Attribute Reduction Algorithm For Neighborhood Rough Sets And Its Application In Classifiers

In recent years, with the rapid development of Internet technology, a large number of companies began to focus on the data generated by the business. They have set up their own data warehouse to collect large amounts of data in order to obtain a decision support and create new value. But with the accumulation of data, the cost of information processing is also rising. And these data are often time-sensitive. If you do not get value from these, a lot of resources will be wasted. So how to acquire knowledge in an efficient way has become a new hotspot from large amounts of data.Neighborhood rough set can not only handle imprecise, inconsistent and incomplete data, but also has a direct role in continuous data. Therefore, in order to obtain higher efficiency, this paper will be based on neighborhood rough set and two aspects, dimensionality reduction and classification, will be improved.The main task of this paper is as follows:1. In the neighborhood rough set, attribute reduction is a way of dimensionality reduction. A particular analysis about the complex operation of calculating the neighborhood elements of each record in neighborhood rough set model is given. Then this paper also describes how to deduce a new property by the existing properties, and using the new property improves the existing attribute reduction algorithm to reduce the number of computing neighborhoods. Experimental results show that the improved algorithm is effective. 2. Through the existing classification algorithms research and analysis, this article also describes a classification algorithm based on neighborhood rough set. This algorithm can determine the category by gradually adding attributes. Meanwhile, the accuracy of this algorithm is similar to that of k NN, but can greatly improve the speed of classification algorithm.