A method and device for vectorization and visualization processing of supply chain data

Abstract

The embodiment of this specification provides a supply chain data vectorization and visualization processing method and device. In the vectorization method, the enterprise master data is obtained, and internal subsidiary enterprise information and external business cooperation enterprise information are extracted from it. Using the feature word extraction model, the word segmentation is extracted from two kinds of information respectively. The two kinds of information and the corresponding word segmentation are respectively input into the text conversion model to obtain the corresponding first and second feature vectors. The obtained eigenvectors are fused to obtain the target eigenvector corresponding to the enterprise master data. Then, the target feature vectors of multiple pieces of enterprise master data are displayed on the interface with appropriate dimensionality reduction algorithms and clustering algorithms through visual interaction, and appropriate algorithm parameters are determined through observation. Finally, possible problem data can be located and cleaned directly based on the clustering results or by visual search. At the same time, the visual search view also supports version rollback and re-correction of previous search and modification records.

Description

technical field [0001] One or more embodiments of this specification relate to the field of data processing, in particular to a method and device for vectorizing and visualizing supply chain data. Background technique [0002] Data is the core and key to the success of an enterprise's digital transformation, and data quality will directly affect the authenticity and reliability of data analysis. With the continuous development of modern Internet technology, data empowerment brings more and more value to large-scale supply chain integration service group companies, and thus drives various business operations and innovative development of enterprises, improves enterprise management level, and leads enterprise transformation Upgrade and continuously create new economic value. From a practical point of view, large-scale supply chain integration service group companies have the characteristics of massive data volume, complex data environment, and potential data defects. They are...

AI Technical Summary

Problems solved by technology

However, due to the incomplete data operation and management system, the incomplete data management process between different subsidiaries, the incomplete implementation of the division of responsibilities, the imprecise checking mechanism in the process of data extraction, cleaning, conversion, collection and distribution, and the discovery of Failure to properly handle data problems in a timely manner may lead to a large amount of irregular redundant data, such as enterprise master data, including internal subsidiary information and external business cooperation enterprise (customer, supplier) information, data such as duplication of master data Quality issues have become a bottleneck restricting the development of digital transformation of enterprises

In the process of implementing digital transformation, if there is no effective way to standardize and deduplicate existing data, and if incremental data is not effectively checked, data quality problems will accumulate over time, which will seriously affect data mining, analysis and application, which has a greater impact on enterprise business operations and management

In the daily data operation and control of enterprises, the processing of these redundant data, such as using traditional cleaning methods, mainly relies on manual data review and processing one by one, so the low efficiency is only applicable to small data sets

For example, although the intelligent cleaning method is faster, the user cannot participate in the data processing execution process, and the reliability of the cleaning results cannot be guaranteed when dealing with complex data problems.

Images