Weighted KNN classification method and system based on support vector machine

Abstract

The invention discloses a weighted KNN classification method based on a support vector machine, and the method comprises the steps: carrying out classification of an original training set through thesupport vector machine, obtaining a weight factor related to each attribute in a feature vector, and enabling the weight factor to represent the correlation between the feature attribute and the classification; introducing the weight factor into the calculation of the distance between the input point and the training set, calculating the most suitable neighbor number and the weight factor of eachinput point by using a KNN method, and finally judging the category of the input point by using a local weighting method. When more feature attributes irrelevant to classification exist in the feature vectors, the dimensionality of the data can be effectively reduced, and the classification precision can be improved. The method is particularly suitable for classification of complex data and massdata, and compared with other classification algorithms, the time complexity is reduced while the classification accuracy is improved. The invention further comprises a system for implementing the weighted KNN classification method based on the support vector machine. The problem that the effect becomes poor when a traditional KNN algorithm faces uneven data set distribution is solved.

Description

technical field [0001] The invention relates to knowledge in the field of data mining such as support vector machines and KNN algorithms, in particular to a support vector machine-based weighted KNN classification method and a system thereof. [0002] technical background [0003] With the further acceleration of the development of information technology, the desire for informatization in all walks of life is becoming increasingly urgent, and informatization construction is in full swing. In the context of informatization, the amount of data generated by the country or enterprises is constantly expanding, and effective data analysis and data mining methods will bring great benefits to the country, society and enterprises. In the era of data explosion, machine learning and artificial intelligence have attracted great attention from the whole society. A large number of data mining technologies have been applied to user analysis, financial investment, medical and health industri...

AI Technical Summary

Problems solved by technology

[0004] The traditional KNN classification method becomes worse when the data set is unevenly distributed. If the accuracy is improved, the complexity of the algorithm will be greatly increased.

Images