Utilizing machine learning technique for emotion learning and aiding mental health issues
Mental health has long been considered a difficult subject to discuss openly, and stigmas still surround it. Poor mental health has become a prevalent issue since people have difficulty talking about and expressing their feelings. The problem has been escalating during this COVID-19 outbreak. In to...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Published: |
Institute of Research and Journals
2022
|
| Online Access: | http://psasir.upm.edu.my/id/eprint/102572/ |
| _version_ | 1848863832565350400 |
|---|---|
| author | Husin, Nor Azura Wan, Gibson Liang Kamaruzaman, Nurul Nadhrah |
| author_facet | Husin, Nor Azura Wan, Gibson Liang Kamaruzaman, Nurul Nadhrah |
| author_sort | Husin, Nor Azura |
| building | UPM Institutional Repository |
| collection | Online Access |
| description | Mental health has long been considered a difficult subject to discuss openly, and stigmas still surround it. Poor
mental health has become a prevalent issue since people have difficulty talking about and expressing their feelings. The problem has been escalating during this COVID-19 outbreak. In today's modern society, it is demonstrated that the technology has the capability of assisting health care providers in assisting their patients with mental health concerns. With this idea, a system named EMOICE, which is a speech emotion recognition system to aid mental health issues, is developed.
Doctors or therapists can utilize this technique to analyze and comprehend their patients' emotions, which will aid them in
making diagnoses. EMOICE can also be used for emotional learning, where people can use empathy and understanding to
deal with mental health concerns. EMOICE will use human speech to extract features such as pitch, voice quality, and voice
spectral, which will be used by the algorithm to learn and produce accurate results. EMOICE will employ machine learning
techniques, and among the classifiers tested and compared, 1D-Convolutional Neural Network (1D-CNN) has a high
accuracy value of 94.78 percent. As a result, this approach can help doctors and therapists better understand their patients'
thoughts and emotions, as well as help patients become more self-aware and develop empathy for others in their community
and the world around them. |
| first_indexed | 2025-11-15T13:39:11Z |
| format | Article |
| id | upm-102572 |
| institution | Universiti Putra Malaysia |
| institution_category | Local University |
| last_indexed | 2025-11-15T13:39:11Z |
| publishDate | 2022 |
| publisher | Institute of Research and Journals |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | upm-1025722024-02-28T08:47:07Z http://psasir.upm.edu.my/id/eprint/102572/ Utilizing machine learning technique for emotion learning and aiding mental health issues Husin, Nor Azura Wan, Gibson Liang Kamaruzaman, Nurul Nadhrah Mental health has long been considered a difficult subject to discuss openly, and stigmas still surround it. Poor mental health has become a prevalent issue since people have difficulty talking about and expressing their feelings. The problem has been escalating during this COVID-19 outbreak. In today's modern society, it is demonstrated that the technology has the capability of assisting health care providers in assisting their patients with mental health concerns. With this idea, a system named EMOICE, which is a speech emotion recognition system to aid mental health issues, is developed. Doctors or therapists can utilize this technique to analyze and comprehend their patients' emotions, which will aid them in making diagnoses. EMOICE can also be used for emotional learning, where people can use empathy and understanding to deal with mental health concerns. EMOICE will use human speech to extract features such as pitch, voice quality, and voice spectral, which will be used by the algorithm to learn and produce accurate results. EMOICE will employ machine learning techniques, and among the classifiers tested and compared, 1D-Convolutional Neural Network (1D-CNN) has a high accuracy value of 94.78 percent. As a result, this approach can help doctors and therapists better understand their patients' thoughts and emotions, as well as help patients become more self-aware and develop empathy for others in their community and the world around them. Institute of Research and Journals 2022-08 Article PeerReviewed Husin, Nor Azura and Wan, Gibson Liang and Kamaruzaman, Nurul Nadhrah (2022) Utilizing machine learning technique for emotion learning and aiding mental health issues. International Journal of Advances in Electronics and Computer Science, 9 (8). 53- 57. ISSN 2394-2835 https://iraj.doionline.org/dx/IJAECS-IRAJ-DOIONLINE-18997 |
| spellingShingle | Husin, Nor Azura Wan, Gibson Liang Kamaruzaman, Nurul Nadhrah Utilizing machine learning technique for emotion learning and aiding mental health issues |
| title | Utilizing machine learning technique for emotion learning and aiding mental health issues |
| title_full | Utilizing machine learning technique for emotion learning and aiding mental health issues |
| title_fullStr | Utilizing machine learning technique for emotion learning and aiding mental health issues |
| title_full_unstemmed | Utilizing machine learning technique for emotion learning and aiding mental health issues |
| title_short | Utilizing machine learning technique for emotion learning and aiding mental health issues |
| title_sort | utilizing machine learning technique for emotion learning and aiding mental health issues |
| url | http://psasir.upm.edu.my/id/eprint/102572/ http://psasir.upm.edu.my/id/eprint/102572/ |