AJAGBE TOLA TAOFIK

Stress hurts both physical and mental health, and can ultimately increase the risk of developing serious health conditions, such as depression and stroke. It has caused a lot of issues in human living conditions with about 41% of adults worldwide reporting experiencing a lot of stress. In comparison, 61.97% were estimated to experience stress in Nigeria frequently. Ablution (water cleansing) has been identified as one of the ways to reduce stress. Although, there have been researches along this line but are they scientific? This work explores the claim and uses spectrophotometric analysis to investigate the effect of ablution on stress reduction. To review, investigate, and identify the strengths and weaknesses of the existing machine learning and deep learning models for investigating the effect of water cleansing on skin as regards stress reduction using spectrophotometric analysis.  A total of 1,216 subjects were recruited where 608 perform ablution regularly and the other 608 were non-ablution performing group. The age ranged between 18 to 71 with average age of 35 years old. Spectrophotometer was used to collect the Forehead, Cheek, Chin and Arm absorbance and reflectance values of the subjects. The spectrophotometric datasets were trained and classified with 3 different machine learning models Convolutional Neural Network (CNN), K-Nearest Neighbour (KNN) and Support Vector Machine (SVM).  Results show that the accuracy of the three models was not better off compared to an ensemble model, where 89.9% accuracy was generated for KNN, 95.1% for SVM and 88.8% for CNN as against our developed ensemble model SpectraKS-CNN which returned 96.4% accuracy. This suggests that an ensemble machine learning models returned higher accuracy of 98.3% in classification, especially when used with diverse datasets.