Hair analysis based on medical history and spatial-temporal data

Abstract

Over the course of time, machine learning has improved the data analysis technique such as face detection and recognition. Many machine learning researches have been implemented in medical treatments. This concept which is proposed is inspired from different aspects of hair scalp and other factors. Spatial-temporal data is very useful in weather forecasting and satellite image analysis. This  technique is implemented to capture necessary data from hair follicle images. Hair is also a subject of human body. There are many factors which can be used to determine health of hair. All these factors including spatial-temporal images, gender, age and hair style are used to predict health of hair. This paper presents machine learning algorithm for analysis of medical data for determining health of hair.  We use the SVM (support-vector machines) model classifier for analysis of data. After  that, we get values such as short, straight, wavy and curly.In this paper, J48 Algorithms were used to obtain an accurate result compared with other algorithms.J48 with bagging is creating different decision trees for same data that why it is given more accurate results, J48 algorithms will split continuous values through using threshold. This paper 1066 samples were tested using cross validation technique, according to the test, it is found 87.14 % was a correctly classified and 12.85 % was incorrectly classifier. So at the end we get a real time performance is 89.5 %. This paper proves the compatible between hair style and Age-Gender.