EXECUTION AND EVALUATION OF K-NEAREST NEIGHBOR FOR IDENTIFICATION AND VISUALIZATION OF BREAST MALIGNANT GROWTH
Gaurav D Saxena1, Dr. Shaik Jumlesha2, K Susmitha3, Morukurthi Sreenivasu4,
Uppalapu Vinod Kumar5, Gottala Surendra Kumar6
1Department of Computer Science, Kamla Nehru Mahavidyalaya, Nagpur, Maharashtra, India
2Professor, Department of Computer Science and Engineering, Annamacharya Institute of Technology and Sciences, Tirupati, India
3Assistant Professor, Department of Computer Science and Engineering, Annamacharya Institute of Technology and Sciences, Tirupati, India
4Associate Professor, Department of Information Technology, GIET Engineering College, JNTUK, Kakinada, Andhra Pradesh, India
5Assistant Professor, Department of Computer Science and Engineering, GIET College of Engineering, JNTUK, Kakinada, Andhra Pradesh, India
6Assistant Professor, Department of Computer Science and Engineering, Shri Vishnu Engineering College for Women, Bhimavaram, India
gauravsaxena@kamlanehrucollege.ac.in1, ahmedsadhiq@gmail.com2, susmitha.karanam10@gmail.com3, msreenivasucse@giet.ac.in4, uvinodkumarcse@gmail.com5, gsurendrakumarcse@svecw.edu.in6
ABSTRACT
Breast malignant growth starts as a sluggish developing irregularity or cancer that begins from the milk channel cell lining. Breast malignant growth can either be obtrusive or not. Harmless breast malignant growths can’t attack other breast tissues; however obtrusive breast diseases can go from the milk conduit or lobule to other breast tissues. The thickness and mass of the breast are consistent in size and structure because of their heterogeneity. In this paper, the K-Nearest Neighbor technique is utilized to break down the Breast Malignant growth Wisconsin (Diagnostic) dataset from the UCI machine learning repository. For the distinguishing proof of Breast threatening development, support vector machines, K-Nearest Neighbors, random forest, calculated relapse, and a combination of various philosophies can be used. Regardless, for a surprisingly long time, we just used the K-Nearest Neighbor approach for getting ready and assurance, which is a controlled AI calculation. The most restricted way between the model point and the arrangement discernments in the dataset is all not completely permanently established in that frame of mind of social event the data (dataset). Directly following executing all of the cycles as indicated by K-Nearest Neighbor computation on the foreordained dataset, we acquired results with 95.21 percent Accuracy.
KEYWORDS: Breast Cancer, Breast Cancer Diagnostic Dataset, Confusion Matrix, K- Nearest Neighbor, Machine Learning.