising:
the acquisition module is used for acquiring historical financial data and real-time financial data from one or more data sources;
the data corresponding module is used for regulating the historical financial data according to historical time points so as to enable the historical data and the current real-time financial data to form a corresponding relation;
the training module is used for performing characteristic processing on the historical financial data according to financial processing standards to obtain training data;
the correction module is used for establishing a layer-by-layer deep learning prediction model according to the historical financial data and continuously correcting the deep learning prediction model according to the training data until the model converges;
and the prediction module is used for inputting the corresponding relation and the real-time financial data into the deep learning prediction model to obtain a financial data prediction result.
9. A prediction device for financial data comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor when executing the computer program implements a method as claimed in any one of claims 1 to 7.
10. A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out the method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010004452.2A CN111192144A (en) | 2020-01-03 | 2020-01-03 | Financial data prediction method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010004452.2A CN111192144A (en) | 2020-01-03 | 2020-01-03 | Financial data prediction method, device, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111192144A true CN111192144A (en) | 2020-05-22 |
Family
ID=70708024
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010004452.2A Pending CN111192144A (en) | 2020-01-03 | 2020-01-03 | Financial data prediction method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111192144A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111612261A (en) * | 2020-05-28 | 2020-09-01 | 洋浦美诺安电子科技有限责任公司 | Financial big data analysis system based on block chain |
| CN111798061A (en) * | 2020-07-08 | 2020-10-20 | 洋浦美诺安电子科技有限责任公司 | Financial data prediction system based on block chain and cloud computing |
| CN111798062A (en) * | 2020-07-08 | 2020-10-20 | 洋浦美诺安电子科技有限责任公司 | Financial data prediction system based on block chain and big data |
| CN112182118A (en) * | 2020-09-29 | 2021-01-05 | 中国平安人寿保险股份有限公司 | Target object prediction method based on multiple data sources and related equipment thereof |
| CN112766596A (en) * | 2021-01-29 | 2021-05-07 | 苏州思萃融合基建技术研究所有限公司 | Building energy consumption prediction model construction method, energy consumption prediction method and device |
| WO2022141877A1 (en) * | 2020-12-31 | 2022-07-07 | 平安科技（深圳）有限公司 | Financial data processing method, apparatus and device, and storage medium |
| CN116700632A (en) * | 2023-08-07 | 2023-09-05 | 湖南中盈梦想商业保理有限公司 | High-reliability financial information data storage method |
| CN112766596B (en) * | 2021-01-29 | 2024-04-16 | 苏州思萃融合基建技术研究所有限公司 | Construction method of building energy consumption prediction model, energy consumption prediction method and device |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120116996A1 (en) * | 2010-11-04 | 2012-05-10 | Investpic, Llc | Method and system for analyzing investment information |
| US20160125317A1 (en) * | 2014-10-31 | 2016-05-05 | Capital One Financial Corporation | System and method for transaction learning |
| CN106991506A (en) * | 2017-05-16 | 2017-07-28 | 深圳先进技术研究院 | Intelligent terminal and its stock trend forecasting method based on LSTM |
| CN107230151A (en) * | 2017-05-27 | 2017-10-03 | 北京奇虎科技有限公司 | Stock certificate data Forecasting Methodology and device, electronic equipment, computer-readable storage medium |
| CN107481048A (en) * | 2017-08-08 | 2017-12-15 | 哈尔滨工业大学深圳研究生院 | A kind of financial kind price expectation method and system based on mixed model |
| CN108460679A (en) * | 2018-02-28 | 2018-08-28 | 电子科技大学 | Merge the depth network intelligence investment system data analysing method of attention mechanism |
| CN109472700A (en) * | 2018-11-20 | 2019-03-15 | 中山大学 | Prediction technique, server and the storage medium of stock price |
| CN109493139A (en) * | 2018-11-08 | 2019-03-19 | 兴业证券股份有限公司 | Similar market judgement, preferred method and storage medium |
| CN109657851A (en) * | 2018-12-12 | 2019-04-19 | 哈尔滨工业大学（深圳） | Forecasting of Stock Prices method, apparatus, equipment and the storage medium of event driven |
| CN109840626A (en) * | 2019-01-16 | 2019-06-04 | 大连理工大学 | A kind of Stock Market Forecasting method based on convolutional neural networks model |
| CN110264013A (en) * | 2019-06-27 | 2019-09-20 | 苏州派维斯信息科技有限公司 | Financial prediction method and system |
-
2020
- 2020-01-03 CN CN202010004452.2A patent/CN111192144A/en active Pending
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120116996A1 (en) * | 2010-11-04 | 2012-05-10 | Investpic, Llc | Method and system for analyzing investment information |
| US20160125317A1 (en) * | 2014-10-31 | 2016-05-05 | Capital One Financial Corporation | System and method for transaction learning |
| CN106991506A (en) * | 2017-05-16 | 2017-07-28 | 深圳先进技术研究院 | Intelligent terminal and its stock trend forecasting method based on LSTM |
| CN107230151A (en) * | 2017-05-27 | 2017-10-03 | 北京奇虎科技有限公司 | Stock certificate data Forecasting Methodology and device, electronic equipment, computer-readable storage medium |
| CN107481048A (en) * | 2017-08-08 | 2017-12-15 | 哈尔滨工业大学深圳研究生院 | A kind of financial kind price expectation method and system based on mixed model |
| CN108460679A (en) * | 2018-02-28 | 2018-08-28 | 电子科技大学 | Merge the depth network intelligence investment system data analysing method of attention mechanism |
| CN109493139A (en) * | 2018-11-08 | 2019-03-19 | 兴业证券股份有限公司 | Similar market judgement, preferred method and storage medium |
| CN109472700A (en) * | 2018-11-20 | 2019-03-15 | 中山大学 | Prediction technique, server and the storage medium of stock price |
| CN109657851A (en) * | 2018-12-12 | 2019-04-19 | 哈尔滨工业大学（深圳） | Forecasting of Stock Prices method, apparatus, equipment and the storage medium of event driven |
| CN109840626A (en) * | 2019-01-16 | 2019-06-04 | 大连理工大学 | A kind of Stock Market Forecasting method based on convolutional neural networks model |
| CN110264013A (en) * | 2019-06-27 | 2019-09-20 | 苏州派维斯信息科技有限公司 | Financial prediction method and system |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112330067A (en) * | 2020-05-28 | 2021-02-05 | 洋浦美诺安电子科技有限责任公司 | Financial big data analysis system based on block chain |
| CN112330067B (en) * | 2020-05-28 | 2021-06-25 | 山东深度网络科技有限公司 | Financial big data analysis system based on block chain |
| CN111612261A (en) * | 2020-05-28 | 2020-09-01 | 洋浦美诺安电子科技有限责任公司 | Financial big data analysis system based on block chain |
| CN111798062B (en) * | 2020-07-08 | 2021-04-02 | 中信消费金融有限公司 | Financial data prediction system based on block chain and big data |
| CN111798061B (en) * | 2020-07-08 | 2020-12-29 | 深圳市荣亿达科技发展有限公司 | Financial data prediction system based on block chain and cloud computing |
| CN111798062A (en) * | 2020-07-08 | 2020-10-20 | 洋浦美诺安电子科技有限责任公司 | Financial data prediction system based on block chain and big data |
| CN111798061A (en) * | 2020-07-08 | 2020-10-20 | 洋浦美诺安电子科技有限责任公司 | Financial data prediction system based on block chain and cloud computing |
| CN112182118A (en) * | 2020-09-29 | 2021-01-05 | 中国平安人寿保险股份有限公司 | Target object prediction method based on multiple data sources and related equipment thereof |
| CN112182118B (en) * | 2020-09-29 | 2023-12-05 | 中国平安人寿保险股份有限公司 | Target object prediction method based on multiple data sources and related equipment thereof |
| WO2022141877A1 (en) * | 2020-12-31 | 2022-07-07 | 平安科技（深圳）有限公司 | Financial data processing method, apparatus and device, and storage medium |
| CN112766596A (en) * | 2021-01-29 | 2021-05-07 | 苏州思萃融合基建技术研究所有限公司 | Building energy consumption prediction model construction method, energy consumption prediction method and device |
| CN112766596B (en) * | 2021-01-29 | 2024-04-16 | 苏州思萃融合基建技术研究所有限公司 | Construction method of building energy consumption prediction model, energy consumption prediction method and device |
| CN116700632A (en) * | 2023-08-07 | 2023-09-05 | 湖南中盈梦想商业保理有限公司 | High-reliability financial information data storage method |
| CN116700632B (en) * | 2023-08-07 | 2023-10-24 | 湖南中盈梦想商业保理有限公司 | High-reliability financial information data storage method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111192144A (en) | Financial data prediction method, device, equipment and storage medium | |
| CN112598513B (en) | Method and device for identifying stockholder risk transaction behaviors | |
| CN115423578B (en) | Bid bidding method and system based on micro-service containerized cloud platform | |
| CN111738762A (en) | Method, device, equipment and storage medium for determining recovery price of poor assets | |
| CN115545886A (en) | Overdue risk identification method, overdue risk identification device, overdue risk identification equipment and storage medium | |
| CN111260189A (en) | Risk control method, risk control device, computer system and readable storage medium | |
| CN112950347A (en) | Resource data processing optimization method and device, storage medium and terminal | |
| CN107172311B (en) | Service evaluation method and terminal equipment | |
| CN111242779B (en) | Financial data characteristic selection and prediction method, device, equipment and storage medium | |
| CN115809930A (en) | Anti-fraud analysis method, device, equipment and medium based on data fusion matching | |
| CN115760404A (en) | Stock reduction scheme generation method, system, terminal and storage medium | |
| CN112288559B (en) | Stock right-based financial data accounting method, device, equipment and storage medium | |
| CN114298825A (en) | Method and device for extremely evaluating repayment volume | |
| CN115049137A (en) | Prediction method and device of transaction yield, storage medium and electronic equipment | |
| CN114612220A (en) | Wind control strategy adjusting method and device, electronic equipment and storage medium | |
| CN114511314A (en) | Payment account management method and device, computer equipment and storage medium | |
| CN113421014A (en) | Target enterprise determination method, device, equipment and storage medium | |
| CN110544165A (en) | credit risk score card creating method and device and electronic equipment | |
| Yeh et al. | Predicting failure of P2P lending platforms through machine learning: The case in China | |
| US11915309B1 (en) | Computer-based systems for calculating risk of asset transfers | |
| US11392573B1 (en) | Systems and methods for generating and maintaining data objects | |
| US20220398660A1 (en) | System and method for computational shelf forecasting | |
| CN114707488B (en) | Data processing method, device, computer equipment and storage medium | |
| CN115760384A (en) | Abnormal behavior recognition method, abnormal behavior recognition device, electronic device, and storage medium | |
| CN114240653A (en) | Noble metal pairing transaction method, device, equipment and readable storage medium |
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 |