Detection of mental health disorders in social media
Tagged: Computer Science
1. Background
Data security has recently risen to the top of the internet's priority list. An intruder gained access to the system in order to gain unlawful access to or usage of the network's information. An attack, a hacking effort, a packet sniffing attempt, or a data stop are all considered incursions. Attacks are designed to breach system security or networks in order to extort money, gain sensitive information, or for other unethical reasons (Kaloudi & Li, 2020). Intrusion occurs when malicious code is introduced into a computer's software, data, or logic, producing a variety of problems, including the capacity to supply and steal sensitive data from the institute, making it available to cybercriminals.
Due to the complexity of mental problems, detecting mental illness via social media might be challenging. This research field has started to develop as a result of the ongoing increase in popularity of social media platforms, which have grown to be an essential component of people's lives in recent years. As a result, social media platforms have evolved to reflect users' personal lives on various levels due to the close relationship between them and their users (Stavropoulos et al., 2018).
In such a situation, individuals reveal a great deal of information to researchers about their lives. For example, supervised machine learning methods like deep neural networks have not typically been used because of the difficulties in obtaining sufficient quantities of annotated training data, which is in addition to diagnosing mental illnesses using social media platforms. (González-Sanguino et al., 2020).
Figure 1: Social media Vs. mental health
Source: Acronym, (2022)
Due to these factors, most of the researchers select the best deep neural network design from a group of architectures that have been successfully used in natural language processing applications. For example, social networking site usage has become hazardous as a result of its quick surge in popularity. However, the development of social networks has an impact on consumers. For example, the social broadcasting stage provides various alternatives while also affecting mental health (Helbich, 2018). The symptoms of these mental illnesses are now frequently only passively observed, delaying therapeutic action. In this project, we propose that social network analysis enables early detection of social network mental disorders. Identifying social network mental disorders is challenging since the mental status cannot be quickly determined from online social activity logs (Ioannidis et al., 2018).
As social networking and messaging apps have exploded, online social networks (OSNs) have entered many people's daily lives. As a result, social networking services have become increasingly popular, leading to harmful usage—a growth in psychological issues, a dependency on cybernetic interactions, and information overload in social networks. As a result, OSNs seem to make it easier for users to connect with others, but they may make it harder for users to interact with others in person (Elbay et al., 2020).
Researchers have suggested that monitoring social media behaviour online enables early active detection of SNMDs. Online social activity records cannot reveal the mental components addressed in conventional diagnostic criteria (questionnaire), making it challenging to identify SNMDs. In addition, social networking websites have attracted the younger generation's attention over the past ten years. However, thanks to advancements in internet technology, it is becoming easier and more convenient to access and connect with these social networking sites on the other side of the planet (Lichtenstein et al., 2018).
Due to the large number of users that interact and socialise on social networking sites, these platforms have become a source of data mining. Additionally, it is feasible to glean considerable knowledge about the user's attitude, feelings, and negativity. In the meantime, despite having mental problems, many people have been using social networking sites to relax and get rid of their worries (Xie et al., 2018).
Figure 2: SNS information and Machine learning prediction
Source: Eurekalert (2022)
The majority of people rely on social networking sites (SNS) to find out the latest information and different points of view on a range of subjects, including sharing, emotions, interpretation, assessment, evaluations, methodologies, assumptions, factors that influence, viewpoints, and opinions, using a range of materials and marketing. YouTube, Instagram, and Snapchat are all less popular and have less widespread availability than Facebook. Additionally, compared to non-users, Facebook users appear to have a higher prevalence of depressive symptoms and make up the great bulk of Facebook users (Robinaugh et al., 2020).
The results show that people with this mental disease turn to social media for comfort by sharing their views because they desire to distance themselves from others. Social networking sites are widely used. Thus, there may be a chance to spot depression at levels that weren't there before. The regular usage of these social networking sites can reveal characteristics of users with mental diseases and people at risk for mental illness. One of the most significant issues nowadays is depression, spreading rapidly. Globally, depression is the main factor that causes mental illness (Li et al., 2020).
AI can manage vast volumes of data: AI can help to improve network security by creating self-contained security systems that can detect and respond to breaches. The number of security warnings that receive on a regular basis might be overwhelming for security professionals. Automatically identifying and responding to attacks has decreased the strain of network security experts and can help detect threats more effectively than earlier strategies (Chan et al., 2019).
Data mining employs various techniques to gather important information about a people's emotional and psychological state. Data and statistical reports can be obtained through a method called mining. These tactics help people comprehend how to manage data flow. Efficient mining techniques are able to organise enormous amounts of data. The information from these social networking sites may be enough to use the tools and methods offered by data mining techniques. A vast amount of data needs to be analysed in the meantime. This provides academics with a more sophisticated way of comprehending social information clearly (Vigo et al., 2020)..
1.2 Social Media and Mental Health
Almost everyone has heard of depression, which has emerged as a prominent topic of debate in the global community discussing public health issues. Observing their activities and behaviours makes it simple to recognise individuals struggling with many problems. The problems are regretful longing, lack of vigour, inferiority complex, empty mind, dread, distress, decreased or increased hunger, difficulty sleeping, guilty sentiments, self-harm, and suicidal thoughts. An inability to manage daily chores, let alone interpersonal connections, may result from all these feelings and thoughts. Therefore this may impact their capacity to live their lives in several ways, which is ultimately society's responsibility (Chua, 2018)
The effect of social media on mental health impedes the delivery of social services on a small scale due to the exponential growth of mental health symptoms. More caregivers are required as the number of individuals experiencing depression, anxiety, low self-esteem, and other problems due to social media usage grows. Mental health care professionals need to be aware of how social media affects mental health to serve persons impacted by this problem effectively. It also requires more instruction on evaluating social media use and its potential impact on mental health. Finally, a deeper understanding of the impacts of social media usage would enable clients to receive more effective and beneficial treatment.
Early detection and intervention can significantly shorten the course of treatment. Understanding the degree of depression may also aid with early prevention strategies. Professionals in mental health haven't done much to address the impact of social media on mental health. Furthermore, there hasn't been much research on how mental health professionals can handle the effects (Caspi & Moffitt, 2018).
However, numerous studies in the literature link social media use to a range of psychiatric issues, including depressive symptoms, anxiety, and low self-esteem. Users may encounter bullying, public shame, and severe reactions to their posts on social media. Comparing their self-image and level of life satisfaction to those of other users may also make them uneasy. Additionally, bad social media habits can cause mood swings, loneliness, and melancholy because of the unfavourable content users read through (Ho et al., 2020).
Figure 3: Healthy ways to cope with mental health using social media
Source: Cuncic (2021)
Social media is a powerful tool for establishing connections with loved ones and others online. Every innovation has benefits and drawbacks, but this one helps close the communication gap in society, encourages long-distance relationships with friends and family, and keeps individuals up to date on various events that are important to their jobs. When it comes to social media, the situation is the same. Social media has made communication simpler, but it has also led to a rise in the number of users of all ages addicted to social networking sites. As a result, regular use of social networking sites can cause addiction, and addictive behaviour can produce delusions, which can be harmful if not recognised and treated promptly (Mills & Rahal, 2019).
Research by Davila et al. (2012) on social websites, especially in the frequent use of Facebook, has shown how it may affect mental health conditions. Additionally, they revealed that more positive interactions were linked to fewer depressive symptoms, while unfavourable interactions were linked to higher depression symptoms (Centre for Mental Health, 2010). Most researchers have studied mental health difficulties in recent years and evaluated the efficacy of various treatments (Bloom et al., 2011).
Furthermore, they have proved that the evidence-based movement has substantially improved mental health cases. But there is a need for an effective model to address the global burden of mental health conditions (Becker et al., 2018). For mental disorder classification, the Machine Learning (ML) technique is widely applied toward facilitating the diagnosis objectively (Gao et al., 2018; Zhang et al., 2015). Henceforth, the present study uses a practical learning approach to analyse mental disorder detection of social network sites.
1.3 Research Problem
After the onset of the disorder, limited treatment prospects and significantly fewer options to alter the illness course are significant obstacles to improving the mental issue prediction results (Millan et al., 2016). In this way, prevention is the most promising strategy to reduce the high close to familial, clinical, personal, economic, and societal costs of mental disorders worldwide (Fusar-Poli, 2017). In most situations, pre-processing and data extraction is necessary to gain more informative features and reduce duplicated information. However, due to its automatic pattern learning from a large amount of data, ML-based deep learning (DL) yields better outcomes (Arbabshirani et al., 2017; Yang et al., 2019). Additionally, it can categorise mental diseases based on actual events. Therefore, the current study focuses on an efficient learning model that detects mental health disorders on social media and then validates the findings in a real-time setting.
1.4 Research aim and objective
This research aims to predict Social Network Mental Disorders Using ML-based learning techniques. The primary objectives of this research are as follows: •To study, design and analyse the real-time data and extract the significant features for identifying mental disorders from those using online social networks. • To capture the social interaction behaviour of a user using effective learning architecture. • Using an effective deep learning model to provide a better quality of life and prevent mental disorders from the general population. • To validate and compare the results with a traditional method like SVM, fuzzy neural network, or RNN method in terms of precision, accuracy, false detection ratio and 10-fold cross-validation.
1.5 Importance of the study
Nowadays, social media is frequently utilised and has the potential to easily lead people who suffer from various diseases, such as stress, anxiety, and depression. The social media platform is the most significant factor impacting a person's mental health. Consequently, being able to detect a disease at an early stage can aid people in avoiding mental illness. The machine learning models are practical and enable quick and straightforward early diagnosis of mental disorders.
1.6 Thesis outline
The structure of the thesis will be discussed as follows; Chapter 1 provides a brief background about the concepts and theory of social network websites and the need for a mental disorder detection model. Also, this section explains the problem statement, research aim and objective. Chapter 2 presents a detailed overview of machine learning-based classifiers and previous studies on mental health disorder detection, deep learning, and feature extraction approaches and then discusses the research gap for the above-discussed studies.
Chapter 3 identifies the research questions and explains the objectives and procedures carried out in this research. A detailed description of all the research approaches, which includes the system architecture, implementation, and process flow to be used for the analysis, is included in this chapter. Chapter 4 reports the findings and proof of concept of this research methodology. Then, the results are validated, and the performance is compared with the traditional method regarding the detection model's precision, recall, accuracy, sensitivity and specificity. And finally, chapter 5 discusses the conclusion and presents the key findings from the study, answering all the research questions proposed in the study along with recommendations for upcoming research.
References
- Acronym, T. 2022. Social Media And Mental Health: Pros, Cons, And Solutions. 2022. Arbabshirani, M.R., Plis, S., Sui, J. & Calhoun, V.D. 2017. Single subject prediction of brain disorders in neuroimaging: Promises and pitfalls. NeuroImage. (145). pp. 137–165. Becker, D., van Breda, W., Funk, B., Hoogendoorn, M., Ruwaard, J. & Riper, H. 2018. Predictive modeling in e-mental health: A common language framework. Internet Interventions. (12). pp. 57–67. Bloom, D.E., Cafiero, E.T., Jane-Llopis, E., Abrahams-Gessel, S., Bloom, L.R., Fathima, S., Feigl, A.B., Gaziano, T., Mowafi, M. & Pandya, A. 2011. The Global Economic Burden of Non-Communicable Diseases: Geneva: World. In: Economic Forum. 2011. Caspi, A. & Moffitt, T.E. 2018. All for One and One for All: Mental Disorders in One Dimension. American Journal of Psychiatry. (175)9,. pp. 831–844. Centre for Mental Health 2010. Centre for Mental Health. 2010. Chua, R.Y.J. 2018. Innovating at Cultural Crossroads: How Multicultural Social Networks Promote Idea Flow and Creativity. Journal of Management. (44)3,. pp. 1119–1146. Cuncic, A. 2021. Mental Health Effects of Reading Negative Comments Online. 2021. Davila, J., Hershenberg, R., Feinstein, B.A., Gorman, K., Bhatia, V. & Starr, L.R. 2012. Frequency and quality of social networking among young adults: Associations with depressive symptoms, rumination, and corumination. Psychology of Popular Media Culture. (1)2,. pp. 72–86. Elbay, R.Y., Kurtulmuş, A., Arpacıoğlu, S. & Karadere, E. 2020. Depression, anxiety, stress levels of physicians and associated factors in Covid-19 pandemics. Psychiatry Research. (290). pp. 113130. Eurekalert 2022. Social media information can predict a wide range of personality traits and
