Social Media Depression Monitoring Model Using Sentiment Analysis

IRJET, 2021

More than three hundred million folks are littered with depression everywhere in the globe. Mental illness detection in social media can be considered a difficult task, mainly due to the complicated nature of mental disorders. In recent years, this research area has started to evolve with the continuous increase in popularity of social media platforms such as Instagram, Twitter, and Facebook, etc that became an integral part of people's life. This close relationship between social media platforms and their users has made these platforms reflect the users' personal life with different limitations. Social networks have been developed as a great platform for its users to communicate with their friends and share their opinions, photos, and videos reflecting their moods, feelings, and sentiments. This creates an opportunity to analyze social network data for user's feelings and sentiments to investigate their moods and attitudes when they are communicating via these platforms. In this paper, we have proposed a depression analysis and detection system. The present study aims to exploit natural language processing and machine learning techniques for analyzing and detecting of depression level of the Twitter user. For this purpose, we used NLP techniques on to the textual data extracted from Twitter to analyze and identify the emotions in the data. We used machine learning classifiers and trained and tested classifiers to analyze the extracted tweets from Twitter and evaluate the tweets to classify whether the user is going through depression or not. All this are done using the Predictive approach for early detection of depression or other mental illnesses.

Nowadays risk of early death is increasing due to mental illness which is mostly caused due to depression. Depression creates suicidal thoughts causing serious impairments in daily life. Sentiment analysis is a hot topic that's been on research for decades, which intends to find the nature of text and classifies into positive, negative and neutral. In today's digital world lot of data is available for sentiment analysis of both image and text. The aim of this paper is to apply natural language processing on Twitter feeds for conducting emotion analysis focusing on depression. Based on a curated word list, the tweets are classified as positive or negative. Naive-Bayes and support vector machine algorithms are used for the classification of the text data obtained from twitter feed. The results are presented using the classification metrics like accuracy, precision, recall and confusion matrix.

ArXiv, 2020

Text sentiment analysis for preliminary depression status estimation of users on social media is a widely exercised and feasible method, However, the immense variety of users accessing the social media websites and their ample mix of vocabularies makes it difficult for commonly applied deep learning-based classifiers to perform. To add to the situation, the lack of adaptability of traditional supervised machine learning could hurt at many levels. We propose a cloud-based smartphone application, with a deep learning-based backend to primarily perform depression detection on Twitter social media. The backend model consists of a RoBERTa based siamese sentence classifier that compares a given tweet (Query) with a labeled set of tweets with known sentiment ( Standard Corpus ). The standard corpus is varied over time with expert opinion so as to improve the model's reliability. A psychologist ( with the patient's permission ) could leverage the application to assess the patient&#3...

INTERNATIONAL SCIENTIFIC AND PRACTICAL CONFERENCE “TECHNOLOGY IN AGRICULTURE, ENERGY AND ECOLOGY” (TAEE2022)

Depression is one of the major causes of increased number of cases of mental illness and suicides all over the globe now a days. A person suffering from anxiety, sadness, depression or suicidal thoughts finds it easier to express his emotions on social media platforms. Thus, the messages or content shared by a person on social media platforms is the best way to detect the mental condition of a person by analysing these messages. Today's situation of pandemic covid19 also increased the cases of depression. In this paper we have used Natural Language processing techniques and deep learning methods to create a model to predict such mental conditions like depression. The model created using LSTM-CNN gives better accuracy of 97% when compared to the other base models of logistic regression, Naïve Bayes, Random Forest and Decision Tree.

Indian Journal of Science and Technology, 2018

Objective: Online social and news media has turned into an extremely mainstream for clients to impart their insights. The objective of this paper is to propose a methodology through which sentiments can be analyzed. Methods/Statistical analysis: The sentiments are helpful for the identification of the depression. In this paper we proposed an algorithm through which tweets are extracted from twitter using R studio and then their sentiments are analyzed i.e. the scores are given to each sentiment by which we identify whether the person is depressed or not. This gives imperative data to basic leadership in different spaces. Findings: Sentiment analysis over Twitter offers associations and people a quick and powerful approach to screen the general population's sentiments towards them and their rivals. To evaluate the assumption examination over twitter we need a dataset that is been extracted from the twitter that would be publicly available for twitter sentiment analysis. We found that through twitter to extract tweets are scored based on their sentiments. The result is unique as we have proposed new algorithm through which twitter sentiments are scored. Applications: This sentiment analysis will be helpful to draw conclusion, whether the person is depressed or not. It can be helpful for prescreening test, diagnostic tool and automation monitoring system.

Indian Journal of Natural Sciences, 2025

A lot of people have depression, which is a mental health problem. Recent statistics show that about 3.8% of the population, including 5% of adults, suffers from depression. Face-to-face interviews and other traditional approaches of finding out if someone is depressed have their limits and don't always work. But social media sentiment analysis could help find people who are likely to get depressed. This essay will talk about how common depression is in society, the problems with traditional ways of finding depression, the possible uses of social media for finding depression, the difficulties of using social media sentiment analysis to find depression, and how well social media sentiment analysis works for finding depression.

SSRN Electronic Journal, 2019

Social media platforms like Twitter which is a microblogging tool enable its users to express their feelings, emotions and opinions through short text messages. Detecting the emotions in a text can help one identify anxiety and depression of an individual. Depression is a mental health problem which can happen to anyone, at any age. There is a lack of systematic and efficient methods to identify the psychological state of an individual. With more than 58 millions tweets generated daily, Twitter can be used in order to detect the sign of depression in a faster way. Recent studies have demonstrated that Twitter can be used to prevent one from taking an extreme step. Our Proposed depression detection and prevention system can detect any depression related words or phrases from Tweets and also classify the type of depression, if detected. This system is proposed in order to diagnose depression and prevent it. Proposed system is using Support vector machine and Naïve Bayes classifier. This hybrid approach works well not only with shorter snippets but also with longer snippets.

Universal Journal of Public Health, 2022

At present the pandemic situation caused by Corona virus syndrome 2019 which is abbreviated as COVID-19, has made the lifespan of men at stake. Not only has it affected the economic condition and created health hazards of the peoples all over the world but also it tells upon their mental state. It is surprising but very difficult to assess the frequency. There are various reasons behind this mental disorder. They are recession from work, confined in house strategy, getting afraid of corona virus, and some more. In this paper, we focus on the use of Natural Language Processing (NLP) procedure to analyze tweets with regard to mental state. Training of significant prototypes has been provided to categorize all tweets into the emotions mentioned below: covid19, covid, COVID-19, covid 19, flu, virus, hantavirus, fever, cough, social distance, lockdown, pandemic, epidemic. We build the EmoCT (Emotion-Covid19-Tweet) dataset to train physically, tagging 1,500 English tweets. In addition to it, two procedures are suggested and distinguished to explore the causes which are creating melancholy and disquiet.

The use of Social Networking Sitesis increasing nowadays, especially by the new generations. The availability of SNS allows users to express their interests, feelings and share daily routines. Many researchers prove that using user-generated content correctly may help determine people's mental health. Mining the UGC could help to predict mental health levels and depression. Depression is a serious mental illness. However, from the user profile, we can collect all the information that relates to a person's mood. In this research, we decided to search how users' posts can help classify users according to mental health levels. We propose a system that uses networking sites as a source of data and screening tool to classify the user using artificial intelligence according to the UGC on SNS. We created a model that classifies the UGC using two different classifiers: Support Vector Machine (SVM), Naïve Bayes,and Transformer.

International Journal of Engineering and Advanced Technology, 2020

Today Micro-blogging has become a popular Internet-user communication tool. Millions of users exchange views on different aspects of their lives. Thus micro blogging websites are a rich source of opinion mining data or Sentiment Analysis (SA) information. Due to the recent emergence of micro blogging, there are a few research works devoted to this subject. We concentrate in our paper on Twitter, one of the prominent micro blogging sites to analyze sentiment of the public. We'll demonstrate, how to gather real-time twitter data for sentiment analysis or opinion mining purposes, and employed algorithms like Term Frequency - Inverse Document Frequency (TF-IDF), Bag of Words (BOW) and Multinomial Naive Bayes ( MNB). We are able to determine positive and negative sentiments for the real-time twitter data using the above chosen algorithms. Experimental evaluations below shows that the algorithms used are efficient and it can be used as a application in detection of the depression of t...