Machine Learning and Natural Language Processing Usage for Psychological Consultation

Abstract

It`s an obvious fact to claim, that the problem of stress has become a vital issue presently. One of the main root cause of this is the escalation of human desires and complexity of those. Most of those desires would be difficult to achieve or not practical at all. When the reality is so harsh, compared with the imaginations, it will incur for stress. More the gap between reality and the perceptions, stress will escalate. As results of people are running after unrealistic or difficult to achieve greener pastures, most of them will end up with becoming a victim of stress. Stress management is a difficult skill to be developed, but in current context, it has become an essential skill to have. This research is based on the concept of internal self-talk. Thought stream captured in-form of text stream will be segmented, according to the cognitive behavioral therapeutic approach. This is technically implemented via POS tagging of the Stanford NLP library. Afterwards machine learning approach is used to train the WEKA engine, according to the OCEAN model, which is a prominent psychological model. Predictions derived from the trained WEKA model, will be presented inform of a report with the help of itext reporting plugin. This report will be used by the psychologist, before providing the treatment to the patient/client. It`s assumed, that this tool will be a good aiding tool, which can reduce the cognitive effort of a consultant. Respective, problem, technical, executional and all important aspects are addressed in detail, within this paper along with required evidences.