International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 13 Issue: 02 | Feb 2026
p-ISSN: 2395-0072
www.irjet.net
Automatic mood detection tool: Self Response Inventory Jiger P. Acharya1, Dr. Milind S. Shah2 1Research Scholar, Gujarat Technological University –Ahmadabad, Gujarat, India 2Professor, Electronics & Communication Dept., S.S.E.C-Bhavnagar, Gujarat, India
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract – Depression is a mood disorder in which negative
telephone lines where question is assessed on time scale of months. After decades scenario was changed with modifications of question and examination period was in weeks. In present situations where technology had made great impact on our life style the risk of depressed mood increase [6],[13],[14],[15],[16],[17],[28].
mood of a person prolonged more than 2 weeks The World Health Organization list depression as major cause of disability. More than 370 million people suffer from it. Proper concern is not given to such burning issue globally. Psychological and clinical treatments are available but quality and availability of clinical professional associated to domain are limited. to resolve this issue an automated ML based approach is suggested for mood detection. Person can check its present mood automatically by its own at no cost. The accuracy of approach is above 90%. Person can repeat the test and if negative mood result continues then person can say to converge toward depressed mood.
1.2 Audio Channel approach Person having depressed mood speech can be identified so speech processing domain deal with such task and classify normal, mild depressed and major depressed person from vocal features [5],[11],[18],[19].
1.3 Visual Channel approach
Key Words: Depression, Positive mood, Negative mood, Psychology, PHQ-9, BDI, CES-D, MOS, SDDS-PC, QIDS, DMI-10, Mild Negative, Depressive disorder
Person having depressed mood can be identified from its visual features since long time person’s expressions are recognized into seven prototypic class and they can further classify as positive and negative expression. Happy is positive emotion or expression where sad, disgust, angry, contempt are negative expressions. Surprise may lead to positive or negative expression according to the association of it with event [4],[7],[8],[9],[20],[21].
1. INTRODUCTION Depression is leading cause of disability globally in all age groups which severely affects the global health index. It is complex and an ongoing problem which may resolve with proper care but may reoccur via various risk factors. The medical community does not fully understand the causes of depression but mostly known various risk factors like environmental factors, psychological, social factors, financial factors, drug addictions change in genetic features, brain's neurotransmitter levels, accidental events like head major head injury and many more. Early detection plays a crucial role as by positive counseling, various physical and mental exercises like meditation and proper medication may resolved depression else situations become worst up to suicidal tendency. Depression can be detected through various behavioral symptoms and during clinical interview by medical expert but the rate of the affected person reach at right time with right medical professional is very low so many researchers from physiological, pattern matching, computer vision have tried to detect depression features from behavior verbal and nonverbal channels with help of various of machine learning algorithms under Artificial Intelligence domain. Neural Network has shown shows great potential to resolve this burning issue of human health care domain to serve mankind.
1.4 Social Media approach In era of social networking technology person spare time more on screen so social media like Facebook, twitter, snap are identified person’s mood and emotions so based on status, text message and updates like picture, snap person’s data in which Text data in various levels and categorize with positive or negative broadly and much specific like emphatic, glad etc. databases for textual emotion analysis [18].From uploaded pictures, emojis updated on various social media platforms person’s mood behaviour can be examined.
1.5 Bio -sensor approach There are various bio sensors are available to check the person’s mental activity and associated parameters like EEG based approach physiological signals receive from brain are measured which are real time in nature and cannot suppress or hide like facial expressions, audio and textual emotions so it is more authentic and precise only drawback is the cost and complexity associate with this approach [18].The result of such sensors is efficient only drawback is the cost and complexity of setup, evaluation and examination criteria in such modalities[19].
1.1 Clinical Interview Psychological and clinical based interview to evaluate depression symptoms are carried out since last 60 years. Initially interviews were taken place in person or on
© 2026, IRJET
|
Impact Factor value: 8.315
|
ISO 9001:2008 Certified Journal
|
Page 821