Ajeenkya D.Y.

Patil School of Engineering (Polytechnic)

Charoli, via Lohegaon, Pune-412 015.

Department of Computer Engineering 2022-23[5th Sem] Report of

Capstone Project Execution and Report
Writing
On
DEPRESSION AND STRESS MONITORING SYSTEM VIA
SOCIAL MEDIA DATA
Submitted by:

Roll Number Name of Student Enrollment Number

KAUSTUBH 2116490119
46

50 VIKRAM 2116490130

51 SARVESH 2116490131

52 NIRANJAN 2116490134

53 SHREYANSHU 2116490142
 Ms. Smita Arude Mam
( Project Guide)
 Ajeenkya D.Y. Patil School of Engineering (Polytechnic)
Charholi,Pune 412-015
Department of Computer Engineering
SEMESTER 5TH- 2023-24

CERTIFICATE
This is to certify that Project report entitled " DEPRESSION AND STRESS
MONITORING SYSTEM VIA SOCIAL MEDIA DATA" is
submitted in the partial fulfilment of requirement for the award of the Diploma in
Computer Engineering by Maharashtra State Board of Technical Education as record of
student's own work carried out by them under the guidance and supervision at D.Y. School
Of Engineering, Pune-412015, during the academic year 2023-24.

ROLL NO. NAME ENROLLMENT NO

46 KAUSTUBH 2116490119

50 VIKRAM 2116490130

51 SARVESH 2116490131

52 NIRANJAN 2116490134

53 SHREYANSHU 2116490142

Place: Charholi (Bk)

Date: / /2023
([Link] Arude Mam)
Project Guide (Smt. Priyanaka Kasare)
Head of Computer Department
 ACKNOWLEDGEMEMT
It is with profoundly sense of gratitude that we acknowledge from our guide
[Link] Arude She has been guide in the true sense of word, a guide who satisfaction from
our word & progress.

We are highly obliged to Ms. Priyanka Kasare. .Head of Computer

Department for aberrance & good co-operation given to us for bringing this project
to almost standard.
We are grateful to our principal Dr. Nagesh Shelke for proceeding
acknowledgement to us in the connection of this project concluding. We appreciate
the assistance of all staff that helps us in for their sincere & obliging help to make
our project successfully.
Action Plan :

[Link] Details of activity Planned start Planned end Name of responsible

date date group members

1. Searching for the topic SHREYANSHU SARVESH

2. Confirmed the topic

4. Collecting information
SARVESH KAUSTUBH

5. Making Proposal VIKRAM NIRANJAN

6. Testing And Analysis SHREYANSHU SARVESH

7. SHREYANSHU SARVESH
Preparation of Final
Project and Report KAUSTUBH VIKRAM NIRANJA

SHREYANSHU SARVESH
8. Final submission of the
Project KAUSTUBH VIKRAM NIRANJA
 ABSTRACT

Stress and Depression is one of the most widely recognized and handicapping
mental issue that relevantly affects society. Automatic health monitoring systems
could be crucial and important to improve depression and stress detection system
using social networking. Sentiment Analysis alludes to the utilization of natural
language processing and content mining approaches planning to recognize feeling
or opinion. Full of feeling Computing is the examination and advancement of
frameworks and gadgets that can perceive, decipher, process, and mimic human
effects. Sentiment Analysis and deep learning techniques could give powerful
algorithms and frameworks to a target appraisal and observing of mental issue and,
specifically of depression and stress. In this paper, the application of sentiment
analysis and deep learning methodologies to depression and stress detection and
monitoring are discussed. In addition, a fundamental plan of an incorporated
multimodal framework for stress and depression checking, that incorporates
estimation investigation and full of feeling processing strategies, is proposed. In
particular, the paper traces the fundamental issues and moves comparative with
the structure of such a framework.
 INTRODUCTION

Keywords – stress and depression; e-health; sentiment analysis, social media,

deep learning
Social media is arguably the richest source of human generated text input.
Opinions, feedbacks, and critiques provided by internet users reflect attitudes and
sentiments towards certain topics. This paper presents a knowledge-based system,
which includes an emotional health monitoring system to detect users with
possible psychological disorders specially depression and stress. Symptoms Of
these psychological disorders are usually observed passively. In this situation,
author argue that online social behaviour extraction offers an opportunity to
actively identify psychological disorder at an early stage. It is difficult to identify the
disorder because the psychological factors considered in standard diagnostic
criteria questionnaire cannot be observed by the registers of online social activities.
Depression and stress are one of the most common and disabling mental disorders,
and has a relevant impact on society. Currently, methods for depression and stress
detection and diagnosis rely on self-reporting coupled with the health care
practitioners informed assessment. The provision of effective health monitoring
systems and diagnostic aids could be crucial and important to improve health
professional’s work and lower healthcare costs. Sentiment and deep learning
technology could help to tackle these objectives by providing effective tools and
systems for objective assessment. Such tools and systems do not aim to replace
the psychologist or psychiatrist, but they could support their decisions. Our
approach, New and innovative for the practice of psychological disorder detection,
it does so do not trust the self-disclosure of those psychological factors through the
questionnaires. Instead, propose a machine learning technique that is detection of
psychological disorder in social networks which exploits the features extracted
from social network data for identify with precision possible cases of disorder
detection. We perform an analysis of the characteristics and we also apply machine
learning in large-scale data sets and analyse features of the two types of
psychological disorders.
M. Al-Qurishi, M. S. Hossain, M. Alrubaian, S. M. M. Rahman, and A. Alamri - In
this paper, author propose an integrated social media content analysis platform that
leverages three levels of features, i.e., user-generated content, social graph
connections, and user profile activities, to analyze and detect anomalous behaviours
that deviate significantly from the norm in large-scale social networks. Several types
of analyses have been conducted for a better understanding of the different user
behaviours in the detection of highly adaptive malicious users.
Huijie Lin, JiaJia, JiezhonQiu, Yongfeng Zhang, LexingXie, Jie Tang, Ling Feng,
and Tat-Seng Chua - In this paper, we find that users stress state is closely related to
that of his/her friends in social media, and we employ a large-scale dataset from real-
world social platforms to systematically study the correlation of users’ stress states
and social interactions. We first define a set of stress-related textual, visual, and
social attributes from various aspects, and then propose a novel hybrid model - a
factor graph model combined with Convolutional Neural Network to leverage tweet
content and social interaction information for stress detection.
 BudhadityaSaha, Thin Nguyen, DinhPhung, SvethaVenkatesh - Mental illness has
a deep impact on individuals, families, and by extension, society as a whole. Social
networks allow individuals with mental disorders to communicate with others
sufferers via online communities, providing an invaluable resource for studies on
textual signs of psychological health problems. Mental disorders often occur in
combinations, e.g., a patient with an anxiety disorder may also develop depression.
AndreyBogomolov, Bruno Lepri, MichelaFerron, Fabio Pianesi, Alex (Sandy)
Pentland- In our paper, propose an alternative approach providing evidence that
daily stress can be reliably recognized based on behavioural metrics, derived from
the user's mobile phone activity and from additional indicators, such as the weather
conditions (data pertaining to transitory properties of the environment) and the
personality traits (data concerning permanent dispositions of individuals). Our
multifactorial statistical model, which is person-independent, obtains the accuracy
score of 72.28% for a 2-class daily stress recognition problem. The model is efficient
to implement for most of multimedia applications due to highly reduced low
dimensional feature space (32d). Moreover, we identify and discuss the indicators
which have strong predictive power.
 BimalViswanath† Alan MisloveMeeyoung Cha Krishna P. Gummadi – In this
paper, study the evolution of activity between users in the Facebook social network
to capture this notion. Also find that links in the activity network tend to come and
go rapidly over time, and the strength of ties exhibits a general decreasing trend of
activity as the social network link ages. For example, only 30% of Facebook user
pairs interact consistently from one month to the next. Interestingly, and find that
even though the links of the activity network change rapidly over time, many graph-
theoretic properties of the activity network remain unchanged.
 I.-R. Glavan, A. Mirica, and B. Firtescu - Social media tools are wide spread in
web communication and are gaining popularity in the communication process
between public institutions and citizens. This study conducts an analysis on how
social media is used by Official Statistical Institutes to interact with citizens and
disseminate information. A linear regression technique is performed to examine
which social media platforms (Twitter or Facebook) is a more effective tool in the
communication process in the official statistics area. Our study suggests that Twitter
is a more powerful tool than Facebook in enhancing the relationship between official
statistics and citizens, complying with several other studies. Next, performed an
analysis Twitter network characteristics discussing “official statistics” using
NodeXL that revealed the unexploited potential of this network by official statistical
agencies.
 E. U. Berbano, H. N. V. Pengson, C. G. V. Razon, K. C. G. Tungcul, and S. V.
Prado - The paper presents further research on neural engineering that focuses on
the classification of emotional, mental, physical and no stress through the use of
Electroencephalography (EEG) signal analysis. Stress is one of the leading causes
of several health-related problems and diseases. Therefore, it becomes necessary for
people to monitor their stress. The human body acquires and responds to stress in
different ways resulting to two classifications of stress namely, mental and
emotional stress. Traditional methods in classifying stress such as through
questionnaires and self-assessment tests are said to be subjective since they rely on
personal judgment. Thus, in this study, stress is classified through an objective
measure which is EEG signal analysis .
Methods: A cross- sectional study was conducted with 100 women with CPP
and 100 without CPP. The Hospital Anxiety and Depression Scale (HADS)
was used to evaluate the presence of anxiety and depression.
Sociodemographic, behavioral and clinical characteristics were
investigated. Fisher’s exact test was used to compare characteristics
between groups. A log-binomial regression model was used, with
adjustment for age, skin color, schooling, body mass index and pain.
Prevalence ratios (PR), together with their 95% confidence intervals (CI), were
calculated to investigate factors associated with anxiety, depression and
MADD.

Anxiety and depression are the mental disorders most commonly investigated
in patients with chronic pain. Studies have shown a high frequency of
psychological problems in women with CPP, although the
rates vary between authors. An association has also
been reported between symptoms of depression and anxiety
and a history of physical and/or sexual abuse both in
women with CPP and in those without CPP T his
emphasizes the importance of evaluating the occurrence of
such issues and its relationship with mental disorders.

In view of the increasing evidence associating a psychosocial context with CPP, the
objective of the present study was to investigate the prevalence of anxiety,
depression and mixed anxiety and depressive disorder (MADD) and the factors
associated with these conditions in women with CPP compared to a group of
controls without CPP.
Variations in Stress Responses

Certain characteristics of a situation are associated with greater stress responses.

These include the intensity or severity of the stressor and controllability of the
stressor, as well as features that determine the nature of the cognitive responses or
appraisals. Life event dimensions of loss, humiliation, and danger are related to the
development of major depression and generalized anxiety (Kendler et al. 2003).
Factors associated with the development of symptoms of PTSD and mental health
disorders include injury, damage to property, loss of resources, bereavement, and
perceived life threat (Freedy et al. 1992, Ironson et al. 1997, McNally 2003).
Recovery from a stressor can also be affected by secondary traumatization
(Pfefferbaum et al. 2003). Other studies have found that multiple facets of stress
that may work synergistically are more potent than a single facet; for example, in
the area of work stress, time pressure in combination with threat (Stanton et al.
2001), or high demand in combination with low control (Karasek & Theorell 1990).

Stress-related outcomes also vary according to personal and environmental factors.

Personal risk factors for the development of depression, anxiety, or PTSD after a
serious life event, disaster, or trauma include prior psychiatric history, neuroticism,
female gender, and other sociodemographic variables (Green 1996, McNally
2003, Patton et al. 2003). There is also some evidence that the relationship
between personality and environmental adversity may be bidirectional (Kendler et
al. 2003). Levels of neuroticism, emotionality, and reactivity correlate with poor
interpersonal relationships as well as “event proneness.” Protective factors that
have been identified include, but are not limited to, coping, resources (e.g., social
support, self-esteem, optimism), and finding meaning. For example, those with
social support fare better after a natural disaster (Madakaisira & O’Brien 1987) or
after myocardial infarction (Frasure-Smith et al. 2000). Pruessner et al.
(1999) found that people with higher self-esteem performed better and had lower
cortisol responses to acute stressors (difficult math problems). Attaching meaning
to the event is another protective factor against the development of PTSD, even
when horrific torture has occurred. Left-wing political activists who were tortured
by Turkey’s military regime had lower rates of PTSD than did nonactivists who
were arrested and tortured by the police (Basoğlu et al. 1994).
Finally, human beings are resilient and in general are able to cope with adverse
situations. A recent illustration is provided by a study of a nationally representative
sample of Israelis after 19 months of ongoing exposure to the Palestinian intifada.
Despite considerable distress, most Israelis reported adapting to the situation
without substantial mental health symptoms or impairment (Bleich et al. 2003).
RNN Classification algorithm
1. Select the dataset.
2. Step2: Feature selection using information gain and ranking
3. Step3: Classification algorithm
4. Step4: Each Feature calculate fx value of input layer
5. Step5: bias class of each feature calculate
6. Step6: Next produce the feature map it goes to forward pass input layer
7. Step7: Calculate the convolution cores in a feature pattern
8. Step8: Produce sub sample layer and feature value.
9. Step9: Back propagation input deviation of the kth neuron in output layer.
Step10: Finally give the selected feature and classification results.
USE CASE DIAGRAM:

I. FOR J=1 NO. OF WORDS IN STOP DESIGN WORDS LIST.

If words (i) == Stop words (j)
Then eliminate words (i) end if, end for.
Tokenization – This technique removes Special characters and images. Initialize
feature vector bg feature = [0,0..0] for token in [Link]() do, if token in dict
then
Token idx=gentix (dict. token)
Bg feature [token idx] ++ else continue end if end for.
Feature Selection - Feature selection is for filtering irrelevant or redundant
features from your dataset. The feature selection is that feature selection keeps
a subset of the original features.
Sentiment Analysis – a) Given training dataset D which consists of documents
belonging to different class say Class A and Class B
b) Calculate the prior probability of class A=number of objects of class A/total
number of objects
Calculate the prior probability of class B=number of objects of class B/total
number of objects
c) Find NI, the total no of frequency of each class
Na=the total no of frequency of a class A
Na=the total no of frequency of a class B
d) Find conditional probability of keywords occurrence given class:
P (value1/Class A) = count/ni (A)
P (value1/Class B) = count/ni (B)
P (value2/Class A) = count/ni (A)
P (value2/Class B) = count/ni (B)
P (value n/Class B) = count/ni (B)
e) Avoid zero frequency problems by applying uniform distribution
f) Assign document to class that has higher probability.
CLASS DIAGRAM:

ACTIVITY DIAGRAM:

ACTIVITY DIAGRAM:
COMPONENT DIAGRAM:
DEPLOYMENT DIAGRAM:
DATA FLOW DIAGRAM

CONCLUSION
In this proposed system, automatically identifying potential online users with
depression and stress is threatening people's health. Thus, users suffering from
depression can be identified and they might be helped before they take any
drastic steps which might have a long-lasting impact. Using the data of the
social networks of the real world as a basis, we study the correlation between
the states of psychological disorder of users and their social interaction
behaviour we recommend the user for health precautions to send by mail for
user interaction.
References

1 The National Center for Complementary and Alternative Medicine

(NCCAM). [Jan 2012]. [Web Page]. [Link]

2 Barnes PM, Bloom B, Nahin RL. Complementary and alternative medicine

use among adults and children: United States, 2007. Natl Health Stat
Report. 2008;(12):1–23. [PubMed]

3 Goyal M, Haythornthwaite J, Levine D, et al. Intensive meditation for

refractory pain and symptoms. J Altern Complement Med. 2010;16(6):627–
31. [PMC free article] [PubMed]

4 Chiesa A, Malinowski P. Mindfulness-based approaches: are they all the

same? J Clin Psychol. 2011;67(4):404–24. [PubMed]