Data mining method, device and equipment based on Bayesian classification algorithm

Abstract

The invention relates to a data mining method, device and equipment for a Bayesian classification algorithm, and the method comprises the steps: obtaining text information, carrying out the preprocessing of the text information, obtaining target text information, processing the target text information, obtaining a text vector of the target text information, inputting the text vector into a Bayesian classification model, and obtaining a Bayesian classification model. Obtaining a category of the text information, and counting high-frequency words of each category in the category to obtain a high-frequency word set of each category; according to the method, the problems that text information in the current internet is redundant, unclassified, too rough and lack of guidance value in the prior art are solved, the data are quickly and efficiently mined and displayed, and the potential value of the data is exerted.

Description

technical field [0001] The invention relates to the technical field of data mining, in particular to a data mining method, device and equipment based on a Bayesian classification algorithm. Background technique [0002] Data mining involves many subject areas and methods, and the main task is to mine useful information from massive information or data. With the development of information technology and smart devices, more and more text data are generated in cyberspace, and the problem of text information overload is increasing. At present, we can easily and quickly obtain a large amount of information, the frequency of obtaining information has increased, and the difficulty of mining key information has also increased. Especially in the e-commerce applications that have emerged with the development of the times, online shopping users can complete the purchase of goods without leaving home, but it is usually difficult to distinguish the quality of the goods. At this time, us...

AI Technical Summary

Problems solved by technology

[0003] The current data mining algorithms include methods such as Bayesian, decision tree, neural network, and association analysis. Decision tree is a basic classification and regression method. In classification problems, it represents the process of classifying instances based on features. Generally, They are all generated from top to bottom, representing a collection of if-then rules. Its advantage is that it is intuitive and generates understandable rules. However, as the complexity of data increases, its branch trees will also increase. Management difficulty

Moreover, it is difficult to discover rules based on multiple variable combinations, and it is difficult to predict continuous fields

The neural network can process continuous signals, has good self-adaptation, self-organization and strong learning functions, and has an information processing mode closer to the human brain. It has many applications in the fields of natural language understanding, image recognition and speech recognition, but Its hidden layer is a black box to the outside world, the factors for drawing conclusions are not obvious, and the interpretability of the output results is poor, which affects people's acceptance of the results

The advantage of the correlation analysis method is that it can produce clear results, and the processing process can be seen, but the data with low confidence in this method may also reflect very important market information, which will be ignored

Images