Published Paper Details:

Data mining is the process of extracting valuable information from a large data source and this is the primary step in the process of knowledge discovery. One among the best classification algorithm for classification is K-nearest neighbors (kNN) for the process of knowledge discovery. Inspired by the huge positive reviews about the tree-based methodology over the last decades, we try to develop a new method for data search under random projection forests (rpForests). Here the random forest projection method is mainly used to search the specified document randomly from the several groups and try to find out the document which is found or not. As we are using K-NN algorithm if the file is not found in one group the same file can be immediately search in opposite group in random manner and if the file is found the search will be immediately stopped at that point and the result will be displayed immediately. If the file is not found then the search continues till the end and the process terminates at the final stage if the file is not present in the entire groups.By conducting various experiments on our proposed model we finally came to a conclusion that our proposed approach is best in finding the solution to search the documents in random projection manner by using K-NN in an accurate manner.