Rapid data mining method and system based on construction model

Abstract

The invention discloses a rapid data mining method and system based on a construction model. The method comprises the steps of constructing a model and quickly mining based on the model. The quick mining process comprises the following steps: performing correlation matching on a query word and a pre-stored result word; establishing a graph storage structure between the query words and the result words by utilizing the incidence relation between the query words and the result words of each data source; constructing a network topology relation graph based on the storage structure of the graph; and outputting a data mining result according to different application requirements input by a user to enable the user to select an output result so as to trigger display of output information. According to the invention, the data mining model is established according to the data mining requirements, so that the accuracy and reusability of data mining are ensured; the method comprises the followingsteps: establishing a network topology relationship in an association relationship between a data source query word and a result word and in a storage structure of a result graph, and constructing asound data mining network to realize multi-dimensional comprehensive data mining; improving data mining efficiency.

Description

technical field [0001] The invention belongs to the technical field of data mining, in particular to a fast data mining method and system based on building a model. Background technique [0002] With the rapid development of Internet technology, the amount of information released by major manufacturers on the network is also increasing. Therefore, it is necessary to study data mining technology so that customers can find the information they need from a large amount of network information. . [0003] In the prior art, after the user creates the training model, it can be determined whether the training model process is accurate through the final output or mining model verification, and whether the entire modeling process is reasonable can be determined through the final matching. Once there is a problem in the intermediate steps, there is no way to verify and give feedback in stages. The reusability and portability are very poor and cannot be standardized. [0004] Moreove...

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to provide a fast data mining method and system based on building a model, by establishing a data mining model according to data mining requirements, to ensure the accuracy and reusability of data mining; The association relationship and the result graph establish the network topology relationship with the storage structure of the graph, build a sound data mining network, and realize multi-dimensional comprehensive data mining; solve the problem of poor reuse of existing data mining, inability to standardize, and inability to mine data in multiple dimensions. incomplete question

Images