سامانه همایش های دانشگاه آزاد اسلامی واحد نجف آباد

صفحه اصلی / International Conference on Artificial Intelligence; City, Industry and Health

A Comparative Study of Emotion Detection Methods Using NLP Techniques on Sentiment140 and IMDb Datasets

نویسندگان :

Bahar Asgari¹ Hamid Rastegari² Vahid Nejati³

1- Department of Computer Engineering, Na.C, Islamic Azad University, Najafabad, Iran 2- Department of Computer Engineering, Na.C, Islamic Azad University, Najafabad, Iran 3- Faculty of Psychology and Educational Sciences, Shahid Beheshti University, Tehran, Iran

کلمات کلیدی :

Emotional Analysis،Mental Illness Detection،Logistic Regression،Support Vector Machine،Natural Language Processing

چکیده :

Emotional analysis plays a crucial role in identifying unknown aspects of human behavior, which nowadays is use by Artificial Intelligence techniques such as Natural Language Processing for the early prediction and diagnosis of mental illness. In recent years, computer science researchers have conducted numerous studies on detecting mental illness through computational methods. However, this domain still requires further exploration of unresolved challenges alongside existing solutions. This study reviews recent developments, key concepts, and issues in this area, focusing on two datasets: Sentiment140 and IMDb. Following preprocessing and normalization, several widely used classification algorithms such as Support Vector Machine, Logistic Regression, Random Forest, Extremely Randomized Trees, and Multi-Layer Perceptron are implemented, and their performance is compared based on accuracy. The experimental results indicate that the existence of a dictionary and the use of combined methods and algorithms such as Support Vector Machine and Logistic Regression on the IMDb dataset will achieve higher accuracy.