Leveraging "AI and ML for IT Employee Well-being: Detecting and Addressing Workplace Depression"

Abstract

The proposed system aims to detect and address depression levels among IT employees, offering recommendations to alleviate their distress. Comprising four key components—Face Recognition and Mood Detection, User Friendly Chat Agent, Activity Recommendations based on moods and severity levels, and Voice Analysis, the system leverages cutting-edge technologies to revolutionize employee well-being in the workplace. The Face Recognition and Mood Detection component employs machine learning techniques to create an advanced system. Using computer vision algorithms, it recognizes employee faces and analyzes facial expressions to gauge their mood. This system not only records employees' mood statuses, encompassing emotions like sadness and anger, but also empowers consultants. When an employee seeks guidance, the consultant can access their mood history, aiding in understanding behavior and tailoring support. The overarching goal is to preemptively mitigate mental stress, rectify moods, and enable consultants to address issues effectively. The User Friendly Chat Agent acts as a secure space for employees to interact with the company. Utilizing advanced natural language processing and AI technology, the agent promotes positive mental health and camaraderie while collecting valuable conversation and mental health trend data. Integrating with the consultant team, it provides personalized support and resources, enhancing well-being across the organization. Incorporating the promising avenue of voice analysis, the third component targets early depression detection. The algorithm dissects vocal characteristics such as pitch and intonation, indicative of emotional states. By processing voice recordings from employees with known depression levels, this approach offers a low-cost, non-invasive method to recognize at-risk individuals and provide timely support. The activity recommendation module proposes suitable activities based on the employee's emotional condition, categorized as mild or severe levels of depression. These activities are straightforward, easily comprehensible, and not require significant time investment. The system keeps activities log for each user, allowing filtering choices for suggested activities based on user approval. Additionally, the system generates music frequencies customized to the user's mood, potentially fostering a positive effect on the employee's mental well-being and alleviating the monotony of the office surroundings. This comprehensive system embodies a holistic approach to tackle employee depression. By integrating state-of-the-art technologies and leveraging real-time data, it aims to create an empathetic, informed, and supportive workplace, ultimately enhancing mental well-being and productivity among IT employees.

Country : Sri Lanka

1 K.A.K. Shyamal2 N.H.I. Chamanki3 K.T.M. Weliwita4 U.K.T. Himasha5 H.M Samadhi Chathuranga Rathnayake6 Lakmal Ponnamperuma

  1. Faculty of Computing, Sri Lanka Institute of Information Technology, Malabe, Sri lanka
  2. Faculty of Computing, Sri Lanka Institute of Information Technology, Malabe, Sri lanka
  3. Faculty of Computing, Sri Lanka Institute of Information Technology, Malabe, Sri lanka
  4. Faculty of Computing, Sri Lanka Institute of Information Technology, Malabe, Sri lanka
  5. Department of Information Technology, Faculty of Computing, Sri Lanka Institute of Information Technology, Malabe, Sri Lanka
  6. School of Psychology, Faculty of Humanities and Science, Sri Lanka Institute of Information Technology, Malabe, Sri Lanka

IRJIET, Volume 7, Issue 11, November 2023 pp. 664-670

doi.org/10.47001/IRJIET/2023.711088

References

  1. Chen, S., Sun, S., & Wang, Y. (2018). Web-based cognitive behavioral therapy for employees with depressive symptoms: A randomized controlled trial. Journal of Medical Internet Research, 20(1), e6.
  2. Smith, K. L., Jones, R. M., & Martinez, E. (2021). Employee perspectives on technology-enabled mental health support: A qualitative study. Journal of Applied Psychology, 106(3), 415-427.
  3. Smith, A. B., Johnson, C. D., Newman, D. A., & Wang, L. (2019). Wearable heart rate monitors for real-time detection of depression symptoms. Journal of Occupational Health Psychology, 24(4), 458465.
  4. Wang, J., Zhang, D., Wan, L., & Zhou, L. (2020). A smartphone app for identifying individuals with depressive symptoms using machine learning techniques. Journal of Affective Disorders, 265, 583-590.
  5. Lovibond, S.H. & Lovibond, P.F. (1995). Manual for the Depression Anxiety & Stress Scales. (2nd Ed.)Sydney: Psychology Foundation.
  6. Advancements and recent trends in emotion recognition using facial image analysis and machine learning models. Tuhin Kundu; Chandran Saravanan. https://ieeexplore.ieee.org/abstract/document/8284512.
  7. Facial Emotion Detection Using Deep Learning. Akriti Jaiswal; A. Krishnama   Raju;      Suman   Deb. https://ieeexplor.ieee.org/abstract/document/9154121
  8. Face Recognition and Emotion Recognition from Facial Expression Using Deep Learning Neural Network. Ali Hussien Mary , Zubaidah Bilal Kadhim, Zainab Saad Sharqi https://iopscience.iop.org/article/10.1088/1757-899X/928/3/032061/meta.
  9. Recognition of Facial Expressions with Principal Component Analysis and Singular Value Decomposition. Mandeep Kaur Lecturer in IT Deptt. SSCET. Badhani ; Rajeev Vashisht Lecturer in IT Deptt. D.A.V.I.E.T. Jalandhar; Nirvair Neeru Lecturer in Computer Sci& Engg. Punjabi university, Patiala jaffe only for women. https://www.researchgate.net/profile/Nirvair Neeru/publication/4958 7784_Recognition_of_Facial_Expressions_with_Principal_Compone nt_Analysis_and_Singular_V.
  10. Artificial Intelligence Chatbot for Depression https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7695525/
  11. Early detection of depression using a conversational AI bot: A nonclinical trial https://journals.plos.org/plosone/article?id=10.1371/journal.pon e.0279743
  12. Using AI chatbots to provide self-help depression interventions for university students https://www.sciencedirect.com/science/article/pii/S2214782922000021.
  13. A time-frequency channel attention and vectorization network for automatic depression level prediction - https://sci-hub.se/https://www.sciencedirect.com/science/article/abs/pii/S0925231221005981.
  14. Depression analysis using convolutional neural networks from speech -              Lang He, Cui Cao - https://www.sciencedirect.com/science/article/pii/S153204641830090X.
  15. Spectral Representation of Behavior Primitives for Depression Analysis - Siyang Song, Shashank Jaiswal, Linlin Shen, and Michel Valstar- https://ieeexplore.ieee.org/document/8976305.
  16. MUBS: A Personalized Recommender System for Behavioral Activation in Mental Health - Darius A. Rohani, Andrea Quemada  Lopategui,Nanna             Tuxen, Maria Faurholt-Jepsen https://www.researchgate.net/publication/341695579_MUBS_A_Pers onalized_Recommender_System_for_Behavioral
  17. A Smartphone-Based Personalized Activity Recommender System for Patients with Depression - Galen Chin-Lun Hung, Pei-Ching Yang, Chen-Yi Wan, https://www.researchgate.net/publication/301451954_A_Smartphone  Based_Personalized_Activity_Recommender_System_for_Patients_with_Depression
  18. Recommendation System to Improve Time Management for People in EducMation Environments - Samaher Al-Janabi, ahdi Abed Salman and Ahmed Fanfakh Recommendation-System-to-Improve-TimeManagement-for-People-in-Education-Environments.pdf(researchgate.net).
  19. M. Chennafi, M. A. Khan, G. Li, and Y. Lian, "Study of Music Effect on Mental Stress Relief Based on Heart Rate Variability."
  20. Hamamci, Z. (2002). The effect of integrating psychodrama and cognitive behavioral therapy on reducing cognitive distortions in interpersonal relationships. Journal of Group Psychotherapy, Psychodrama & Sociometry.
  21. Ranjbar E, Memari AH, Hafizi S, Shayestehfar M, Mirfazeli FS, Eshghi MA. Depression and exercise: A clinical review and management guideline. Asian J Sports Med. 2015.