An Evaluation of Classification Techniques for Depression, Anxiety and Stress Assessment
S. T Arokkiya Mary, L Jabasheela
S. T Arokkiya Mary
Available Online February 2018.
- https://doi.org/10.2991/pecteam-18.2018.13How to use a DOI?
- In the past decades, Depression, Anxiety and Stress (DAS) became a serious health issue in the society over all parts of the world. It can be measured by Depression, Anxiety and Stress Scale (DASS) which consists of a set of questionnaires. Several techniques are proposed to validate the DASS level and to identify the levels of severity. The goal of this paper is to predict the DAS level using 5 different classifiers namely Logistic Regression (LR), Multilayer Perceptron (MLP), J48, Reduces Error Pruning (REP) Tree and Classification and Regression Trees (CART) algorithm. The major intention of this study is to compare the performance of these 5 classifiers based on classification accuracy, kappa value, precision, recall, F-score and ROC. We assessed the classifiers of the DASS-21 version suing the samples collected from 600 students from Puducherry, India. For experimentation, WEKA is used as a simulation tool and the results reveal that the MLP achieves better performance when compared to other classifiers. MLP attains the classification accuracy of 90.33%, 92% and 90.33% for Depression, Anxiety and Stress dataset respectively.
- Open Access
- This is an open access article distributed under the CC BY-NC license.
Cite this article
TY - CONF AU - S. T Arokkiya Mary AU - L Jabasheela PY - 2018/02 DA - 2018/02 TI - An Evaluation of Classification Techniques for Depression, Anxiety and Stress Assessment BT - International Conference for Phoenixes on Emerging Current Trends in Engineering and Management (PECTEAM 2018) PB - Atlantis Press SP - 64 EP - 69 SN - 2352-5401 UR - https://doi.org/10.2991/pecteam-18.2018.13 DO - https://doi.org/10.2991/pecteam-18.2018.13 ID - ArokkiyaMary2018/02 ER -