International Journal of Innovative Technology and Exploring Engineering (IJITEE) ISSN: 2278-3075, Volume-9 Issue-3, January 2020 3489 Published By: Blue Eyes Intelligence Engineering & Sciences Publication Retrieval Number: B7537129219/2020©BEIESP DOI: 10.35940/ijitee.B7537.019320 Stress Detection Methodology based on Social Media Network: A Proposed Design S. M. Chaware, Chaitanya Makashir, Chinmayi Athavale, Manali Athavale, Tejas Baraskar Abstract: Mental disorders can be recognized by how a person behaves, feels, perceives, or thinks over a period of a lifetime. Nowadays, a large number of people are feeling stressed with the rapid pace of life. Stress and depression may lead to mental disorders. Work pressure, working environment, people we interact, schedule of the day, food habits, etc. are some of the major reasons behind building stress among the people. Thus, stress can be detected through some conventional medical symptoms such as headache, rapid heartbeats, feeling low energy, chest pain, frequent colds, infections, etc. The stress also may reflect in normal behavior while carrying out day-to-day activities. Individuals may share their day-to-day activities and interact with friends on social media. Thus, it may be possible to detect stress through social network data. There are many ways to detect stress levels. Some of the instruments are used to detect stress while there is a medical test to know the stress level. Also, there are apps that analyze the behavior of the person to detect stress. Many researchers had tried to use machine learning techniques including the use of various algorithms such as Decision Tree, Naïve Bayes, Random Forest, etc. which gives a lower accuracy of 70% on average. In this paper, we are using a closeness of stress levels with social media data shared by many users. In our proposed system design, Facebook posts are being accessed using a token. Further, we recommend the use of machine learning algorithms such as Conventional Neural Network (CNN) to extract Facebook posts, Transductive Support Vector Machine (TSVM) to classify posts and K-Nearest Neighbors (KNN) to recommend nearby hospitals. With the help of these algorithms, we predict the stress level of the person as positive, negative. Thus, we are expecting more accuracy to detect the stress along with the preventive recommendation. We have proposed a methodology to detect stress because severe stress may lead to self-harming activities and also it may affect the lives of people around us. Thus, stress detection has become extremely important and we are expecting that our proposed model may detect it with more accuracy. Keywords: Social Media, Mental Disorder, Conventional Neural Network, Transductive Support Vector Machine, K- Nearest Neighbors, Facebook I. INTRODUCTION Mental disorders are threatening people’s health. They are considered to be a major factor of change the mood of a user and the user goes into a depression. Nowadays users can be stressed due to social interactions of social networks. The rapid increase of mental disorders or stress has become a great challenge to human health and quality of life. Revised Manuscript Received on January 06, 2020. S. M. Chaware, Department of Computer Engineering, Savitribai Phule Pune University. Chaitanya Makashir, Department of Computer Engineering, Savitribai Phule Pune University. Chinmayi Athavale, Department of Computer Engineering, Savitribai Phule Pune University. Manali Athavale, Tejas Baraskar, Department of Computer Engineering, Savitribai Phule Pune University. It is difficult to timely detect mental disorders or stress for proactive care. Thus, there is significantly important to detect mental disorder before it turns into severe problems. Our proposed design join hands to detect stress to avoid further consequences such as going into depression, self- harming acts, etc. Once stress is detected, people can take the help of stress management methodologies such as meditation, ‘smile and laugh’, reading motivational books, etc. A person can also follow proper treatment suggested by doctors, consultants. But for this, there is a need to suggest nearby hospitals so that a person gets help as quickly as possible. There are also some techniques that are implemented to detect the mental state of mind using different machine learning algorithms. For this, real-world social media data has been analyzed. But algorithms like Decision Tree, Naïve Bayes, Random Forest failed to achieve expected accuracy. These algorithms gave an approximate accuracy of 70%. II. LITERATURE REVIEW Nowadays people are constantly using social media to reflect their lives over the internet. Social media platforms like Facebook, Twitter, Snapchat, Instagram, LinkedIn, Tumblr, Pinterest, etc. engage people more than one-to-one human interactions. Though social media has provided a platform to facilitate the sharing of thoughts, feelings, career interests, etc. on the internet, unfortunately, it’s overuse leads to addiction to social media and stress. Research says that symptoms of mental disorder can be noticed from interactions over social media so that delays in treatment can be avoided. The emphasis is on Cyber- Relationship addiction, Net compulsion, Information overload to detect social network mental disorders [1]. Features like social relationships, self-disclosure or self- esteem, loneliness, bursting temporal behavior, etc. are analyzed. To build the SNMD-based Tensor Model, the Transductive Support Vector Machine (TSVM) is used that gave an accuracy of 84.3%. Mining online social behavior provides an opportunity to detect mental disorders based on features extracted from data logs of online social networks [1].The main emphasis of previous studies is on the classification of emotions of tweets, posts gathered from social media platforms like Twitter, Facebook. This is because these platforms are the most frequently used platforms. Preprocessing includes classification of a dataset into a training dataset and testing dataset to carry out tokenization further [2]. Next to it, pre-processing of tweets is done which includes removing handles, removing URLs, timings of tweets, #hashtag, etc. [2]. Support Vector Machine (SVM) and Decision tree algorithms are implemented to obtain positive or negative results.