Mining Educational Data Using Mknn Classification To Decress The Dropout Rate In Technical Educational

To increase the student graduation rates and decrease the student dropout rate some predictive measures should be taken. This can be achieved through data mining methods. Educational Data mining concerns with developing methods for discovering knowledge from data that come from educational domain. Attribute selection is done by MATLAB; the data processing tool mainly works in prediction and classification of knowledge. Here Modified K-Nearest Neighbor, k means, Gaussian Mixture Model, fuzzy c means, classification is used to classify and neural network is used to compare the MKNN, GMM, Fuzzy C means classification algorithms and predict the algorithm which is highly accuracy and time computing process. The proposed KNN classification is called Modified K-Nearest Neighbor (MKNN). The main idea is to classify an input query according to the most frequent tag in set of neighbor tags. MKNN can be considered a kind of weighted KNN, so that the query label is estimated by weighting the neighbors of the query. The procedure computes the frequencies of the same labeled neighbors to the total number of neighbors.