Fund party data configuration method and device based on big data and storage medium

Abstract

The invention relates to a big data technology, and provides a fund party data allocation method and device based on big data and a storage medium, and the method comprises the steps: carrying out the stage classification of a financing demand obtained from a financing party in advance, and obtaining a financing stage corresponding to the financing demand; constructing a sparse matrix of the financing party according to the financing stage, and performing financing data filling on the sparse matrix to obtain a demand matrix of the financing party; calculating the degree of adaptation between the demand matrix and table data in historical data configuration tables obtained from a fund party in advance, and selecting the historical data configuration table corresponding to the maximum degree of adaptation in the degrees of adaptation as a template table of the demand matrix; and performing data adjustment on the table data in the template table by using fund flow data obtained from a fund party in advance, and performing financing configuration on the financing party according to the template table after data adjustment. According to the invention, the fund party data allocation efficiency can be improved.

Description

Technical Field

[0001] The present invention relates to the field of big data technology, and in particular to a funding party data allocation method, device and storage medium based on big data. Background Art

[0002] For different financing parties, the funding party must specify a funding data allocation plan that is suitable for the former, but different financing parties are quite different, and there are also obvious differences in funding data allocation plans at different times. For example, some companies may need short-term funding support to cope with sudden operational needs, while other companies may need long-term funds to support their development strategies. Therefore, timely adjustment and allocation of funding plans can better meet the actual needs of the financing party.

[0003] Some institutions may be slow to formulate funding plans due to technical and system limitations. For example, traditional manual operations and paper document processes may be time-consuming and unable to meet the needs of quickly formulating funding plans. Summary of the invention

[0004] The present invention provides a method, device and storage medium for allocating funding data based on big data, the main purpose of which is to solve the problem of low efficiency in allocating funding data.

[0005] To achieve the above purpose, the present invention provides a funding data allocation method based on big data, comprising:

[0006] Classify the financing needs obtained in advance from the financing party by stages to obtain the financing stage corresponding to the financing needs;

[0007] Constructing a sparse matrix of the financing party according to the financing stage, and filling the sparse matrix with financing data to obtain a demand matrix of the financing party;

[0008] Calculate the degree of fit between the demand matrix and the table data in the historical data matching table pre-acquired from the funding party, and select the historical data matching table corresponding to the maximum degree of fit among the degrees of fit as the template table of the demand matrix;

[0009] The table data in the template table is adjusted using the fund flow data obtained in advance from the funding party, and the financing party is allocated financing based on the template table after the data adjustment.

[0010] Optionally, the step of classifying the financing needs pre-acquired from the financing party to obtain the financing stage corresponding to the financing needs includes:

[0011] Use the word frequency-inverse document frequency algorithm to extract keywords from the financing needs obtained in advance from the financing party;

[0012] The keywords are converted into data vectors, and the financing needs are classified into stages using the data vectors and a pre-trained classification model to obtain the financing stage corresponding to the financing needs.

[0013] Optionally, the extracting keywords in the financing demand obtained in advance from the financing party by using a word frequency-inverse document frequency algorithm includes:

[0014] Performing text segmentation on the financing demand, and performing word frequency statistics on the financing demand after the text segmentation to obtain the word frequency of the financing demand;

[0015] Generate an inverse document frequency of each word in the financing demand using a preset corpus;

[0016] Generating distribution probabilities of words in the financing demand one by one according to the word frequency and the inverse document frequency;

[0017] The words in the financing demand are arranged in descending order according to the size of the distribution probability, the words in the descending order are selected using a preset number threshold, and the selected words are determined to be keywords of the financing demand.

[0018] Optionally, constructing the sparse matrix of the financing party according to the financing stage includes:

[0019] Determining the number of columns of the sparse matrix to be constructed according to the total number of stages of the financing stage;

[0020] Determine the number of rows of the sparse matrix to be constructed according to a preset industry field;

[0021] A sparse matrix of the financing party is constructed according to the number of rows and the number of columns.

[0022] Optionally, filling the sparse matrix with financing data to obtain the financing party's demand matrix includes:

[0023] Filling the sparse matrix with data according to the financing data, and weighting the matrix elements in the sparse matrix after the data filling according to the preset field weights of the industry field to obtain weighted data of each matrix element;

[0024] A demand matrix of the financing party is generated according to the weighted data.

[0025] Optionally, the calculating the degree of compatibility between the demand matrix and the table data in the historical data matching table pre-acquired from the funding party includes:

[0026] Standardizing the table data in the historical data allocation table to obtain standardized data;

[0027] The standardized data is numerically compared with the matrix elements in the demand matrix item by item, and the compatibility of the demand matrix with the table data in the historical data matching table pre-acquired from the funding party is generated according to the comparison result of the item by item numerical comparison.

[0028] Optionally, the adjusting the table data in the template table by using the capital flow data obtained in advance from the funding party includes:

[0029] Performing flow trend analysis on the capital flow data obtained in advance from the funding party to obtain the flow trend of the capital flow data;

[0030] Performing seasonal fluctuation analysis on the fund flow data obtained in advance from the funding party to obtain the seasonal fluctuation of the fund flow data;

[0031] The table data in the template table is adjusted according to the flow trend and the seasonal fluctuation.

[0032] Optionally, the performing flow trend analysis on the capital flow data obtained in advance from the funding party to obtain the flow trend of the capital flow data includes:

[0033] Determine the timestamp of the fund flow data pre-acquired from the funding party, and generate time series data of the fund flow data pre-acquired from the funding party according to the timestamp;

[0034] The time series data is subjected to moving average processing using a preset moving window, and the flow trend of the capital flow data is generated according to the time series data after the moving average processing.

[0035] In order to solve the above problem, the present invention further provides an electronic device, the electronic device comprising:

[0036] at least one processor; and,

[0037] a memory communicatively connected to the at least one processor; wherein,

[0038] The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the above-mentioned big data-based funding party data allocation method.

[0039] In order to solve the above problems, the present invention also provides a computer-readable storage medium, in which at least one computer program is stored. The at least one computer program is executed by a processor in an electronic device to implement the above-mentioned big data-based funding party data allocation method.

[0040] The embodiment of the present invention can accurately locate the current financing stage of the financing party by classifying the financing demand by stage, so as to match the historical data in a targeted manner, improve the accuracy and efficiency of matching, construct the information of the financing party into a sparse matrix, form a demand matrix by filling the financing data, and quickly screen out the most suitable historical data as a template by calculating the adaptability of the demand matrix and the table data in the historical data matching table of the funding party. By quantifying the matching degree, the matching efficiency is improved, the full amount traversal is avoided, and the calculation speed is improved. The historical data matching table with the highest adaptability is selected as the template, and the mode and structure of the historical data are used to guide the current demand, so that the data matching is faster and more accurate. The template table is adjusted by using the latest funding party capital flow data, so that it can be ensured that the matching result not only considers the historical data, but also reflects the current capital flow situation, enhances the real-time and dynamic adaptability, and the whole process is highly dependent on data drive and algorithm processing. The automated processing flow significantly improves the efficiency. Therefore, the funding party data matching method, device and storage medium based on big data proposed in the present invention can solve the problem of low efficiency when performing funding party data matching. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 A schematic diagram of a flow chart of a method for allocating funding data based on big data provided by an embodiment of the present invention;

[0042] Figure 2 A schematic diagram of the structure of an electronic device for implementing the big data-based funding party data allocation method provided in one embodiment of the present invention.

[0043] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0044] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0045] The embodiment of the present application provides a method for allocating data of a funding party based on big data. The execution subject of the method for allocating data of a funding party based on big data includes but is not limited to at least one of the electronic devices such as a server, a terminal, etc. that can be configured to execute the method provided by the embodiment of the present application. In other words, the method for allocating data of a funding party based on big data can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc. The server can be an independent server, or it can be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (Content Delivery Network, CDN), and big data and artificial intelligence platforms.

[0046] Reference Figure 1 FIG. 1 is a flow chart of a method for allocating data of a funding party based on big data provided by an embodiment of the present invention. In this embodiment, the method for allocating data of a funding party based on big data includes:

[0047] S1. Classify the financing needs obtained in advance from the financing party into stages to obtain the financing stage corresponding to the financing needs.

[0048] In the embodiment of the present invention, the stage classification of the financing demand obtained in advance from the financing party to obtain the financing stage corresponding to the financing demand includes:

[0049] The keyword frequency-inverse document frequency algorithm is used to extract the keywords of the financing demand obtained in advance from the financing party;

[0050] The keywords are converted into data vectors, and the financing needs are classified into stages using the data vectors and a pre-trained classification model to obtain the financing stage corresponding to the financing needs.

[0051] In an embodiment of the present invention, it is assumed that there is a start-up enterprise that specializes in providing an online agricultural product sales platform. They are seeking seed round financing, and their financing needs include expanding the supply chain network, improving logistics and warehousing systems, and improving procurement and inventory management efficiency.

[0052] Furthermore, the required information can be obtained legally and effectively through some public channels and professional financial data service platforms.

[0053] For example: using a third-party financial data platform, there are many professional financial data providers, such as Wind, Hang Seng Juyuan, etc. They collect and organize financial market data through legal channels, including information on the financing party.

[0054] For example: using open source data interface packages, among which Tushare is a well-known free, open source Python financial data interface package that provides financial data in various categories including stocks, fundamentals, macro, news, etc. Through Tushare, you can obtain the public financial data of certain financing parties under the premise of compliance.

[0055] For example: many financing parties will publish key financial information and business updates on their official websites. At the same time, listed companies are also required to publish regular financial reports and important announcements on the websites of the China Securities Regulatory Commission or exchanges. Although this information is scattered, it provides the public with a direct way to understand the financial status of a company.

[0056] For example: Various financial institutions and research institutions regularly publish in-depth analysis reports on specific industries or companies. These reports are usually based on public information and corporate surveys, and can provide in-depth market analysis and financial forecasts.

[0057] For example: Some academic institutions and professional databases provide access to research results in the fields of economics and finance, which often include detailed analysis of corporate financial data.

[0058] For example: government finance departments, central banks and financial regulatory agencies regularly publish financial market statistics and analysis reports, which may contain relevant data of the financing party.

[0059] To sum up, although the financial data of the financing party is confidential, relevant information can still be obtained through the above-mentioned public channels.

[0060] Furthermore, keywords are first extracted from the financing demand description provided by the financing party. In this example, the financing demand description includes the following keywords: supply chain network, logistics, warehousing and procurement. The word frequency-inverse document frequency algorithm can be used to identify and extract these keywords. The word frequency-inverse document frequency algorithm will evaluate the importance of keywords based on the frequency of the words in the document and the frequency in the entire corpus, so as to determine which words are the keywords that best describe the financing needs.

[0061] Furthermore, the extracted keywords are converted into numerical vectors. A pre-trained word embedding model, such as Word2Vec, can be used to map each keyword to a vector representation in a high-dimensional space.

[0062] For example, “expanding the supply chain network” is transformed into a 300-dimensional vector, “improving the logistics and warehousing system” is transformed into another 300-dimensional vector, and so on.

[0063] In detail, the converted keyword vector is input into a pre-trained classification model, which may be a support vector machine (SVM) model. After the model is trained, the financing stage to which the financing demand belongs can be predicted based on the input keyword vector.

[0064] Furthermore, based on the output of the model, determine the financing stage to which the financing demand belongs. For example, if the probability of "seed round financing" is the highest in the probability distribution output by the model, then the financing demand can be classified as the seed round financing stage.

[0065] In detail, the method of extracting keywords from the financing demand obtained in advance from the financing party by using the word frequency-inverse document frequency algorithm includes:

[0066] Performing text segmentation on the financing demand, and performing word frequency statistics on the financing demand after the text segmentation to obtain the word frequency of the financing demand;

[0067] Generate an inverse document frequency of each word in the financing demand using a preset corpus;

[0068] Generating distribution probabilities of words in the financing demand one by one according to the word frequency and the inverse document frequency;

[0069] The words in the financing demand are arranged in descending order according to the size of the distribution probability, the words in the descending order are selected using a preset number threshold, and the selected words are determined to be keywords of the financing demand.

[0070] In detail, first, the financing demand description is segmented to split the text into words or phrases. For example, the following results are obtained after word segmentation: “expand”, “supply chain”, “network”, “improve”, “logistics”, “and”, “warehousing”, “system”, “enhance”, “purchase”, “and”, “inventory”, “management”, and “efficiency”.

[0071] Furthermore, a word frequency statistics is performed on the financing demand after word segmentation: the frequency of each word in the text is calculated, among which the word frequency of “expansion” is 1, the word frequency of “supply chain” is 1, the word frequency of “network” is 1, the word frequency of “improvement” is 1, the word frequency of “logistics” is 1, the word frequency of “and” is 1, the word frequency of “warehousing” is 1, the word frequency of “system” is 1, the word frequency of “improvement” is 1, the word frequency of “procurement” is 1, the word frequency of “and” is 1, the word frequency of “inventory” is 1, the word frequency of “management” is 1, and the word frequency of “efficiency” is 1.

[0072] Furthermore, an inverse document frequency is generated using a preset corpus, wherein the inverse document frequency represents the logarithm of the inverse of the frequency of a word appearing in the corpus. In this example, assuming that there are a large number of documents in the corpus, the inverse document frequency of each word is obtained after calculation, and the word frequency and the inverse document frequency are multiplied to obtain the distribution probability of the word in the financing demand.

[0073] Furthermore, the words in the financing demand are arranged in descending order according to the size of the distribution probability, and the words in the descending order are selected using a preset number threshold. Determining that the selected words are keywords of the financing demand means arranging the words in descending order according to the size of the probability distribution, and then selecting the front words as keywords according to the preset number threshold. For example, the front words are: supply chain network, logistics, warehousing and procurement, wherein the preset number threshold is specified.

[0074] In detail, the method of extracting keywords from the financing demand obtained in advance from the financing party by using the word frequency-inverse document frequency algorithm includes:

[0075] Performing text segmentation on the financing demand, and performing word frequency statistics on the financing demand after the text segmentation to obtain the word frequency of the financing demand;

[0076] Generate an inverse document frequency of each word in the financing demand using a preset corpus;

[0077] Generate a word frequency-inverse document frequency matrix of the financing demand according to the word frequency and the inverse document frequency;

[0078] Establishing an association relationship between the word frequency-inverse document frequency matrix and each word in the financing demand;

[0079] The matrix elements in the word frequency-inverse document frequency matrix are filtered using a preset word frequency-inverse document frequency threshold, and the words in the financing demand are selected according to the matrix elements obtained after the element filtering and the association relationship, and the words obtained after the target selection are determined to be keywords of the financing demand.

[0080] In detail, the inverse document frequency represents the logarithm of the inverse frequency of a word appearing in a corpus. Assuming there is a corpus containing a large number of financing demand descriptions, the inverse document frequency of each word can be calculated.

[0081] Furthermore, generating a term frequency-inverse document frequency matrix according to the term frequency and the inverse document frequency means multiplying the term frequency and the inverse document frequency to obtain the term frequency-inverse document frequency value of each word, and then forming a matrix.

[0082] In detail, keyword screening according to a preset word frequency-inverse document frequency threshold means filtering elements in the word frequency-inverse document frequency matrix according to the preset word frequency-inverse document frequency threshold, and selecting words corresponding to elements above the threshold as keywords.

[0083] Furthermore, according to the keyword list obtained in the above steps, these words are determined to be keywords of financing needs.

[0084] S2. Construct a sparse matrix of the financing party according to the financing stage, fill the sparse matrix with financing data, and obtain a demand matrix of the financing party.

[0085] In the embodiment of the present invention, the step of constructing the sparse matrix of the financing party according to the financing stage includes:

[0086] Determining the number of columns of the sparse matrix to be constructed according to the total number of stages of the financing stage;

[0087] Determine the number of rows of the sparse matrix to be constructed according to a preset industry field;

[0088] A sparse matrix of the financing party is constructed according to the number of rows and the number of columns.

[0089] In detail, suppose there is a company that is in the process of financing, and the financing process is divided into three stages: the start-up stage, the development stage, and the mature stage. It is hoped that a sparse matrix of the company's supply chain indicators will be constructed based on these stages.

[0090] Furthermore, the number of columns of the sparse matrix to be constructed is determined according to the total number of stages of the financing stage. In this example, there are three stages, so the number of columns of the sparse matrix is ​​3.

[0091] In detail, the number of rows of the sparse matrix to be constructed is determined according to the preset industry field. Assuming that the industry fields of interest include manufacturing, retail and logistics, the number of rows of the sparse matrix is ​​3.

[0092] In detail, a sparse matrix of the financing parties is constructed according to the number of rows and columns, where each row of the matrix represents an industry sector and each column represents a financing stage, which will be filled to reflect the supply chain indicators of the company at different stages.

[0093] For example, a sparse matrix of the financing party may be constructed according to the number of rows and columns:

[0094] Industry Start-up phase Development Stage Mature stage manufacturing 0 0 0 Retail 0 0 0 Logistics 0 0 0

[0095] In the embodiment of the present invention, the step of filling the sparse matrix with financing data to obtain the financing party's demand matrix includes:

[0096] Filling the sparse matrix with data according to the financing data, and weighting the matrix elements in the sparse matrix after the data filling according to the preset field weights of the industry field to obtain weighted data of each matrix element;

[0097] A demand matrix of the financing party is generated according to the weighted data.

[0098] In detail, to fill the sparse matrix according to the financing data, first, the actual financing data needs to be used to fill the previously constructed sparse matrix. These data may be obtained from the company's internal system or external financing data providers to reflect the company's actual performance at different financing stages.

[0099] In detail, the populated data is weighted according to the preset industry field weights. In this step, the importance of different industry fields in the company's supply chain indicators needs to be taken into account. For example, manufacturing may be more critical to supply chain efficiency, so it may be given a higher weight when weighting. These weights may be set by professionals, industry standards or internal company decisions.

[0100] In detail, the demand matrix of the financing party is generated, and the demand matrix of the financing party can be generated by weighting the filled sparse matrix.

[0101] For example: Assume that there is actual financing data to fill this sparse matrix, and there are preset weights for industry fields, and then the data is weighted to generate a demand matrix.

[0102] Assume that the filled sparse matrix is ​​as follows:

[0103] Industry Start-up phase Development Stage Mature stage manufacturing 50 60 70 Retail 30 40 50 Logistics 20 25 30

[0104] Assume that there are the following preset weights for the industry sectors, where the weight of manufacturing is 0.4, the weight of retail is 0.3, and the weight of logistics is 0.3. Now perform a weighted calculation on each filled data to obtain the following demand matrix:

[0105] Industry Start-up phase Development Stage Mature stage manufacturing 50*0.4 60*0.4 70*0.4 Retail 30*0.3 40*0.3 50*0.3 Logistics 20*0.3 25*0.3 30*0.3

[0106] Furthermore, the demand matrix obtained after calculation is:

[0107] Industry Start-up phase Development Stage Mature stage manufacturing 20 24 28 Retail 9 12 15 Logistics 6 7.5 9

[0108] In detail, this demand matrix reflects the company's supply chain needs and indicators at different financing stages and industry fields, providing an important reference for corporate decision-making.

[0109] S3. Calculate the degree of compatibility between the demand matrix and the table data in the historical data matching table pre-acquired from the funding party, and select the historical data matching table corresponding to the maximum degree of compatibility as the template table of the demand matrix.

[0110] In the embodiment of the present invention, the calculation of the degree of compatibility between the demand matrix and the table data in the historical data matching table pre-acquired from the funding party includes:

[0111] Standardizing the table data in the historical data allocation table to obtain standardized data;

[0112] The standardized data is numerically compared with the matrix elements in the demand matrix item by item, and the compatibility of the demand matrix with the table data in the historical data matching table pre-acquired from the funding party is generated according to the comparison result of the item by item numerical comparison.

[0113] In detail, the tabular data in the historical data matching table are standardized to ensure that they have similar dimensions and distributions, wherein the standardization may adopt methods such as mean normalization or Z-score normalization.

[0114] In detail, comparing the standardized data with the matrix elements in the demand matrix item by item numerically refers to comparing the standardized historical data with the corresponding elements in the demand matrix item by item numerically, which can be achieved by calculating their differences or correlations. The purpose of the comparison is to evaluate the similarity between the data in the demand matrix and the historical data.

[0115] In detail, assuming that the historical data allocation table contains indicators such as the number of suppliers and logistics efficiency, these indicators need to be standardized to ensure that they have similar dimensions and distributions.

[0116] For example, for the number of suppliers, it can be normalized so that its value is between 0 and 1; for logistics efficiency, the Z-score normalization method can be used to transform it into a distribution with a mean of 0 and a standard deviation of 1.

[0117] In detail, for each supplier, its historical supplier quantity and logistics efficiency are compared with the corresponding matrix elements in the demand matrix.

[0118] In detail, based on the results of item-by-item numerical comparison, the degree of fit between the demand matrix and the tabular data in the historical data matching table is generated, where the degree of fit can be a score or an evaluation level used to indicate the degree of match between the demand matrix and the historical data. For example, a fit score can be given based on the size of the difference or the degree of correlation to evaluate the accuracy and reliability of the demand matrix.

[0119] For example, if the supplier quantity forecast in the demand matrix is ​​not much different from the supplier quantity in the historical data, and the logistics efficiency forecast is highly correlated with the logistics efficiency in the historical data, then it can be considered that the degree of fit between the demand matrix and the historical data matching table is high. Otherwise, the degree of fit is low.

[0120] In summary, through these steps, the compatibility of the demand matrix with the tabular data in the historical data allocation table obtained in advance from the funding party can be evaluated, thereby providing a reference for subsequent decision-making.

[0121] In the embodiment of the present invention, the selection of the historical data matching table corresponding to the maximum fitness degree among the fitness degrees as the template table of the demand matrix can be explained by the following example:

[0122] Assume that two historical data allocation tables are obtained in advance from the funding party, namely Table A and Table B, and their data are as follows:

[0123] Form A:

[0124] Industry Start-up phase Development Stage Mature stage manufacturing 18 22 25 Retail 8 11 14 Logistics 5 8 10

[0125] Form B:

[0126] Industry Start-up phase Development Stage Mature stage manufacturing 21 25 30 Retail 10 13 16 Logistics 7 9 11

[0127] Furthermore, it is necessary to calculate the degree of fit between the demand matrix and the two historical data matching tables. The Pearson correlation coefficient method can be used to calculate the differences or correlations between the data in various industry fields and development stages.

[0128] In detail, assuming that after calculation, the fit between Table A and the demand matrix is ​​0.85, and the fit between Table B is 0.78, according to the calculation results, the historical data matching table with the highest fit will be selected as the template table of the demand matrix, that is, Table A will be selected as the template. This means that the data in Table A is the most compatible with the demand matrix and can be used as a reference for future supply chain management decisions.

[0129] S4. Using the fund flow data obtained in advance from the funding party, the table data in the template table is adjusted, and financing is allocated to the financing party according to the template table after the data adjustment.

[0130] In the embodiment of the present invention, the step of adjusting the table data in the template table by using the capital flow data obtained in advance from the funding party includes:

[0131] Performing flow trend analysis on the capital flow data obtained in advance from the funding party to obtain the flow trend of the capital flow data;

[0132] Performing seasonal fluctuation analysis on the fund flow data obtained in advance from the funding party to obtain the seasonal fluctuation of the fund flow data;

[0133] The table data in the template table is adjusted according to the flow trend and the seasonal fluctuation.

[0134] In detail, assuming that the cash flow data of logistics cost and inventory cost is obtained from the funding party, the data in the template table can be adjusted by following the steps below:

[0135] First, conduct a flow trend analysis on the cash flow data of logistics costs and inventory costs. This includes calculating moving averages or other trend indicators to understand the overall change trend of the data. For example, it may be found that logistics costs have obvious fluctuations in a certain seasonal cycle, which may be due to increased logistics demand caused by holidays or seasonal sales activities.

[0136] Next, seasonal fluctuation analysis is performed on the cash flow data of logistics costs and inventory costs. This can be achieved by fitting seasonal models or calculating seasonal indicators. For example, it may be found that logistics costs have higher seasonal fluctuations in certain months of each year, while they are relatively stable in other months.

[0137] Finally, based on the results of flow trend analysis and seasonal fluctuation analysis, the logistics cost and inventory cost data in the template table can be adjusted. For flow trend analysis, the data can be adjusted to reflect changes in the overall trend, such as increasing the budget for logistics costs during seasonal demand peaks; for seasonal fluctuation analysis, the data can be adjusted for specific months or seasonal events to reflect expected seasonal fluctuations.

[0138] For example: There is a template table that records the monthly logistics cost and inventory cost data for the past few years. By analyzing these data, the following trends and seasonal fluctuations are found: logistics costs show an increasing trend year by year; in December of each year, logistics costs usually increase significantly due to Christmas sales activities.

[0139] Furthermore, consider the growth trend of logistics costs in the past few years and adjust the logistics cost budget accordingly in the template form. For example, if the logistics cost increased by 10% last year, this year's budget will be increased by 10% accordingly to reflect the changes in long-term trends.

[0140] Furthermore, in response to seasonal fluctuations in December, the logistics cost budget for that month can be adjusted in the template table. Assuming that the logistics cost in December last year was twice that of other months, and sales activities are expected to be more intense this year, the logistics cost budget for December this year may be increased to three times that of last year.

[0141] In this way, by adjusting the data according to flow trend analysis and seasonal fluctuation analysis, the impact of long-term trends and seasonal fluctuations on logistics costs and inventory costs can be more accurately reflected, thereby formulating more reasonable budgets and decisions.

[0142] In detail, the flow trend analysis of the fund flow data obtained in advance from the funding party to obtain the flow trend of the fund flow data includes:

[0143] Determine the timestamp of the fund flow data pre-acquired from the funding party, and generate time series data of the fund flow data pre-acquired from the funding party according to the timestamp;

[0144] The time series data is subjected to moving average processing using a preset moving window, and the flow trend of the capital flow data is generated according to the time series data after the moving average processing.

[0145] In detail, timestamps are determined from the fund flow data obtained from the funding party, and these timestamps may represent the time point of each transaction or data record. Then, these timestamps and the corresponding fund flow data are combined into time series data, in which each timestamp corresponds to a fund flow data point.

[0146] Furthermore, in order to smooth the data and capture its trend, a preset moving window is used to perform moving average processing on the time series data, where the moving average is a method of calculating the average of a series of consecutive data points, where each average is calculated within a window of a fixed size. For example, if the window size is selected as 3, for the nth data point, the moving average will calculate the average of the three data points n-2, n-1 and n as the smoothed value of the nth data point.

[0147] In detail, by analyzing the time series data after moving average processing, the flow trend of the capital flow data can be obtained, which may include identifying whether the overall trend of the data is rising, falling or remaining stable, as well as possible cyclical or seasonal fluctuations.

[0148] Furthermore, by observing the changing trend of the moving average, we can have a clearer understanding of the overall trend of capital flow data, thereby providing more valuable information for subsequent data adjustments.

[0149] In summary, this process uses time series analysis techniques, especially moving average processing, to reveal trends and patterns in capital flow data and provide guidance for subsequent data adjustments.

[0150] In detail, the seasonal fluctuation analysis of the fund flow data obtained in advance from the funding party to obtain the seasonal fluctuation of the fund flow data includes:

[0151] Performing local regression fitting on the capital flow data obtained in advance from the funding party to obtain the trend component of the capital flow data;

[0152] Perform trend elimination on the fund flow data according to the trend component, and calculate the average deviation between each time point of the fund flow data after trend elimination and the same time point in a preset season;

[0153] The seasonal fluctuation of the fund flow data is generated according to the mean deviation.

[0154] In detail, local regression fitting is performed on the capital flow data obtained in advance from the funding party, wherein local regression is a method for fitting data trends, which is different from global regression in that it fits a local model around each data point.

[0155] In detail, select an appropriate regression model and use the local regression method to fit the capital flow data to obtain the trend component of the data. This trend component represents the trend of medium- and long-term changes in the data. Among them, the regression model can choose linear regression, polynomial regression or local weighted regression. The specific choice needs to consider the characteristics of the data and the purpose of fitting.

[0156] For example: If the cash flow data shows a linear trend, you can choose a linear regression model. The linear regression model assumes that there is a linear relationship between the dependent variable (cash flow data) and the independent variable (time), and describes the trend of the data by fitting a straight line. This model is simple and intuitive, easy to understand, and works well in some cases.

[0157] Furthermore, the fund flow data is detrended using the trend component, which means subtracting the trend component from the original data in order to better analyze the seasonal fluctuations of the data. Detrending can be achieved by simply subtracting the trend component from the original data.

[0158] In detail, after trend removal, the average deviation of each time point of the fund flow data from the same time point in the preset season is calculated, which means comparing the difference between the data value at each time point and the average value of all time points in the same season. By calculating these deviations, the fluctuation of the fund flow data in each seasonal cycle, that is, seasonal fluctuation, can be understood.

[0159] In detail, by analyzing the calculated average deviation, seasonal fluctuations in the fund flow data can be generated. These fluctuations represent the changing patterns of the data in different seasonal cycles. The analysis of seasonal fluctuations can help understand the typical behavior of fund flow data in a specific time period, so as to make corresponding decisions or adjustments.

[0160] In detail, by analyzing the calculated average deviation, the seasonal fluctuation of capital flow data can be generated. This means that the calculated average deviation is normalized to obtain the seasonal coefficient, which can usually be achieved by dividing the deviation by the average value or standard deviation within the seasonal cycle, and then multiplying the normalized seasonal coefficient by the value of the corresponding time point in the trend component to obtain an estimate of the seasonal fluctuation. In this way, the seasonal fluctuation can be superimposed on the original data while retaining the trend.

[0161] In summary, this process reveals the seasonal fluctuations in fund flow data through local regression fitting, trend removal, and mean deviation calculation, thereby helping to better understand the cyclical characteristics and changing patterns of the data.

[0162] In an embodiment of the present invention, allocating financing to the financing party based on the template form adjusted by the data refers to using a series of data adjusted by flow trend analysis and seasonal fluctuation analysis, wherein the series of data can be logistics cost and inventory cost data, to determine the financing needs and capital allocation plan of the financing party.

[0163] Specifically, by analyzing and adjusting the logistics cost and inventory cost data, it is possible to more accurately predict the financing party's capital needs in different time periods. For example, after considering the annual growth trend of logistics costs, it can be predicted that the financing party may need more funds in the next few years to cope with the cost growth. After considering the seasonal fluctuations in logistics costs in December each year due to Christmas sales activities, it can be predicted that the financing party may need additional funds in December each year to support sales activities.

[0164] Therefore, based on these adjusted data, the financing party's financing allocation can more accurately match its actual funding needs, ensuring sufficient funding support during critical periods while avoiding excessive or insufficient financing. In this way, the adjusted template form can serve as an important reference for the financing party's financing decision-making, helping them to formulate more reasonable funding strategies and decisions.

[0165] The embodiment of the present invention can accurately locate the current financing stage of the financing party by classifying the financing demand by stage, so as to match the historical data in a targeted manner, improve the accuracy and efficiency of matching, construct the information of the financing party into a sparse matrix, form a demand matrix by filling the financing data, and quickly screen out the most suitable historical data as a template by calculating the adaptability of the demand matrix and the table data in the historical data matching table of the funding party. By quantifying the matching degree, the matching efficiency is improved, the full amount traversal is avoided, and the calculation speed is improved. The historical data matching table with the highest adaptability is selected as the template, and the mode and structure of the historical data are used to guide the current demand, so that data matching is faster and more accurate. The template table is adjusted by using the latest funding party capital flow data, so that it can be ensured that the matching result not only considers the historical data, but also reflects the current capital flow situation, enhances the real-time and dynamic adaptability, and the whole process is highly dependent on data-driven and algorithm processing. The automated processing flow significantly improves the efficiency. Therefore, the funding party data matching method based on big data proposed in the present invention can solve the problem of low efficiency when performing funding party data matching.

[0166] like Figure 2 1 is a schematic diagram of the structure of an electronic device for implementing a funding party data allocation method based on big data provided by an embodiment of the present invention.

[0167] The electronic device 1 may include a processor 10, a memory 11, a communication bus 12 and a communication interface 13, and may also include a computer program stored in the memory 11 and executable on the processor 10, such as a funding data allocation program based on big data.

[0168] In some embodiments, the processor 10 may be composed of an integrated circuit, for example, a single packaged integrated circuit, or a plurality of integrated circuits with the same or different functions, including one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and combinations of various control chips. The processor 10 is the control core (Control Unit) of the electronic device, and uses various interfaces and lines to connect the various components of the entire electronic device, and executes or executes the programs or modules stored in the memory 11 (for example, executing the funding data allocation program based on big data, etc.), and calls the data stored in the memory 11 to execute various functions of the electronic device and process data.

[0169] The memory 11 includes at least one type of readable storage medium, and the readable storage medium includes flash memory, mobile hard disk, multimedia card, card-type memory (for example: SD or DX memory, etc.), magnetic memory, disk, optical disk, etc. In some embodiments, the memory 11 may be an internal storage unit of an electronic device, such as a mobile hard disk of the electronic device. In other embodiments, the memory 11 may also be an external storage device of an electronic device, such as a plug-in mobile hard disk, a smart memory card (SmartMediaCard, SMC), a secure digital (SecureDigital, SD) card, a flash card (FlashCard), etc. equipped on the electronic device. Further, the memory 11 may also include both an internal storage unit of the electronic device and an external storage device. The memory 11 can not only be used to store application software and various types of data installed in the electronic device, such as the code of the funding party data allocation program based on big data, but also can be used to temporarily store data that has been output or is to be output.

[0170] The communication bus 12 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. The bus is configured to realize connection and communication between the memory 11 and at least one processor 10, etc.

[0171] The communication interface 13 is used for communication between the above-mentioned electronic device and other devices, including a network interface and a user interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a WI-FI interface, a Bluetooth interface, etc.), which is generally used to establish a communication connection between the electronic device and other electronic devices. The user interface may be a display (Display), an input unit (such as a keyboard (Keyboard)), and optionally, the user interface may also be a standard wired interface, a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, and an OLED (Organic Light-Emitting Diode, organic light-emitting diode) touch device, etc. Among them, the display may also be appropriately referred to as a display screen or a display unit, which is used to display information processed in the electronic device and to display a visual user interface.

[0172] The figure only shows an electronic device with components. Those skilled in the art will understand that the structure shown in the figure does not constitute a limitation on the electronic device, and may include fewer or more components than shown in the figure, or combine certain components, or arrange the components differently.

[0173] For example, although not shown, the electronic device may also include a power source (such as a battery) for supplying power to various components. Preferably, the power source may be logically connected to the at least one processor 10 through a power management device, so that the power management device can realize functions such as charging management, discharging management, and power consumption management. The power source may also include any components such as one or more DC or AC power sources, recharging devices, power failure detection circuits, power converters or inverters, power status indicators, etc. The electronic device may also include a variety of sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be repeated here.

[0174] It should be understood that the embodiment is for illustration only and the scope of the patent application is not limited to this structure.

[0175] The funding party data allocation program based on big data stored in the memory 11 of the electronic device 1 is a combination of multiple instructions. When running in the processor 10, it can achieve:

[0176] Classify the financing needs obtained in advance from the financing party by stages to obtain the financing stage corresponding to the financing needs;

[0177] Constructing a sparse matrix of the financing party according to the financing stage, and filling the sparse matrix with financing data to obtain a demand matrix of the financing party;

[0178] Calculate the degree of fit between the demand matrix and the table data in the historical data matching table pre-acquired from the funding party, and select the historical data matching table corresponding to the maximum degree of fit among the degrees of fit as the template table of the demand matrix;

[0179] The table data in the template table is adjusted using the fund flow data obtained in advance from the funding party, and the financing party is allocated financing based on the template table after the data adjustment.

[0180] Specifically, the specific implementation method of the processor 10 for the above instructions can refer to the description of the relevant steps in the corresponding embodiment of the accompanying drawings, which will not be repeated here.

[0181] Furthermore, if the module / unit integrated in the electronic device 1 is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. The computer-readable storage medium can be volatile or non-volatile. For example, the computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a mobile hard disk, a magnetic disk, an optical disk, a computer memory, and a read-only memory (ROM).

[0182] The present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor of an electronic device, the computer program can implement:

[0183] Classify the financing needs obtained in advance from the financing party by stages to obtain the financing stage corresponding to the financing needs;

[0184] Constructing a sparse matrix of the financing party according to the financing stage, and filling the sparse matrix with financing data to obtain a demand matrix of the financing party;

[0185] Calculate the degree of fit between the demand matrix and the table data in the historical data matching table pre-acquired from the funding party, and select the historical data matching table corresponding to the maximum degree of fit among the degrees of fit as the template table of the demand matrix;

[0186] The table data in the template table is adjusted using the fund flow data obtained in advance from the funding party, and the financing party is allocated financing based on the template table after the data adjustment.

[0187] In several embodiments provided by the present invention, it should be understood that the disclosed devices and methods can be implemented in other ways.

[0188] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0189] In addition, each functional module in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0190] It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0191] The blockchain referred to in this invention is a new application mode of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanism, encryption algorithm, etc. Blockchain is essentially a decentralized database, a string of data blocks generated by cryptographic methods. Each data block contains a batch of network transaction information, which is used to verify the validity of its information (anti-counterfeiting) and generate the next block. Blockchain can include the underlying blockchain platform, platform product service layer, and application service layer.

[0192] The embodiments of the present application can acquire and process relevant data based on big data technology. Among them, artificial intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0193] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A funding data allocation method based on big data, characterized in that: The method comprises: Classify the financing needs obtained in advance from the financing party by stages to obtain the financing stage corresponding to the financing needs; Constructing a sparse matrix of the financing party according to the financing stage, and filling the sparse matrix with financing data to obtain a demand matrix of the financing party; Calculate the degree of fit between the demand matrix and the table data in the historical data matching table pre-acquired from the funding party, and select the historical data matching table corresponding to the maximum degree of fit among the degrees of fit as the template table of the demand matrix; The table data in the template table is adjusted using the fund flow data obtained in advance from the funding party, and the financing party is allocated financing based on the template table after the data adjustment.

2. The method for allocating funding data based on big data as claimed in claim 1, characterized in that: The step of classifying the financing needs obtained in advance from the financing party to obtain the financing stage corresponding to the financing needs includes: Use the word frequency-inverse document frequency algorithm to extract keywords from the financing needs obtained in advance from the financing party; The keywords are converted into data vectors, and the financing needs are classified into stages using the data vectors and a pre-trained classification model to obtain the financing stage corresponding to the financing needs.

3. The method for allocating funding data based on big data as claimed in claim 2, characterized in that: The method of extracting keywords from the financing demand obtained in advance from the financing party by using the word frequency-inverse document frequency algorithm includes: Performing text segmentation on the financing demand, and performing word frequency statistics on the financing demand after the text segmentation to obtain the word frequency of the financing demand; Generate an inverse document frequency of each word in the financing demand using a preset corpus; Generating distribution probabilities of words in the financing demand one by one according to the word frequency and the inverse document frequency; The words in the financing demand are arranged in descending order according to the size of the distribution probability, the words in the descending order are selected using a preset number threshold, and the selected words are determined to be keywords of the financing demand.

4. The method for allocating funding data based on big data as claimed in claim 1, characterized in that: The constructing the sparse matrix of the financing party according to the financing stage includes: Determining the number of columns of the sparse matrix to be constructed according to the total number of stages of the financing stage; Determine the number of rows of the sparse matrix to be constructed according to a preset industry field; A sparse matrix of the financing party is constructed according to the number of rows and the number of columns.

5. The method for allocating funding data based on big data as claimed in claim 1, characterized in that: The step of filling the sparse matrix with financing data to obtain the financing party's demand matrix includes: Filling the sparse matrix with data according to the financing data, and weighting the matrix elements in the sparse matrix after the data filling according to the preset field weights of the industry field to obtain weighted data of each matrix element; A demand matrix of the financing party is generated according to the weighted data.

6. The method for allocating funding data based on big data as claimed in claim 1, characterized in that: The calculating the adaptability of the demand matrix and the table data in the historical data matching table pre-acquired from the funding party includes: Standardizing the table data in the historical data allocation table to obtain standardized data; The standardized data is numerically compared with the matrix elements in the demand matrix item by item, and the compatibility between the demand matrix and the table data in the historical data matching table pre-acquired from the funding party is generated according to the comparison result of the item by item numerical comparison.

7. The method for allocating funding data based on big data as claimed in claim 1, characterized in that: The step of adjusting the table data in the template table by using the capital flow data obtained in advance from the funding party includes: Performing flow trend analysis on the capital flow data obtained in advance from the funding party to obtain the flow trend of the capital flow data; Performing seasonal fluctuation analysis on the fund flow data obtained in advance from the funding party to obtain the seasonal fluctuation of the fund flow data; The table data in the template table is adjusted according to the flow trend and the seasonal fluctuation.

8. The method for allocating funding data based on big data as claimed in claim 7, characterized in that: The flow trend analysis of the capital flow data obtained in advance from the funding party to obtain the flow trend of the capital flow data includes: Determine the timestamp of the fund flow data pre-acquired from the funding party, and generate time series data of the fund flow data pre-acquired from the funding party according to the timestamp; The time series data is subjected to moving average processing using a preset moving window, and the flow trend of the capital flow data is generated according to the time series data after the moving average processing.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the big data-based funding party data allocation method as described in any one of claims 1 to 8.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, it implements the big data-based funding data allocation method as described in any one of claims 1 to 8.

Images