#support vector machine

Diabetes Mellitus is a disorder in metabolism of carbohydrates, and due to lack of the pancreatic hormone insulin sugars in the body are not oxidized to produce energy. Diabetic Retinopathy is a disorder of the retina resulting in impairment or vision loss. Improper blood sugar control is the main cause of diabetic retinopathy. That is the reason why early detection of retinopathy is crucial to prevent vision loss. Appearance of exudates, microaneurysms and hemorrhages are the early indications. In this study, we propose an algorithm for detection and classification of diabetic retinopathy. The proposed algorithm is based on the combination of various image processing techniques, which includes Contrast Limited Adaptive Histogram Equalization, Green channelization, Filtering and Thresholding. The objective measurements such as homogeneity, entropy, contrast, energy, dissimilarity, asm, correlation, mean and standard deviation are computed from processed images. These measurements are finally fed to Support Vector Machine and k Nearest Neighbors classifiers for classification and their results were analysed and compared. 

Face identification is a fundamental report field of example recognition. Today, it has created decent enthusiasm for researchers in these fields, similar to PC vision and example identification. We will affirm that crafted by the face acknowledgment framework is chosen by the best approach to portion factor vector absolutely and to convey them into a class appropriately. A technique perfection to help features acknowledgment pace by intertwining the part and size of Gabors delineations of the features as a fresh out of a box new portrayal, inside the spot of the arrangement pictured, however, the physicist portrayals were generally utilized, strikingly inside the calculations on worldwide methodologies, the physicist part was never misused, trailed by a face acknowledgment algorithmic principle, upheld the important part investigating approach and Support Vector Machine is utilized as a shiny advanced classifier as design acknowledgment. The presentation for the anticipated algorithmic principle is to try the overall population. It is generally utilized databases of Face Recognition is Grand Challenge v2 face and ORL databases. The test domino effect on databases shows that the blends to the greatness on the piece of physicist choices can do talented outcomes. 

Theoretical Classification of information has become a significant research region. The way toward ordering archives into predefined classifications dependent on their substance is Text characterization. It is the mechanized task of common language writings to predefined classifications. The essential prerequisite of content recovery frameworks is content characterization, which recover messages because of a client inquiry, and content getting frameworks, which change message here and there, for example, responding to questions, creating outlines or removing information. In this paper we are concentrating the different grouping calculations. Order is the way toward isolating the information to certain gatherings that can demonstration either conditionally or freely. Our fundamental point is to show the examination of the different characterization calculations like K nn, Na¯ve Bayes, Decision Tree, Random Forest and Support Vector Machine SVM with quick digger and discover which calculation will be generally reasonable for the clients. 

Skin cancer is one of the major types of cancers with an increasing incidence over the past decades. Accurately diagnosing skin lesions to discriminate between benign and skin lesions is crucial.J48 Algorithm and SVM SUPPORT VECTOR MACHINE based techniques to estimate effort. In this work proposed system of the project is using data mining techniques for collecting the datasets for skin cancer. So that system can overcome to diagnosing the disease quickly and accuracy. Comparing to other algorithm proposed algorithm has more accuracy. When we have to using two kind of algorithm .They are J48, SVM. J48 Algorithm produced better accuracy more than SVM algorithm. The accuracy of the proposed system is 90.2381 . It means this prediction is very close to the actual values.