Data asset analysis method and system based on XGBoost regression and convolutional network, and medium

Abstract

The invention discloses a data asset analysis method and system based on XGBoost regression and a convolutional network, and a medium. The method comprises the following steps: acquiring data asset information to form a data link to obtain an image type data asset feature set and a non-image type data asset feature set; performing regression analysis by taking the non-image data asset feature set as input of an XGBoost algorithm to obtain a first result; taking the image type data asset feature set and the non-image type data asset feature set together as the input of the XGBoost algorithm to carry out regression analysis to obtain a second result; and calculating the target deviation ratio, and comparing the target deviation ratio with a preset ratio threshold value to output a final result. According to the invention, common processing of image type data assets and non-image type data assets can be realized, so that the range coverage of the analyzed data assets is wide; meanwhile, the accuracy and reliability of data asset analysis can be further ensured by setting a deviation value; and an XGBoost algorithm is utilized to expand the supervision right and strength of data asset value analysis.

Description

technical field [0001] The present invention relates to the technical field of data analysis and processing, and more specifically, to a data asset analysis method, system and medium based on XGBoost regression and convolutional network. Background technique [0002] With the rapid development of big data and artificial intelligence, today's data not only provides innovation for science, but also directly creates wealth—data generates a huge economic scale in the process of flow and operation: 8% of the EU's GDP comes from data produced in. There is a growing recognition that data is not just a resource but an asset. However, data assetization faces many technical difficulties. One is the ambiguity of data property rights: data is reproducible, and today’s Internet generally follows the unspoken rule of "who collects who owns", which makes problems of privacy violations and data leakage abound; in addition , because data can be circulated and used through the network, it i...

AI Technical Summary

Problems solved by technology

However, data assetization faces many technical difficulties. One is the ambiguity of data property rights: data is reproducible, and today’s Internet generally follows the unspoken rule of "who collects who owns", which makes problems of privacy violations and data leakage abound; in addition , because data can be circulated and used through the network, it is difficult to define the ultimate owner of the data; the second is the problem of data islands: the security and attribution of circulating data are difficult to guarantee, so the non-circulation of data has become people's choice, thus forming Data islands reduce the validity and value of data; the third is the difficulty of pricing and valuation of data: due to the non-competition and unlimited sharing of data, the potential value of data is large and the value chain is long; Different angles of demand for the same data make them use the same data to mine different values, making it difficult to statically and accurately determine the value of data

This kind of packaging and selling method mostly conducts data asset pricing based on factors such as information quantity and information quality, and cannot withdraw the value potential and real value of digital assets; another example is some e-commerce platforms based on the total number of consumers, the purchasing power of The three indicators of conversion power regard brand consumer data as assets. Through the full link perspective of data, consumer data assets can be evaluated, optimized, and operated, so that brand owners can intuitively see the corresponding consumers. assets, estimating their commercial value and using them to aid in their marketing decisions

However, this method ignores the characteristics of the fast update speed of digital assets and the need for dynamic adjustment of pricing; in addition, the indicators for determining the value of data assets need to be determined for different subjects, instead of solidly selecting three monotonous indicators to evaluate all assets; In addition, some existing data asset management platforms describe the characteristics of information assets through metadata and classify and manage them in the form of catalogs. However, the algorithm process is complicated, and customers still cannot see the internal structure of the "algorithm black box". The right to self-choice cannot be realized

Images