Abstract
The skin dehydration level can be used to infer serious health conditions in patients since diseases like cardiovascular abnormality, diabetes and cancer symptoms do exhibit correlation with skin disorders. Therefore a systematic analysis of human skin hydration levels is critical for multiple health care applications. Motivated by this, in this study we proposed a unique approach of measuring body hydration levels against different body postures using skin conductance electrical activity. In this paper, we report the collection, processing and analysis techniques used in the analysis of skin conductance data. Subsequently in order to predict body hydration levels we employed state-of-the-art machine learning models using the skin conductance data and achieved 81.82% and 73.91 % recognition accuracy for the data of standing and sitting postures,respectively using KNN model.