CN-120780888-B - Enterprise data processing method and system based on structure tree

CN120780888BCN 120780888 BCN120780888 BCN 120780888BCN-120780888-B

Abstract

The invention provides an enterprise data processing method and system based on a structural tree, which can process enterprise data based on the structural tree, a genetic algorithm, an ant colony algorithm and the like. The business data of the enterprise can be rapidly decomposed, classified, stored or extracted in a structural tree mode, and the corresponding algorithm is adopted for processing, so that the relative accuracy and rationality of later calculation are effectively ensured. The individuals in the population in the genetic algorithm continuously carry out genetic operation, so that the individuals in the population continuously evolve towards the global optimal solution, the search space coverage area is larger, the better parallel search capability is achieved, the ant colony algorithm has positive feedback characteristics due to the continuous accumulation effect of pheromones, the algorithm is promoted to converge towards the optimal solution more quickly, multi-point search can be carried out on the solution space, and the robustness and the optimizing capability are stronger than those of other algorithms.

Inventors

WANG LEI
MU YAN
ZHOU ZHIJUAN
LIN XINCHEN
Bai Hanguang
WANG JIAQI

Assignees

兵器装备集团财务有限责任公司

Dates

Publication Date: 20260512
Application Date: 20250626

Claims (8)

1. A method for processing enterprise data based on a structure tree, comprising: The method comprises the steps that a server obtains first camping data of a first enterprise, and data classification is conducted on the first camping data according to preset multi-level dimensions to obtain a first structure tree and a first extraction table; The server receives second operation data of a second enterprise added by the analysis end, classifies the second operation data to obtain a second structure tree and a second extraction table, extracts operation specification information of the second structure tree, and performs partition processing on the first structure tree and the first extraction table based on the operation specification information to obtain a plurality of information areas; Decomposing the information of each dimension in the second business data according to a preset analysis point to obtain business specification information corresponding to each dimension, wherein the business specification information at least comprises a time period; generating a second structure tree and a second extraction table after the second operation data is classified based on the dimension and the time period; Processing and screening the first structure tree based on the second structure tree to obtain a plurality of information areas, synchronously adjusting the first extraction table and the second extraction table, wherein each information area has a corresponding category dimension and a corresponding time dimension; Retaining child nodes of a category dimension corresponding to the second structural tree in the first structural tree; determining the operation specification information of the grandchild node corresponding to the child node of each category dimension; Repartitioning the grandchild nodes reserved in the first structural tree according to the operation specification information by taking the second structural tree as a reference, and then removing non-corresponding nodes to obtain an updated first structural tree, a first extraction table and a corresponding information area; The server constructs information points corresponding to each information area, generates a corresponding first camping function according to corresponding preset multi-level dimensions, and fuses the first camping function based on the similarity of the first enterprise and the second enterprise to obtain a comprehensive function; And the server predicts the second operation data camp based on the comprehensive function to obtain prediction data and credibility information, and feeds the prediction data and credibility information back to the analysis end.
2. The method for processing structured tree based enterprise data processing as claimed in claim 1, wherein, The server obtains first camping data of a first enterprise, classifies the first camping data according to preset multi-level dimensions to obtain a first structure tree and a first extraction table, and the method comprises the following steps: the server screens enterprises in the database at intervals of a preset time period, and adds a first label to the corresponding enterprises after judging that the first condition is reached; The method comprises the steps that a server obtains first camping data of a first enterprise, wherein the first camping data at least comprises financial information, personnel information and stakeholder information; And decomposing the first camping data according to different types of dimensions based on the first time sequence interval to obtain a first structure tree and a first extraction table of time and types of multistage dimensions, wherein the first structure tree and the information in the first extraction table are correspondingly arranged.
3. The method for processing structured tree based enterprise data processing as claimed in claim 1, wherein, Decomposing the information of each dimension in the second operation data according to the preset analysis point to obtain operation specification information corresponding to each dimension, including: Acquiring a maximum span time value of information of each dimension in the second operation data, and if the maximum span time value is greater than or equal to a preset time value, acquiring preset operation specification information; And if the maximum span time value is smaller than the preset time value, equally dividing the maximum span time value based on the number of the preset analysis points to obtain the calculated operation specification information.
4. The method for processing structured tree based enterprise data processing as claimed in claim 1, wherein, The server constructs information points corresponding to each information area and generates a corresponding first camping function according to corresponding preset multi-level dimensions, and the first camping function is fused to obtain a comprehensive function based on the similarity of a first enterprise and a second enterprise, and the method comprises the following steps: The server reserves grandchild nodes in each initial information area based on preset or calculated operation specification information to obtain information points according to the child nodes of the first structure tree as the initial information areas; performing functional statistics on all the information points according to the time dimension of the information points to generate corresponding first campaigns, wherein each information area corresponds to one first campaigns; calculating a similarity weight generated based on the similarity of the first enterprise and the second enterprise and adding the similarity weight to a corresponding first structure tree; And fusing the first warp function pairs based on the similar weights in the first structural tree to obtain a comprehensive function.
5. The method for processing structured tree based enterprise data processing as claimed in claim 4, wherein, The calculating generates a corresponding similarity weight based on the similarity of the first enterprise and the second enterprise, and fuses the similarity weight and the first camping function to obtain a comprehensive function, and the calculating comprises the following steps: Extracting a first attribute tag and a second attribute tag of a first enterprise and a second enterprise, and calculating a quantized difference value of the first attribute tag and the second attribute tag to obtain an attribute difference value under each dimension; comparing the original first structure tree with the second structure tree to obtain a child node difference value and a grandchild node difference value; and obtaining the similarity of the first enterprise and the second enterprise based on the attribute difference value, the child node difference value and the grandchild node difference value, respectively carrying out percentage processing on the similarity of all the first enterprises and the second enterprise to obtain similar weights, and carrying out treelizing fusion on the first camping function structure based on the similar weights to obtain the comprehensive function.
6. The method for processing structured tree based enterprise data processing as claimed in claim 5, wherein, The step of treeing and fusing the first camping function structure based on the similar weight to obtain a comprehensive function comprises the following steps: the parent node of each first enterprise is used as a child node and then connected with the same parent node to obtain an analysis structure tree, and the similar weight is corresponding to the child node of the analysis structure tree; determining the dimension weight of each grandchild node in the analysis structure tree; and predicting the second operation data camp based on the analysis structure tree to obtain prediction data and reliability information, and feeding the prediction data and the reliability information back to the analysis end.
7. An enterprise data processing method applying any one of claims 1 to 6, comprising: Step S1, initializing ant colony related parameters and related parameters of a genetic algorithm in the enterprise revenue trend prediction calculation process, wherein the ant colony related parameters and the related parameters comprise ant colony size, pheromone update rate, population size of the genetic algorithm and probability of intersection and variation; s2, generating a group of initial solutions by using an ant colony algorithm, and calculating the fitness of each solution; Selecting a part of individuals with higher fitness from the initial solution by using a genetic algorithm as a population; S3, performing crossing and mutation operation on the population to generate a next generation of individuals; S4, calculating the fitness of the next generation of individuals; And S5, outputting a result if the specified iteration times are reached or a preset satisfaction solution is found, otherwise, returning to the step S3, and continuing to optimize.
8. An enterprise data processing system based on a structural tree, comprising: The acquisition module is used for enabling the server to acquire first camping data of a first enterprise, and classifying the first camping data according to preset multi-level dimensions to obtain a first structure tree and a first extraction table; the extraction module is used for enabling the server to receive second operation data of a second enterprise added by the analysis end, classifying the second operation data to obtain a second structure tree and a second extraction table, extracting operation specification information of the second structure tree, and carrying out partition processing on the first structure tree and the first extraction table based on the operation specification information to obtain a plurality of information areas; Decomposing the information of each dimension in the second business data according to a preset analysis point to obtain business specification information corresponding to each dimension, wherein the business specification information at least comprises a time period; generating a second structure tree and a second extraction table after the second operation data is classified based on the dimension and the time period; Processing and screening the first structure tree based on the second structure tree to obtain a plurality of information areas, synchronously adjusting the first extraction table and the second extraction table, wherein each information area has a corresponding category dimension and a corresponding time dimension; Retaining child nodes of a category dimension corresponding to the second structural tree in the first structural tree; determining the operation specification information of the grandchild node corresponding to the child node of each category dimension; Repartitioning the grandchild nodes reserved in the first structural tree according to the operation specification information by taking the second structural tree as a reference, and then removing non-corresponding nodes to obtain an updated first structural tree, a first extraction table and a corresponding information area; The generation module is used for enabling the server to construct information points corresponding to each information area, generating a corresponding first camping function according to corresponding preset multi-level dimensions, and fusing the first camping function based on the similarity of the first enterprise and the second enterprise to obtain a comprehensive function; and the prediction module is used for enabling the server to predict the second operation data revenue based on the comprehensive function to obtain prediction data and reliability information and feeding the prediction data and the reliability information back to the analysis end.

Description

Enterprise data processing method and system based on structure tree Technical Field The present invention relates to data processing technologies, and in particular, to a method and a system for processing enterprise data based on a structure tree. Background Enterprise revenue prediction is an important tool for enterprise strategic planning, budgeting, and investment decisions, often estimated in combination with historical data, market analysis, and statistical models. In the business process of enterprises, certain rules and market periods often exist, the prior art cannot reduce the data statistics of manpower through a large data multidimensional analysis mode, and the data statistics efficiency for revenue prediction is improved under the condition that the revenue of the enterprises is relatively accurate. Disclosure of Invention The embodiment of the invention provides a structural tree-based enterprise data processing method and system, which can be used for technically predicting enterprise revenue, reduce manpower in a multi-dimensional analysis mode of big data, improve the statistical efficiency of predicted data and improve the prediction efficiency under the condition of ensuring relatively accurate enterprise revenue. In a first aspect of an embodiment of the present invention, there is provided a method for processing enterprise data based on a structure tree, including: The method comprises the steps that a server obtains first camping data of a first enterprise, and data classification is conducted on the first camping data according to preset multi-level dimensions to obtain a first structure tree and a first extraction table; The server receives second operation data of a second enterprise added by the analysis end, classifies the second operation data to obtain a second structure tree and a second extraction table, extracts operation specification information of the second structure tree, and performs partition processing on the first structure tree and the first extraction table based on the operation specification information to obtain a plurality of information areas; The server constructs information points corresponding to each information area, generates a corresponding first camping function according to corresponding preset multi-level dimensions, and fuses the first camping function based on the similarity of the first enterprise and the second enterprise to obtain a comprehensive function; And the server predicts the second operation data camp based on the comprehensive function to obtain prediction data and credibility information, and feeds the prediction data and credibility information back to the analysis end. Optionally, in one possible implementation manner of the first aspect, the server obtains first camping data of the first enterprise, classifies the first camping data according to a preset multi-level dimension to obtain a first structural tree and a first extraction table, and includes: the server screens enterprises in the database at intervals of a preset time period, and adds a first label to the corresponding enterprises after judging that the first condition is reached; The method comprises the steps that a server obtains first camping data of a first enterprise, wherein the first camping data at least comprises financial information, personnel information and stakeholder information; And decomposing the first camping data according to different types of dimensions based on the first time sequence interval to obtain a first structure tree and a first extraction table of time and types of multistage dimensions, wherein the first structure tree and the information in the first extraction table are correspondingly arranged. Optionally, in one possible implementation manner of the first aspect, the receiving, by the server, the second operation data of the second enterprise added by the analysis end to classify the second operation data into a second structure tree and a second extraction table, extracting operation specification information of the second structure tree, and performing partition processing on the first structure tree and the first extraction table based on the operation specification information to obtain a plurality of information areas, where the method includes: Decomposing the information of each dimension in the second business data according to a preset analysis point to obtain business specification information corresponding to each dimension, wherein the business specification information at least comprises a time period; generating a second structure tree and a second extraction table after the second operation data is classified based on the dimension and the time period; And processing the first structure tree based on the second structure tree, screening to obtain a plurality of information areas, synchronously adjusting the first extraction table and the second extraction table, wherein each information area has a corresponding ca