Method, system, device and terminal for assigning null values to multi-dimensional data of power supplier

Abstract

The application belongs to the technical field of data processing, and discloses a power supplier multi-dimensional data null value assignment method, system, device and terminal. Basic data of the supplier multi-dimension is input. According to the null value rate of each data dimension, the data is divided into a basic data set and a multi-null value data set. Each column of the multi-null value data set and each supplier data null value are calculated by the basic data. Comparison and analysis are carried out by using an industry change factor, and the multi-null data set is assigned. The multi-null data set is taken as the basic data set, and the basic data set is taken as the multi-null data set, so that the assignment of the basic data set is completed. The basic data set and the multi-null data set are merged, and the assigned data set is output. The power supplier multi-dimensional data null value assignment method based on industry ranking selection provided by the application improves the scientificity of the null value assignment, reduces the influence of subjective assignment on data ranking, and has better objectivity.

Description

Technical Field

[0001] This invention belongs to the field of data processing technology, and in particular relates to a method, system, device and terminal for assigning null values ​​to multidimensional data of power suppliers. Background Technology

[0002] Currently, the power industry chain is gradually evolving and upgrading towards a new power system industry chain, posing new requirements for the power material manufacturing and supply system. Companies should fully grasp the evolutionary trajectory of the new power system industry development, and, relying on the foundation of big data on power materials, conduct comprehensive and systematic assessments of material suppliers' qualifications, capabilities, equipment production, testing, and operation under the new circumstances, thereby continuously improving the supplier manufacturing system assessment mechanism. However, in the multi-dimensional data supplier manufacturing system assessment based on big data, a large number of null values ​​appear in different dimensions of the data, and a large amount of supplier data cannot be comprehensively collected. Therefore, how to assign null values ​​to supplier data in different dimensions becomes the basis for determining the multi-dimensional evaluation of suppliers.

[0003] Traditional data assignment methods primarily rely on simple statistical approaches such as direct assignment and averaging. These methods fail to fully consider the unique characteristics of the industry's data, leading to significant subjectivity in the assigned values. Furthermore, assigning null values ​​can severely impact supplier rankings, resulting in a lack of scientific rigor in the overall data application. Therefore, there is an urgent need to design a new multidimensional data null value assignment method.

[0004] Based on the above analysis, the problems and shortcomings of the existing technology are as follows:

[0005] (1) In the existing evaluation of the multi-dimensional data supplier manufacturing system based on big data, a large number of null values ​​appeared in the data of different dimensions, and a large amount of supplier data could not be collected comprehensively.

[0006] (2) Traditional methods of assigning values ​​through direct assignment, average values, and other statistical methods do not fully consider the data characteristics of the entire industry, resulting in a high degree of subjectivity in the assigned data. The method of assigning null values ​​has a significant impact on the actual ranking of suppliers, which to some extent leads to a lack of scientific basis in the overall data application. Summary of the Invention

[0007] To address the problems existing in the prior art, this invention provides a method, system, device, and terminal for assigning null values ​​to multi-dimensional data of power suppliers, and particularly relates to a method, system, medium, device, and terminal for assigning null values ​​to multi-dimensional data of power suppliers selected based on industry ranking.

[0008] This invention is implemented as follows: a method for assigning null values ​​to multidimensional data of power suppliers. The method includes: obtaining a multidimensional dataset; decomposing the multidimensional dataset into a basic dataset and a multidimensional dataset; distinguishing the null value data columns of suppliers and establishing an indicator based on industry ranking changes; selecting suppliers to assign average values ​​to the multidimensional dataset and the basic dataset respectively, and adjusting the results using the ranking changes after the assignment; finally merging and outputting the dataset.

[0009] Furthermore, the method for assigning null values ​​to multidimensional data of power suppliers includes the following steps:

[0010] Step 1: Input the supplier's multidimensional basic data; based on the null value rate of each data dimension, divide the data into a basic dataset and a multi-null value dataset;

[0011] Step 2: Calculate the null values ​​for each supplier in each column of the multi-null value dataset based on the basic data, perform comparative analysis using industry change factors, and assign values ​​to the multi-null dataset.

[0012] Step 3: Use the completed multi-short dataset as the base dataset, and the base dataset as the multi-short dataset, to complete the assignment of values ​​to the base dataset;

[0013] Step 4: Merge the base dataset and the multi-void dataset, and output the assigned dataset.

[0014] Furthermore, the data set decomposition process in step one includes:

[0015] (1) Obtain the supplier multidimensional dataset;

[0016] (2) Calculate the null value rate of all data in the dataset, and then calculate the average a and median b of all null value rates; when a≥b, select b as the null value splitting rate c, otherwise take a as the null value splitting rate c;

[0017] (3) When the null value rate d of the data column and / or dimension is greater than or equal to the null value splitting rate c, it is a multi-null dataset E; otherwise, it is a basic dataset F.

[0018] (4) Divide the dataset into a multi-null value dataset E and a basic dataset F.

[0019] Furthermore, the assignment process for the multi-spatial dataset in step two includes:

[0020] (1) Receive the multi-null value dataset E and the basic dataset F;

[0021] (2) Select the one-dimensional data with the smallest null value rate in F, and calculate the similar data of the null value position in dataset E in turn; select the most similar N data, take the average value and assign it to the null value position, until all null value positions are assigned;

[0022] (3) Calculate the average rank of each data row in E according to the basic weight; calculate the average rank of the rows in the E+ assigned column, and calculate the influence factor Q;

[0023] (4) When Q ≥ 1.5 times the number of data, N = N + 2 and repeat steps (2) to (4); when there is no data that meets the requirements, take the empty value with the smallest Q value as the filling value;

[0024] (5) Add the data column with the completed assignment to E to form new E and F, and repeat steps (1) to (5) to assign data to the next column;

[0025] (6) Output data E.

[0026] Furthermore, the calculation process for assigning values ​​to empty bits in step (2) includes:

[0027] Given a dataset E and rows with empty values, rank each dimension of E; calculate the rank difference between each dimension of the empty rows and the data in each row, and sum the rank differences W.

[0028] Rank the data according to the size of the W value, select the N data rows with the smallest W value; obtain the average number of the N rows in the assignment column, and use the average value to assign a value to the data column. Null values ​​are not included in the calculation.

[0029] Furthermore, the calculation process of the impact factor in step (3) includes:

[0030] 1) Input the data set E and the column to be assigned;

[0031] 2) Rank each dimension of E and take the average of each row, then rank P1 according to the average.

[0032] 3) Add the assigned column to E and calculate P2 according to step 2);

[0033] 4) Calculate the ranking change for each row in turn: Z = (P1*P1 - P2*P2) / 2P1, Q = Z1 + Z2 + ... + Zn;

[0034] 5) Output data impact factor Q.

[0035] Furthermore, the assignment process for the basic dataset in step three includes:

[0036] (1) Receive dataset E;

[0037] (2) Divide the columns in E that have no null values ​​into E, and the columns that have null values ​​into F;

[0038] (3) Perform step two to obtain dataset E.

[0039] Another objective of this invention is to provide a power supplier multidimensional data null value assignment system that applies the aforementioned power supplier multidimensional data null value assignment method. The power supplier multidimensional data null value assignment system includes:

[0040] The dataset partitioning module is used to input multidimensional basic data from suppliers; based on the null value rate of each data dimension, the data is divided into a basic dataset and a multi-null value dataset.

[0041] The multi-void dataset assignment module is used to calculate the null values ​​for each supplier's data in each column of the multi-void dataset based on the basic data, and to perform comparative analysis using industry change factors to assign values ​​to the multi-void dataset.

[0042] The base dataset assignment module is used to assign values ​​to the base dataset by using the completed multi-single dataset as the base dataset and the base dataset as the multi-single dataset.

[0043] The dataset merging module is used to merge the base dataset and the multi-empty dataset and output the assigned dataset.

[0044] Another object of the present invention is to provide a computer device, the computer device including a memory and a processor, the memory storing a computer program, and when the computer program is executed by the processor, causing the processor to perform the steps of the described method for assigning null values ​​to multidimensional data of power suppliers.

[0045] Another object of the present invention is to provide a computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the steps of the described method for assigning null values ​​to multidimensional data of a power supplier.

[0046] Another objective of this invention is to provide an information data processing terminal for implementing the aforementioned power supplier multi-dimensional data null value assignment system.

[0047] Based on the above technical solutions and the technical problems solved, the advantages and positive effects of the technical solution to be protected by this invention are as follows:

[0048] First, addressing the technical problems existing in the prior art and the difficulty of solving them, this paper closely analyzes, in conjunction with the technical solution to be protected by this invention and the results and data obtained during the research and development process, how the technical solution of this invention solves the technical problems, and the inventive technical effects brought about by solving these problems. The specific description is as follows:

[0049] This invention proposes a method for assigning null values ​​to multidimensional data of power suppliers based on industry ranking. This method assigns null values ​​to multidimensional supplier manufacturing process evaluation data. By distinguishing the null value data columns of suppliers and using the ranking changes after assignment for self-adjustment, it can effectively improve the assignment of null values, thereby promoting the usability of multidimensional supplier data and enhancing the scientific nature of the overall evaluation system.

[0050] The present invention provides a method for assigning null values ​​to multidimensional data of power suppliers selected based on industry rankings. Compared with traditional methods, this method fully considers the data characteristics and industry features of multidimensional data, as well as the correlations between different data sets. It assigns values ​​through specific similar datasets, making the scientific nature of null value assignment more realistic. This method improves the scientific rigor of null value assignment and reduces the impact of subjective assignment on data ranking, thus exhibiting better objectivity.

[0051] Second, considering the technical solution as a whole or from a product perspective, the technical effects and advantages of the technical solution to be protected by this invention are specifically described as follows:

[0052] The method for assigning null values ​​to multidimensional data of power suppliers provided by this invention establishes an indicator based on changes in industry rankings to make a scientific judgment on the impact of null value assignment. On the other hand, it selects suppliers with similar profile tags to assign values ​​by averaging, which can more objectively reflect the relationship between data and make null values ​​closer to reality, thereby improving the comprehensiveness and scientific nature of the basic data.

[0053] Third, as supplementary evidence of the inventive step of the claims of this invention, it is also reflected in the following important aspects:

[0054] The technical solution of this invention solves a technical problem that people have long desired to solve but have been unable to achieve:

[0055] This invention addresses the problem of numerous null values ​​in the calculation of comprehensive multi-dimensional data from power industry suppliers. Traditional solutions often subjectively select null value assignment methods such as mean or average. This invention, however, considers other dimensions of supplier information and then uses a label-like approach to select similar suppliers. Null values ​​are assigned based on the values ​​of these similar suppliers, significantly solving the current problem of non-targeted assignment. Furthermore, this invention creatively proposes evaluation criteria for null value assignment, constructs an industry ranking difference factor calculation, and judges the scientific validity of the assignment, thus ensuring the scientific accuracy of null value assignment to the greatest extent possible. Attached Figure Description

[0056] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the embodiments of the present invention will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0057] Figure 1 This is a flowchart of the method for assigning null values ​​to multidimensional data of power suppliers provided in an embodiment of the present invention;

[0058] Figure 2 This is a schematic diagram of the method for assigning null values ​​to multidimensional data of power suppliers provided in this embodiment of the invention;

[0059] Figure 3 This is a flowchart of the data set decomposition method provided in an embodiment of the present invention;

[0060] Figure 4 This is a flowchart of the multi-void dataset assignment method provided in this embodiment of the invention;

[0061] Figure 5 This is a flowchart of the null value assignment calculation method provided in the embodiments of the present invention;

[0062] Figure 6 This is a flowchart of the impact factor calculation method provided in the embodiments of the present invention;

[0063] Figure 7 This is a flowchart of the basic dataset assignment method provided in the embodiments of the present invention. Detailed Implementation

[0064] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0065] To address the problems existing in the prior art, this invention provides a method, system, device, and terminal for assigning null values ​​to multidimensional data of power suppliers. The invention will be described in detail below with reference to the accompanying drawings.

[0066] I. Explanatory and Illustrative Embodiments. To enable those skilled in the art to fully understand how the present invention is specifically implemented, this section provides an explanatory and illustrative description of the embodiments described in the claims.

[0067] like Figure 1 As shown, the method for assigning null values ​​to multidimensional data of power suppliers provided in this embodiment of the invention includes the following steps:

[0068] S101, Obtain the supplier's multidimensional dataset;

[0069] S102, decompose the multidimensional dataset into a basic dataset and a multinumeric dataset;

[0070] S103, differentiate the supplier null data column and establish an indicator based on changes in industry ranking;

[0071] S104: Select suppliers to assign average values ​​to the multi-null value dataset and the basic dataset respectively, adjust the ranking changes after the assignment, and finally merge and output the dataset.

[0072] As a preferred embodiment, such as Figure 2 As shown, the method for assigning null values ​​to multidimensional data of power suppliers provided in this embodiment of the invention specifically includes the following steps:

[0073] S1: Input the supplier's multidimensional basic data;

[0074] S2: Divide the data into basic dataset and multi-null dataset based on the null value rate of each data dimension;

[0075] S3: Calculate the null values ​​for each supplier's data in each column of the multi-null value dataset based on the basic data, compare and analyze them using industry change factors, and assign values ​​to the multi-null dataset.

[0076] S4: Use the completed multi-short dataset as the base dataset, and the base dataset as the multi-short dataset, to complete the assignment of values ​​to the base dataset;

[0077] S5: Merge the two datasets and output the assigned dataset.

[0078] like Figure 3 As shown, the decomposition process of the S2 data set provided in this embodiment of the invention includes:

[0079] S21: Obtain the supplier cube obtained in S1;

[0080] S22: Calculate the vacancy rate of all data in the dataset, then calculate the average a and median b of all vacancy rates. When a ≥ b, select b as the vacancy splitting rate c; otherwise, take a as the vacancy splitting rate c.

[0081] S23: If the null value rate d of the data column (dimension) is greater than or equal to the null value splitting rate c, it is a multi-null dataset E; otherwise, it is a basic dataset F.

[0082] S24: Data set E (multiple null values) and underlying dataset F.

[0083] like Figure 4 As shown, the assignment process for the S3 multi-space dataset provided in this embodiment of the invention includes:

[0084] S31: Receive datasets E and F;

[0085] S32: Select the one-dimensional data with the smallest null value rate in F, calculate the similar data of the null value position in dataset E in turn, select the most similar N data, take the average value and assign it to the null value position, until all null value positions are assigned.

[0086] S33: Calculate the average rank of each data row in E according to the basic weight, then calculate the average rank of the rows in the E+ assigned column, and calculate its influence factor Q;

[0087] S34: When Q ≥ 1.5 times the number of data points, repeat S32~S34 with N = N + 2. When there are no data points that meet the requirements, fill the blank with the smallest Q value.

[0088] S35: Add the data column with the completed assignment to E to form new E and F, and then repeat S31 to S35 to assign data to the next column.

[0089] S36: Output data E.

[0090] like Figure 5 As shown, the calculation of assigning null values ​​in S32 provided by the embodiment of the present invention is further explained, and is implemented in the following way:

[0091] S321: Input data set E and empty rows;

[0092] S322: Rank E for each dimension;

[0093] S323: Calculate the ranking difference between each dimension of the null value row and the data of each row, and sum the ranking differences W;

[0094] S324: Rank the data according to the size of W value and select the N data rows with the smallest W value;

[0095] S325: Rank the data according to the size of W value, select the N data rows with the smallest W value, and obtain the average number of the N rows in the assigned column (null values ​​are not included in the calculation);

[0096] S326: Assign a value to this column using the average.

[0097] like Figure 6 As shown in the embodiment of the present invention, the calculation of the influence factor in S33 is further explained, and is achieved in the following manner:

[0098] S331: Input data set E and the column to be assigned;

[0099] S332: Rank each dimension of E, take the average of each row, and then rank P1 according to the average.

[0100] S333: Add the assigned column to E and calculate P2 according to S332;

[0101] S334: Calculate the ranking change for each row in sequence: Z = (P1*P1 - P2*P2) / 2P1, Q = Z1 + Z2 + ... + Zn;

[0102] S335: Output data impact factor Q.

[0103] like Figure 7 As shown, the process of assigning values ​​to the basic dataset in S4 provided in this embodiment of the invention includes:

[0104] S41: Receive data E;

[0105] S42: Divide the columns in E that have no null values ​​into E, and the columns that have null values ​​into F;

[0106] S43: Execute S3;

[0107] S44: Obtain dataset E.

[0108] The multi-dimensional data null value assignment system for power suppliers provided in this embodiment of the invention includes:

[0109] The dataset partitioning module is used to input multidimensional basic data from suppliers; based on the null value rate of each data dimension, the data is divided into a basic dataset and a multi-null value dataset.

[0110] The multi-void dataset assignment module is used to calculate the null values ​​for each supplier's data in each column of the multi-void dataset based on the basic data, and to perform comparative analysis using industry change factors to assign values ​​to the multi-void dataset.

[0111] The base dataset assignment module is used to assign values ​​to the base dataset by using the completed multi-single dataset as the base dataset and the base dataset as the multi-single dataset.

[0112] The dataset merging module is used to merge the base dataset and the multi-empty dataset and output the assigned dataset.

[0113] II. Application Examples. To demonstrate the inventiveness and technical value of the technical solution of this invention, this section provides application examples of the technical solution of the claims on specific products or related technologies.

[0114] The method of this invention has been applied to the evaluation mechanism of the power supplier manufacturing system. When the average null value rate of the manufacturing evaluation system exceeded 54%, this method was applied to reduce the overall null value rate to 0%. Historical data was used for verification. The ranking after null value assignment had a similarity of more than 90% with the objective supplier bidding ranking, which is much greater than the similarity of assignment by average value and other methods. In particular, the null value assignment method of this invention and the subsequent value assignment were compared with historical data and the similarity of the assignment was greater than 85%.

[0115] III. Evidence of the Relevant Effects of the Embodiments. The embodiments of the present invention have achieved some positive effects during research and development or use, and indeed possess significant advantages compared to existing technologies. The following description, in conjunction with data, charts, and other materials from the experimental process, illustrates these advantages.

[0116] This invention references the supplier aggregation method. Compared with traditional direct assignment methods and initial value assignment methods, the data determined by this method is closer to the educational and categorical, and its industry ranking impact factor is lower.

[0117] It should be noted that embodiments of the present invention can be implemented in hardware, software, or a combination of both. The hardware portion can be implemented using dedicated logic; the software portion can be stored in memory and executed by a suitable instruction execution system, such as a microprocessor or dedicated-design hardware. Those skilled in the art will understand that the above-described devices and methods can be implemented using computer-executable instructions and / or included in processor control code, for example, such code provided on a carrier medium such as a disk, CD, or DVD-ROM, a programmable memory such as read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The devices and modules of the present invention can be implemented by hardware circuitry such as very large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field-programmable gate arrays, programmable logic devices, etc., or by software executed by various types of processors, or by a combination of the above-described hardware circuitry and software, such as firmware.

[0118] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications, equivalent substitutions, and improvements made by those skilled in the art within the scope of the technology disclosed in the present invention, and within the spirit and principles of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for assigning null values ​​to multidimensional data of an electricity supplier, characterized in that, The power supplier multi-dimensional data null value assignment method comprises the following steps: The power supplier multi-dimensional data null value assignment method comprises the following steps: Step one, input the supplier multi-dimensional basic data; according to the null value rate of each data dimension, divide the data into a basic data set and a multi-null value data set; Step two, calculate the null value of each column of the multi-null value data set for each supplier data in the basic data, compare and analyze by using the industry change factor, and assign values to the multi-null data set; Step three, complete the assignment of the basic data set by taking the completed multi-null data set as the basic data set and the basic data set as the multi-null data set; Step four, merge the basic data set and the multi-null data set, and output the assigned data set; The decomposition process of the data set in step one comprises: (1) obtain the supplier multi-dimensional data set; (2) calculate the null value rate of all data in the data set, and then calculate the average value a and the median b of all null value rates; when a≥b, take b as the null value division rate c, otherwise take a as the null value division rate c; (3) when the data column and / or dimension null value rate d≥null value division rate c, it is a multi-null value data set E, otherwise it is a basic data set F; (4) divide the data set into a multi-null value data set E and a basic data set F; The assignment process of the multi-null data set in step two comprises: (1) receive the multi-null value data set E and the basic data set F; (2) select the dimension with the smallest null value rate in F, and calculate the similar data in the null value position E data set in turn; select the most similar N, take the average value to assign the null value position, and repeat steps (2) and (4) until all null value positions are assigned; (3) calculate the average ranking of each data row in E according to the basic weight; calculate the average ranking of the row of the E+assigned column, and calculate the influence factor Q; (4) when Q≥1.5 times the number of data, N=N+2, repeat steps (2) and (4); when the required data is not reached at last, take the smallest Q value of the null value as the filling value; (5) add the assigned data column to E to form new E and F, and repeat steps (1)-(5) to assign values to the next column of data; (6) output data E; The calculation process of the null value position assignment in step (2) comprises: Input data set E and null value row, rank each dimension of E; calculate the ranking difference of each dimension of the null value row and each row of data, and sum the ranking difference W; Rank according to the size of W value, select N data rows with the smallest W value; get the average value of N rows in the assigned column, and assign values to the data column by using the average value, and the null value is not involved in the calculation; The calculation process of the influence factor in step (3) comprises: 1) input data set E and assigned column; 2) rank each dimension of E and take the average value of each row, and rank P1 according to the average value; 3) add the assigned column to E according to step 2) to calculate P2; 4) Calculate the rank change for each row in turn Q = Z1+ Z2+... + Zn; 5) output the data impact factor Q.

2. The power provider multi-dimensional data null assignment method of claim 1, wherein, The assignment process of the base data set in step three includes: (1) receive the data set E; (2) divide the columns without null values in E as E, and divide the columns with null values as F; (3) perform step two to obtain the data set E.

3. A power supplier multi-dimensional data null assignment system applying the power supplier multi-dimensional data null assignment method according to any one of claims 1 to 2, characterized by The power supplier multi-dimensional data null value assignment system includes: a data set division module for inputting the base data of the supplier multi-dimension; dividing the data into a base data set and a multi-null value data set according to the null value rate of each data dimension; a multi-null data set assignment module for calculating the null value of each column of each supplier data of the multi-null value data set by the base data, comparing and analyzing by using the industry change factor, and assigning the multi-null data set; a base data set assignment module for taking the completed multi-null data set as the base data set, taking the base data set as the multi-null data set, and completing the assignment of the base data set; a data set merging module for merging the base data set and the multi-null data set and outputting the assigned data set.

4. A computer device, comprising: The computer device includes a memory and a processor, the memory stores a computer program, and the computer program is executed by the processor to make the processor execute the steps of the power supplier multi-dimensional data null value assignment method according to any one of claims 1-2.

5. A computer readable storage medium, storing a computer program, the computer program being executed by a processor to make the processor execute the steps of the power supplier multi-dimensional data null value assignment method according to any one of claims 1-2.

6. An information data processing terminal, characterized by The information data processing terminal is used to realize the power supplier multi-dimensional data null value assignment system according to claim 3.

Images