Cervical cancer is a dangerous disease that occurs in advancing sovereignty where public awareness is low. Screening Papnicolaou, or Pap test, is the most common test to detect cancer, which develops in the uterus and occurs in most women. Image processing algorithms play a leading role in the segmentation of the cancerous region in cervical images. The cancer region is segmented in cervical cancer images by fuzzy support vector machine (FSVM) algorithm. The cervical cancer region was separated from background images. The k-means classification algorithm is an existing algorithm applied to cervical cancer images. The existing and proposed segmentation results are compared using quality measurement techniques such as accuracy and precision. From this evaluation, proposed algorithm is provided the highest accuracy (98%) when compared to the previous algorithm. In this chapter, we described about the machine learning models, determination of the cancer region and collected the data from the cervical images for diagnosis of cancer level using support vector machine (SVM) classifier and k-nearest neighbours (KNN) classifier and classify it is a benign or malignant stages. The accuracy, sensitivity, and specificity are which calculated for classification accuracy. The accuracy is 97 percentages is gained using SVM compared to the KNN.