Enterprise cost optimization method and system based on big data

Abstract

The invention provides an enterprise cost optimization method based on big data, and belongs to the technical field of data processing, and the method comprises the steps: constructing a model according with the features of a target enterprise through historical data, screening out a matched enterprise similar to the features of the target enterprise through the big data, and carrying out the optimization of the data of the target enterprise through the data of the matched enterprise, therefore, the cost of the target enterprise is closer to that of different enterprises in the same industry. The invention provides an enterprise cost optimization system based on big data and a related device.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a method and system for enterprise cost optimization based on big data. Background Art

[0002] As businesses expand and market competition intensifies, cost optimization has become a key means of enhancing their competitiveness. Traditional cost optimization methods typically rely on manual experience or simple statistical analysis, making them difficult to cope with complex market environments and the multi-dimensional data within enterprises.

[0003] With the development of technologies such as the Internet of Things (IoT), sensor technology, and cloud computing, enterprises are able to collect and store massive amounts of operational data in real time, including production costs, supply chain data, and sales figures. Distributed computing frameworks (such as Hadoop and Spark) and efficient data processing algorithms enable enterprises to rapidly process and analyze large amounts of data. Big data encompasses not only structured data (such as financial statements) but also unstructured data (such as text, images, and video), providing enterprises with a more comprehensive analytical perspective.

[0004] In the related art, in the process of optimizing enterprises based on big data, the optimization process only uses simple data and cannot combine the relevant characteristics of the enterprise, so the accuracy is not high. Summary of the Invention

[0005] The embodiments of the present application provide a method and system for enterprise cost optimization based on big data to improve the above problems.

[0006] To achieve the above objectives, this application adopts the following technical solutions:

[0007] In a first aspect, the present application provides an enterprise cost optimization method based on big data, the method comprising:

[0008] Identify the target enterprise and obtain the historical data of the target enterprise, where the historical data includes historical cost data, historical net profit data, and historical turnover data;

[0009] Aligning the timestamps of the historical cost data, the historical net profit data, and the historical turnover data, and generating multiple training sets, wherein each training set includes a historical cost data, a historical net profit data, and a historical turnover data;

[0010] Import multiple training sets into the neural network model, which is used to iteratively calculate historical cost data, historical net profit data, and historical turnover data, and output a matching index;

[0011] Obtaining the target enterprise's current cost data, current net profit data, and current turnover data, and determining a target matching index based on the current cost data, current net profit data, and current turnover data;

[0012] Based on big data, multiple matching historical data of multiple companies in the same industry are obtained, where the matching historical data includes matching historical cost data, matching historical net profit data, and matching historical turnover data. The multiple matching historical data are input into a neural network model, and multiple predicted matching indexes are obtained based on the output results of the neural network model.

[0013] Comparing the plurality of predicted matching indices with the target matching indices, and determining at least three matching indices closest to the target matching indices from the plurality of predicted matching indices as reference matching indices;

[0014] Obtain matching historical data corresponding to the reference matching index and determine the reference cost ratio of the target enterprise, wherein the reference cost ratio is the ratio of the matching historical cost data to the matching historical net profit data in the corresponding matching historical data;

[0015] Adjusts current cost data based on a reference cost scale.

[0016] In conjunction with the first aspect, in some implementations, the timestamps of the historical cost data, the historical net profit data, and the historical turnover data are aligned, and multiple training sets are generated, where each training set includes one historical cost data, one historical net profit data, and one historical turnover data, including:

[0017] The historical data is divided into training set, validation set and test set, where the total number of training sets is the same as the total number of validation sets and test sets.

[0018] In conjunction with the first aspect, in some embodiments, multiple training sets are imported into a neural network model, and the neural network model is used to iteratively calculate historical cost data, historical net profit data, and historical turnover data, and output a matching index, including:

[0019] Normalize historical cost data, historical net profit data, and historical turnover data so that the data ranges are within the same range;

[0020] The historical cost data, historical net profit data and historical turnover data are used as three different dimensions, which constitute the characteristic vector of the matching index;

[0021] Iteratively calculate historical cost data, historical net profit data, and historical turnover data, and stop calculating when the conditions are met and output the trained neural network model.

[0022] In conjunction with the first aspect, in some embodiments, historical cost data, historical net profit data, and historical turnover data are iteratively calculated, and the calculation is stopped and the trained neural network model is output when a condition is met, including:

[0023] Match the results calculated at each iteration with the data in the validation set and determine the difference value;

[0024] When the difference value is less than the preset value, the iterative calculation is stopped and the trained neural network model is output.

[0025] In conjunction with the first aspect, in some embodiments, matching the result of each iterative calculation with the data in the validation set and determining the difference value includes:

[0026] Get the result of each iterative calculation and compare it with the corresponding data in the validation set. If the result of each iterative calculation is greater than the corresponding data in the validation set, the difference value is:

[0027] T=(x1-y1)x1

[0028] Among them, x1 is the result of each iterative calculation, and y1 is the corresponding data in the validation set.

[0029] In conjunction with the first aspect, in some embodiments, the result of each iterative calculation is obtained and compared with the corresponding data in the validation set. If the result of each iterative calculation is less than the corresponding data in the validation set, the difference value is:

[0030] T=(y1-x1)x1

[0031] Among them, x1 is the result of each iterative calculation, and y1 is the corresponding data in the validation set.

[0032] In conjunction with the first aspect, in some embodiments, comparing multiple predicted matching indices with a target matching indices, and determining at least three matching indices closest to the target matching indices from the multiple predicted matching indices as reference matching indices includes:

[0033] Sort the matching indexes of multiple predictions by their size, and remove the maximum and minimum matching indexes among the multiple predictions;

[0034] The remaining multiple matching indices are matched with the target matching index, and the three closest matching indices are determined and used as reference matching indices.

[0035] In conjunction with the first aspect, in some embodiments, obtaining matching historical data corresponding to a reference matching index and determining a reference cost ratio of the target enterprise, wherein the reference cost ratio is the ratio of matching historical cost data to matching historical net profit data in the corresponding matching historical data, includes:

[0036] Obtain matching historical data corresponding to a reference matching index, and calculate a proportional relationship between matching historical cost data and matching historical net profit data in a set of matching historical data corresponding to the reference matching index;

[0037] Get the average of multiple sets of proportional relationships and use it as the reference cost ratio.

[0038] In a second aspect, the present application provides an enterprise cost optimization system based on big data, which is configured as follows:

[0039] Identify the target enterprise and obtain the historical data of the target enterprise, where the historical data includes historical cost data, historical net profit data, and historical turnover data;

[0040] Aligning the timestamps of the historical cost data, the historical net profit data, and the historical turnover data, and generating multiple training sets, wherein each training set includes a historical cost data, a historical net profit data, and a historical turnover data;

[0041] Import multiple training sets into the neural network model, which is used to iteratively calculate historical cost data, historical net profit data, and historical turnover data, and output a matching index;

[0042] Obtaining the target enterprise's current cost data, current net profit data, and current turnover data, and determining a target matching index based on the current cost data, current net profit data, and current turnover data;

[0043] Based on big data, multiple matching historical data of multiple companies in the same industry are obtained, where the matching historical data includes matching historical cost data, matching historical net profit data, and matching historical turnover data. The multiple matching historical data are input into a neural network model, and multiple predicted matching indexes are obtained based on the output results of the neural network model.

[0044] Comparing the plurality of predicted matching indices with the target matching indices, and determining at least three matching indices closest to the target matching indices from the plurality of predicted matching indices as reference matching indices;

[0045] Obtain matching historical data corresponding to the reference matching index and determine the reference cost ratio of the target enterprise, wherein the reference cost ratio is the ratio of the matching historical cost data to the matching historical net profit data in the corresponding matching historical data;

[0046] Adjusts current cost data based on a reference cost scale.

[0047] In conjunction with the second aspect, in some embodiments, the system is configured to:

[0048] Align the timestamps of historical cost data, historical net profit data, and historical turnover data, and generate multiple training sets, where each training set includes a historical cost data, a historical net profit data, and a historical turnover data, including:

[0049] The historical data is divided into training set, validation set and test set, where the total number of training sets is the same as the total number of validation sets and test sets.

[0050] In conjunction with the second aspect, in some embodiments, the system is configured to:

[0051] Multiple training sets are imported into the neural network model. The neural network model is used to iteratively calculate historical cost data, historical net profit data, and historical turnover data, and output matching indexes, including:

[0052] Normalize historical cost data, historical net profit data, and historical turnover data so that the data ranges are within the same range;

[0053] The historical cost data, historical net profit data and historical turnover data are used as three different dimensions, which constitute the characteristic vector of the matching index;

[0054] Iteratively calculate historical cost data, historical net profit data, and historical turnover data, and stop calculating when the conditions are met and output the trained neural network model.

[0055] In conjunction with the second aspect, in some embodiments, the system is configured to:

[0056] When the conditions are met, stop the calculation and output the trained neural network model, including:

[0057] Match the results calculated at each iteration with the data in the validation set and determine the difference value;

[0058] When the difference value is less than the preset value, the iterative calculation is stopped and the trained neural network model is output.

[0059] In conjunction with the second aspect, in some embodiments, the system is configured to:

[0060] The results calculated at each iteration are matched against the data in the validation set and the difference values are determined, including:

[0061] Get the result of each iterative calculation and compare it with the corresponding data in the validation set. If the result of each iterative calculation is greater than the corresponding data in the validation set, the difference value is:

[0062] T=(x1-y1)x1

[0063] Among them, x1 is the result of each iterative calculation, and y1 is the corresponding data in the validation set.

[0064] In conjunction with the second aspect, in some embodiments, the system is configured to:

[0065] Get the result of each iterative calculation and compare it with the corresponding data in the validation set. If the result of each iterative calculation is less than the corresponding data in the validation set, the difference value is:

[0066] T=(y1-x1)x1

[0067] Among them, x1 is the result of each iterative calculation, and y1 is the corresponding data in the validation set.

[0068] In conjunction with the second aspect, in some embodiments, the system is configured to:

[0069] Comparing the multiple predicted matching indices with the target matching indices, determining at least three matching indices closest to the target matching indices from the multiple predicted matching indices and using them as reference matching indices, including:

[0070] Sort the matching indexes of multiple predictions by their size, and remove the maximum and minimum matching indexes among the multiple predictions;

[0071] The remaining multiple matching indices are matched with the target matching index, and the three closest matching indices are determined and used as reference matching indices.

[0072] In conjunction with the second aspect, in some embodiments, the system is configured to:

[0073] Obtain matching historical data corresponding to the reference matching index and determine the reference cost ratio of the target enterprise, where the reference cost ratio is the ratio of the matching historical cost data to the matching historical net profit data in the corresponding matching historical data, including:

[0074] Obtain matching historical data corresponding to a reference matching index, and calculate a proportional relationship between matching historical cost data and matching historical net profit data in a set of matching historical data corresponding to the reference matching index;

[0075] Get the average of multiple sets of proportional relationships and use it as the reference cost ratio.

[0076] According to a third aspect of an embodiment of the present invention, an electronic device is provided, including:

[0077] At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the method proposed in the first aspect of the embodiment of the present invention.

[0078] A fourth aspect of an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method provided in the first aspect of the embodiment of the present invention.

[0079] In summary, the above method and system have the following technical effects:

[0080] The embodiments of the present application propose a big data-based enterprise cost optimization method and system. The method aligns the timestamps of historical cost data, historical net profit data, and historical turnover data, and then imports multiple training sets into a neural network model. The neural network model is used to iteratively calculate the historical cost data, historical net profit data, and historical turnover data and output a matching index. Then, the current cost data, current net profit data, and current turnover data of a target enterprise are obtained, and a target matching index is determined based on the current cost data, current net profit data, and current turnover data. Then, based on the big data, multiple matching historical data of multiple enterprises in the same industry are obtained. The matching historical data includes matching historical cost data, matching historical net profit data, and matching historical turnover data. The multiple matching historical data are input into the neural network model, and multiple predicted matching indices are obtained based on the output results of the neural network model. Then, the multiple predicted matching indices are compared with the target matching indices. At least three matching indices closest to the target matching indices are determined from the multiple predicted matching indices and used as reference matching indices. Then, matching historical data corresponding to the reference matching indices are obtained, a reference cost ratio of the target enterprise is determined, and finally, the current cost data is adjusted based on the reference cost ratio. The embodiment of the present application proposes a method and system for enterprise cost optimization based on big data, which constructs a model that conforms to the characteristics of the target enterprise through historical data, uses big data to screen out matching enterprises with similar characteristics to the target enterprise, and uses the data of the matching enterprises to optimize the data of the target enterprise, so that the cost of the target enterprise is closer to that of different enterprises in the same industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0081] Figure 1 This is a flowchart of a big data-based enterprise cost optimization method proposed in this application. DETAILED DESCRIPTION

[0082] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0083] This application proposes a method for enterprise cost optimization based on big data, which includes the following steps:

[0084] S101: Determine a target enterprise and obtain historical data of the target enterprise, wherein the historical data includes historical cost data, historical net profit data, and historical turnover data.

[0085] Understandably, the first step is to identify and determine the target company, which forms the foundation for subsequent analysis. Once the target company is selected, the next step is to collect and organize its historical financial data. This data is crucial for understanding the company's operating conditions and financial health. Specifically, the historical data to be obtained includes but is not limited to the following: historical cost data, historical net profit data, and historical turnover data. Historical cost data can help understand the total costs incurred by a company to produce goods or provide services over different time periods; historical net profit data reflects the company's net income after deducting all costs and expenses; and historical turnover data shows the company's total sales over various periods and is a key indicator of a company's market performance.

[0086] S102: Align the timestamps of the historical cost data, the historical net profit data, and the historical turnover data, and generate multiple training sets, wherein each training set includes a historical cost data, a historical net profit data, and a historical turnover data.

[0087] As you can understand, effective data analysis and predictive model construction first require the organization and preprocessing of historical data. Specifically, historical cost data, historical net profit data, and historical turnover data are precisely aligned according to their respective timestamps. This process ensures the temporal consistency of the data and provides an accurate foundation for subsequent analysis. Once alignment is complete, the next step is to generate multiple training sets, each containing a specific set of historical cost data, a corresponding set of historical net profit data, and a corresponding set of historical turnover data. This approach creates a series of data samples that will be used to train and validate the predictive model.

[0088] As an implementation method, the historical data may also be divided into a training set, a validation set, and a test set, wherein the total number of the training set is the same as the total number of the validation set and the test set.

[0089] It is understandable that when processing and analyzing historical data, an important step is to reasonably divide this data into different parts to facilitate subsequent model training and evaluation. Specifically, we will divide these historical data into three main sets: training set, validation set, and test set. The role of the training set is to allow the machine learning model to learn the patterns and characteristics in the data through this part of the data; the validation set is used to adjust the model parameters during the model training process to prevent overfitting and help select the best model structure; finally, the test set is used to evaluate the generalization ability of the model after the model training is completed, that is, the performance of the model on unseen data. It is worth noting that in this division process, the total number of samples in the training set is consistent with the total number of samples in the validation set and test set, so as to ensure that each part is representative, making the training and evaluation of the model more accurate and effective.

[0090] S103: Importing multiple training sets into the neural network model, the neural network model is used to iteratively calculate the historical cost data, historical net profit data and historical turnover data, and output a matching index.

[0091] Exemplarily, the historical cost data, historical net profit data and historical turnover data are normalized so that the data range is within the same range. Then, the historical cost data, historical net profit data and historical turnover data are used as three different dimensions, and the three dimensions constitute the characteristic vector of the matching index. Finally, the historical cost data, historical net profit data and one historical turnover data are iteratively calculated, and the calculation is stopped when the conditions are met and the trained neural network model is output.

[0092] It can be understood that the result of each iterative calculation can be matched with the data in the validation set, and the difference value can be determined. When the difference value is less than a preset value, the iterative calculation is stopped and the trained neural network model is output.

[0093] As you can understand, we need to normalize the historical cost data, historical net profit data, and historical turnover data to bring them into a common range. This ensures comparability of data of different magnitudes and dimensions in subsequent analysis. Next, the normalized historical cost data, historical net profit data, and historical turnover data are used as three independent dimensions, which together form the feature vector used to calculate the matching index. A feature vector is a key concept in machine learning, representing the characteristics of data in mathematical form. Finally, we will perform iterative calculations using the historical cost data, historical net profit data, and one historical turnover data point in these feature vectors. During this iterative process, we will continuously adjust the parameters of the neural network model until a preset stopping condition is met, such as reaching a certain number of iterations, reaching a certain error threshold, or no longer significantly improving model performance. Once these conditions are met, the calculation stops, and the trained neural network model is output. This model can be used to predict future costs, net profit, and turnover, or for data analysis in other related fields.

[0094] Regarding how to determine the error value, for example, obtain the result of each iterative calculation and compare it with the corresponding data in the validation set. If the result of each iterative calculation is greater than the corresponding data in the validation set, the difference value is:

[0095] T=(x1-y1)x1

[0096] Among them, x1 is the result of each iterative calculation, and y1 is the corresponding data in the validation set.

[0097] Similarly, if the result of each iterative calculation is less than the corresponding data in the validation set, the difference value is:

[0098] T=(y1-x1)x1

[0099] Among them, x1 is the result of each iterative calculation, and y1 is the corresponding data in the validation set.

[0100] S104: Obtain the current cost data, current net profit data, and current turnover data of the target enterprise, and determine the target matching index based on the current cost data, current net profit data, and current turnover data.

[0101] As you can understand, this index reflects a company's operational efficiency across multiple dimensions, including cost control, profitability, and market performance. By comparing the target match index with the industry average or the indices of other competing companies, we can further assess the target company's market competitiveness and financial health.

[0102] S105: Based on big data, multiple matching historical data of multiple companies in the same industry are obtained, where the matching historical data include matching historical cost data, matching historical net profit data, and matching historical turnover data. The multiple matching historical data are input into a neural network model, and multiple predicted matching indexes are obtained according to the output results of the neural network model.

[0103] As you can understand, by leveraging big data technology, we can collect and analyze historical data from multiple companies in the same industry. This data covers multiple dimensions, including matching historical costs, matching historical net profits, and matching historical sales. This rich historical matching data is then fed into a neural network model. By learning and processing this data, the model outputs a series of predictions. Based on these outputs, we can derive multiple predicted matching indices, which can help us better understand market trends and company performance.

[0104] S106: Compare the multiple predicted matching indices with the target matching indices, and determine at least three matching indices closest to the target matching indices from the multiple predicted matching indices and use them as reference matching indices.

[0105] For example, multiple predicted matching indices can be sorted according to index size, the maximum and minimum values among the multiple predicted matching indices can be eliminated, and then the remaining multiple matching indices can be matched with the target matching index to determine the three closest matching indices and use them as reference matching indices.

[0106] S107: Obtain matching historical data corresponding to the reference matching index, and determine the reference cost ratio of the target enterprise, wherein the reference cost ratio is the ratio of the matching historical cost data to the matching historical net profit data in the corresponding matching historical data.

[0107] For example, the matching historical data corresponding to the reference matching index can be obtained, and the proportional relationship between the matching historical cost data and the matching historical net profit data in a set of matching historical data corresponding to the reference matching index can be calculated. Then, the average value of multiple sets of proportional relationships can be obtained and used as the reference cost ratio.

[0108] S108: Adjust the current cost data based on the reference cost ratio.

[0109] It is understandable that the current cost data can be adjusted accordingly based on the reference cost ratio mentioned.

[0110] The present application provides an enterprise cost optimization method based on big data. The method aligns the timestamps of historical cost data, historical net profit data, and historical turnover data, and then imports multiple training sets into a neural network model. The neural network model is used to iteratively calculate the historical cost data, historical net profit data, and historical turnover data and output a matching index. Then, the current cost data, current net profit data, and current turnover data of a target enterprise are obtained, and a target matching index is determined based on the current cost data, current net profit data, and current turnover data. Then, based on the big data, multiple matching historical data of multiple enterprises in the same industry are obtained. The matching historical data includes matching historical cost data, matching historical net profit data, and matching historical turnover data. The multiple matching historical data are input into the neural network model, and multiple predicted matching indices are obtained based on the output results of the neural network model. Then, the multiple predicted matching indices are compared with the target matching indices. At least three matching indices closest to the target matching indices are determined from the multiple predicted matching indices and used as reference matching indices. Then, matching historical data corresponding to the reference matching indices are obtained, a reference cost ratio of the target enterprise is determined, and finally, the current cost data is adjusted based on the reference cost ratio. The embodiment of the present application proposes a method for enterprise cost optimization based on big data, which constructs a model that conforms to the characteristics of the target enterprise through historical data, uses big data to screen out matching enterprises with similar characteristics to the target enterprise, and uses the data of the matching enterprises to optimize the data of the target enterprise, so that the cost of the target enterprise is closer to that of different enterprises in the same industry.

[0111] Based on the same inventive concept, the present application embodiment proposes a big data-based enterprise cost optimization system, which is configured as follows:

[0112] Identify the target enterprise and obtain the historical data of the target enterprise, where the historical data includes historical cost data, historical net profit data, and historical turnover data;

[0113] Aligning the timestamps of the historical cost data, the historical net profit data, and the historical turnover data, and generating multiple training sets, wherein each training set includes a historical cost data, a historical net profit data, and a historical turnover data;

[0114] Import multiple training sets into the neural network model, which is used to iteratively calculate historical cost data, historical net profit data, and historical turnover data, and output a matching index;

[0115] Obtaining the target enterprise's current cost data, current net profit data, and current turnover data, and determining a target matching index based on the current cost data, current net profit data, and current turnover data;

[0116] Based on big data, multiple matching historical data of multiple companies in the same industry are obtained, where the matching historical data includes matching historical cost data, matching historical net profit data, and matching historical turnover data. The multiple matching historical data are input into a neural network model, and multiple predicted matching indexes are obtained based on the output results of the neural network model.

[0117] Comparing the plurality of predicted matching indices with the target matching indices, and determining at least three matching indices closest to the target matching indices from the plurality of predicted matching indices as reference matching indices;

[0118] Obtain matching historical data corresponding to the reference matching index and determine the reference cost ratio of the target enterprise, wherein the reference cost ratio is the ratio of the matching historical cost data to the matching historical net profit data in the corresponding matching historical data;

[0119] Adjusts current cost data based on a reference cost scale.

[0120] In conjunction with the second aspect, in some embodiments, the system is configured to:

[0121] Align the timestamps of historical cost data, historical net profit data, and historical turnover data, and generate multiple training sets, where each training set includes a historical cost data, a historical net profit data, and a historical turnover data, including:

[0122] The historical data is divided into training set, validation set and test set, where the total number of training sets is the same as the total number of validation sets and test sets.

[0123] In conjunction with the second aspect, in some embodiments, the system is configured to:

[0124] Multiple training sets are imported into the neural network model. The neural network model is used to iteratively calculate historical cost data, historical net profit data, and historical turnover data, and output matching indexes, including:

[0125] Normalize historical cost data, historical net profit data, and historical turnover data so that the data ranges are within the same range;

[0126] The historical cost data, historical net profit data and historical turnover data are used as three different dimensions, which constitute the characteristic vector of the matching index;

[0127] Iteratively calculate historical cost data, historical net profit data, and historical turnover data, and stop calculating when the conditions are met and output the trained neural network model.

[0128] In conjunction with the second aspect, in some embodiments, the system is configured to:

[0129] When the conditions are met, stop the calculation and output the trained neural network model, including:

[0130] Match the results calculated at each iteration with the data in the validation set and determine the difference value;

[0131] When the difference value is less than the preset value, the iterative calculation is stopped and the trained neural network model is output.

[0132] In conjunction with the second aspect, in some embodiments, the system is configured to:

[0133] The results calculated at each iteration are matched against the data in the validation set and the difference values are determined, including:

[0134] Get the result of each iterative calculation and compare it with the corresponding data in the validation set. If the result of each iterative calculation is greater than the corresponding data in the validation set, the difference value is:

[0135] T=(x1-y1)x1

[0136] Among them, x1 is the result of each iterative calculation, and y1 is the corresponding data in the validation set.

[0137] In conjunction with the second aspect, in some embodiments, the system is configured to:

[0138] Get the result of each iterative calculation and compare it with the corresponding data in the validation set. If the result of each iterative calculation is less than the corresponding data in the validation set, the difference value is:

[0139] T=(y1-x1)x1

[0140] Among them, x1 is the result of each iterative calculation, and y1 is the corresponding data in the validation set.

[0141] In conjunction with the second aspect, in some embodiments, the system is configured to:

[0142] Comparing the multiple predicted matching indices with the target matching indices, determining at least three matching indices closest to the target matching indices from the multiple predicted matching indices and using them as reference matching indices, including:

[0143] Sort the matching indexes of multiple predictions by their size, and remove the maximum and minimum matching indexes among the multiple predictions;

[0144] The remaining multiple matching indices are matched with the target matching index, and the three closest matching indices are determined and used as reference matching indices.

[0145] In conjunction with the second aspect, in some embodiments, the system is configured to:

[0146] Obtain matching historical data corresponding to the reference matching index and determine the reference cost ratio of the target enterprise, where the reference cost ratio is the ratio of the matching historical cost data to the matching historical net profit data in the corresponding matching historical data, including:

[0147] Obtain matching historical data corresponding to a reference matching index, and calculate a proportional relationship between matching historical cost data and matching historical net profit data in a set of matching historical data corresponding to the reference matching index;

[0148] Get the average of multiple sets of proportional relationships and use it as the reference cost ratio.

[0149] The present application provides an enterprise cost optimization system based on big data. The system aligns the timestamps of historical cost data, historical net profit data, and historical turnover data, and then imports multiple training sets into a neural network model. The neural network model is used to iteratively calculate the historical cost data, historical net profit data, and historical turnover data and output a matching index. Then, the system obtains the current cost data, current net profit data, and current turnover data of a target enterprise and determines a target matching index based on the current cost data, current net profit data, and current turnover data. Then, based on the big data, multiple matching historical data of multiple enterprises in the same industry are obtained. The matching historical data includes matching historical cost data, matching historical net profit data, and matching historical turnover data. The multiple matching historical data are input into the neural network model. Multiple predicted matching indices are obtained based on the output results of the neural network model. Then, the multiple predicted matching indices are compared with the target matching indices. At least three matching indices closest to the target matching indices are determined from the multiple predicted matching indices and used as reference matching indices. Then, the matching historical data corresponding to the reference matching indices are obtained to determine a reference cost ratio of the target enterprise. Finally, the current cost data is adjusted based on the reference cost ratio. The embodiment of the present application proposes a big data-based enterprise cost optimization system, which constructs a model that conforms to the characteristics of the target enterprise through historical data, uses big data to screen out matching enterprises with similar characteristics to the target enterprise, and uses the data of the matching enterprises to optimize the data of the target enterprise, so that the cost of the target enterprise is closer to that of different enterprises in the same industry.

[0150] Based on the same inventive concept, an embodiment of the present application further provides an electronic device, the electronic device comprising:

[0151] At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the enterprise cost optimization method based on big data of an embodiment of the present application.

[0152] In addition, to achieve the above-mentioned purpose, an embodiment of the present application also proposes a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the enterprise cost optimization method based on big data of an embodiment of the present application.

[0153] The following is a detailed introduction to the various components of electronic equipment:

[0154] The term "processor" is the control center of an electronic device and may be a single processor or a collective term for multiple processing elements. For example, the processor may be one or more central processing units (CPUs), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement embodiments of the present invention, such as one or more digital signal processors (DSPs) or one or more field programmable gate arrays (FPGAs).

[0155] Optionally, the processor can perform various functions of the electronic device by running or executing a software program stored in the memory, and calling data stored in the memory.

[0156] The memory is used to store the software program for executing the solution of the present invention, and the execution is controlled by the processor. The specific implementation method can refer to the above method embodiment and will not be repeated here.

[0157] Alternatively, the memory may be a read-only memory (ROM) or other type of static storage device that can store static information and instructions, a random access memory (RAM) or other type of dynamic storage device that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, an optical disc storage (including a compact disc, laser disc, optical disc, digital versatile disc, Blu-ray disc, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory may be integrated with the processor or exist independently and be coupled to the processor through an interface circuit of the electronic device, and this is not specifically limited in the embodiments of the present invention.

[0158] A transceiver is used to communicate with network devices or terminal devices.

[0159] Optionally, the transceiver may include a receiver and a transmitter, wherein the receiver is used to implement a receiving function, and the transmitter is used to implement a sending function.

[0160] Optionally, the transceiver may be integrated with the processor, or may exist independently and be coupled to the processor via an interface circuit of the router, which is not specifically limited in the embodiment of the present invention.

[0161] In addition, the technical effects of the electronic device can refer to the technical effects of the data transmission method in the above method embodiment, and will not be repeated here.

[0162] It should be understood that the processor in the embodiments of the present invention may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc.

[0163] It should also be understood that the memory in the embodiments of the present invention may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic random access memory (DRAM), synchronous DRAM (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct rambus RAM (DR RAM).

[0164] The above embodiments can be implemented in whole or in part through software, hardware (such as circuits), firmware, or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. A computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer program are loaded or executed on a computer, the process or function according to the embodiments of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. Computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via a wired method (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains a collection of one or more available media. The available media can be magnetic media (such as floppy disks, hard disks, tapes), optical media (such as DVDs), or semiconductor media. The semiconductor media can be a solid-state drive.

[0165] It should be understood that the term "and / or" as used herein simply describes a relationship between associated objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A alone, A and B together, or B alone. A and B can be singular or plural. Furthermore, the character " / " as used herein generally indicates an "or" relationship between the associated objects, but it may also indicate an "and / or" relationship. For specific understanding, please refer to the context.

[0166] In this disclosure, "at least one" means one or more, and "plurality" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, "at least one of a, b, or c" can mean: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or plural.

[0167] It should be understood that in various embodiments of the present invention, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0168] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present invention.

Claims

1. A method for enterprise cost optimization based on big data, characterized in that: The method comprises: Determine a target enterprise and obtain historical data of the target enterprise, wherein the historical data includes historical cost data, historical net profit data, and historical turnover data; Aligning the timestamps of the historical cost data, the historical net profit data, and the historical turnover data, and generating a plurality of training sets, wherein each training set includes one historical cost data, one historical net profit data, and one historical turnover data; Importing the plurality of training sets into a neural network model, wherein the neural network model is used to iteratively calculate the historical cost data, the historical net profit data, and the historical turnover data, and output a matching index; Obtaining current cost data, current net profit data, and current turnover data of the target enterprise, and determining a target matching index based on the current cost data, the current net profit data, and the current turnover data; Acquire multiple matching historical data of multiple companies in the same industry based on big data, wherein the matching historical data includes matching historical cost data, matching historical net profit data, and matching historical turnover data, input the multiple matching historical data into the neural network model, and obtain multiple predicted matching indexes based on output results of the neural network model; Comparing the plurality of predicted matching indices with the target matching indices, and determining at least three matching indices closest to the target matching indices from the plurality of predicted matching indices as reference matching indices; Obtaining the matching historical data corresponding to the reference matching index, and determining a reference cost ratio of the target enterprise, wherein the reference cost ratio is a ratio of the matching historical cost data to the matching historical net profit data in the corresponding matching historical data; The current cost data is adjusted based on the reference cost ratio.

2. The enterprise cost optimization method based on big data according to claim 1, characterized in that: Aligning the timestamps of the historical cost data, the historical net profit data, and the historical turnover data, and generating multiple training sets, wherein each training set includes one historical cost data, one historical net profit data, and one historical turnover data, including: The historical data is divided into a training set, a validation set, and a test set, wherein the total number of the training set is the same as the total number of the validation set and the test set.

3. The enterprise cost optimization method based on big data according to claim 2, characterized in that: Importing the plurality of training sets into a neural network model, the neural network model is used to iteratively calculate the historical cost data, the historical net profit data, and one of the historical turnover data, and output a matching index, including: Normalizing the historical cost data, the historical net profit data, and the historical turnover data so that the data ranges are within the same range; The historical cost data, the historical net profit data, and the historical turnover data are used as three different dimensions, and the three dimensions constitute a feature vector of the matching index; The historical cost data, the historical net profit data and the historical turnover data are iteratively calculated, and when a condition is met, the calculation is stopped and the trained neural network model is output.

4. The enterprise cost optimization method based on big data according to claim 3 is characterized in that: Iteratively calculating the historical cost data, the historical net profit data, and one of the historical turnover data, and stopping the calculation when a condition is met and outputting the trained neural network model, including: Match the results calculated at each iteration with the data in the validation set and determine the difference value; When the difference value is less than a preset value, the iterative calculation is stopped and the trained neural network model is output.

5. The enterprise cost optimization method based on big data according to claim 4 is characterized in that: The results calculated at each iteration are matched against the data in the validation set and the difference values are determined, including: Obtain the result of each iterative calculation and compare it with the corresponding data in the validation set. If the result of each iterative calculation is greater than the corresponding data in the validation set, the difference value is: T=(x1-y1)x1 Wherein, x1 is the result calculated in each iteration, and y1 is the corresponding data in the validation set.

6. The enterprise cost optimization method based on big data according to claim 5, characterized in that: Obtain the result of each iterative calculation and compare it with the corresponding data in the validation set. If the result of each iterative calculation is smaller than the corresponding data in the validation set, the difference value is: T=(y1-x1)x1 Wherein, x1 is the result calculated in each iteration, and y1 is the corresponding data in the validation set.

7. The enterprise cost optimization method based on big data according to claim 1, characterized in that: Comparing the plurality of predicted matching indices with the target matching indices, and determining at least three matching indices closest to the target matching indices from the plurality of predicted matching indices as reference matching indices, including: Sort the matching indexes of the multiple predictions by index size, and remove the maximum and minimum values of the matching indexes of the multiple predictions; The remaining multiple matching indexes are matched with the target matching index, and the three closest matching indexes are determined and used as reference matching indexes.

8. The enterprise cost optimization method based on big data according to claim 1, characterized in that: Obtain the matching historical data corresponding to the reference matching index, and determine a reference cost ratio of the target enterprise, wherein the reference cost ratio is the ratio of the matching historical cost data to the matching historical net profit data in the corresponding matching historical data, including: Acquire the matching historical data corresponding to the reference matching index, and calculate a proportional relationship between the matching historical cost data and the matching historical net profit data in a set of the matching historical data corresponding to the reference matching index; An average value of multiple groups of the proportional relationships is obtained and used as the reference cost ratio.

9. An enterprise cost optimization system based on big data, characterized in that: The system is configured to: Determine a target enterprise and obtain historical data of the target enterprise, wherein the historical data includes historical cost data, historical net profit data, and historical turnover data; Aligning the timestamps of the historical cost data, the historical net profit data, and the historical turnover data, and generating a plurality of training sets, wherein each training set includes one historical cost data, one historical net profit data, and one historical turnover data; Importing the plurality of training sets into a neural network model, wherein the neural network model is used to iteratively calculate the historical cost data, the historical net profit data, and the historical turnover data, and output a matching index; Obtaining current cost data, current net profit data, and current turnover data of the target enterprise, and determining a target matching index based on the current cost data, the current net profit data, and the current turnover data; Acquire multiple matching historical data of multiple companies in the same industry based on big data, wherein the matching historical data includes matching historical cost data, matching historical net profit data, and matching historical turnover data, input the multiple matching historical data into the neural network model, and obtain multiple predicted matching indexes based on output results of the neural network model; Comparing the plurality of predicted matching indices with the target matching indices, and determining at least three matching indices closest to the target matching indices from the plurality of predicted matching indices as reference matching indices; Obtaining the matching historical data corresponding to the reference matching index, and determining a reference cost ratio of the target enterprise, wherein the reference cost ratio is a ratio of the matching historical cost data to the matching historical net profit data in the corresponding matching historical data; The current cost data is adjusted based on the reference cost ratio.

10. An electronic device, characterized in that: include: at least one processor; And, a memory communicatively connected to at least one of the processors; wherein the memory stores instructions that can be executed by at least one of the processors, and the instructions are executed by at least one of the processors to enable at least one of the processors to execute the method according to any one of claims 1 to 8.

Images