RSIS International

RSIS
Submission Deadline-20th August 2024
August 2024 Issue : Publication Fee: 30$ USD Submit Now
Submission Deadline-20th August 2024
Special Issue on Education: Publication Fee: 30$ USD Submit Now
Submission Deadline-20th August 2024
Special Issue on Public Health: Publication Fee: 30$ USD Submit Now

International Journal of Research and Innovation in Applied Science (IJRIAS) |Volume VII, Issue XI, November 2022|ISSN 2454-6194

Detection of Face Emotion and Music Recommendation System using Machine Learning

Danish Ali1*, Md. Tahmidul Huque2, Jafreen Jafor Godhuli3, Naeem Ahmed4
1,4Department of Computer Science, GPGC Haripur, Pakistan
2,3Department of Computer Science and Engineering, Bangladesh University of Business and Technology, Bangladesh
*Corresponding Author

IJRISS Call for paper

Abstract: Face emotion detection has recently attracted a lot of interest because of its uses in computer vision and the field of human-computer interaction. Various methods and applications were suggested and put into use as a result of the ongoing research in this area. In this study, we present an emotion-recognition recommender system that can identify a user’s feelings and offer a selection of suitable songs that might lift his spirits. To gather information and enable us to give the users a selection of music tracks that are effective at lifting the users’ spirits, a quick search was undertaken to learn how music may impact the user mood in the short term. The suggested system recognizes emotions, and if the individual is feeling down, a special playlist including the best kinds of music will be played to lift his spirits. On the other hand, if a favorable mood is recognized, an appropriate playlist will be offered that contains several genres of music that will amplify the pleasant feelings. Principal Component Analysis (PCA) methods and the Fisher Face algorithm are used to implement the suggested recommender system.

Keywords: Machine Learning, Face Emotion Recognition, Music Recommendation System, CNN, Deep Learning, Classification

I. INTRODUCTION

Around the world, there are different kinds of face expressions such as fear when the inner eyebrows are curved upward and are elevated and drawn together [1]. The eyes are vigilant and tense. Disgust, when there is relaxation in brows and eyelids. Additionally, the top lip is lifted and curled, usually asymmetrically [2]. When someone is happy, they have relaxed brows, an open mouth, and lips that are drawn back toward their ears. Astonishment, when the brows are lifted [3]. The lower eyelid is relaxed, while the upper eyelid is open widely [4]. The jaw is also opened. When treating depression, music therapy is seen as a beneficial addition to regular care [5]. There is a wide range of music that is available, along with a wide range of tempos, amplitudes, and moods [6]. Additionally, there have been significant improvements in emotional intelligence [7]. Geometric-based and appearance-based feature extraction techniques are the two most well-known methods [8]. For the same, there are several methods and techniques accessible. However, the issue is still unresolved since the two tasks are carried out in distinct ways [9]. In order to improve technology and even our daily lives, we must attempt to merge the two technologies [10]. In our modern life, we spend a lot time in front of mobiles and computers [11]. While working with these technologies, we get tired and frustrated after sometime. So, the motivation for this project is to get the face emotion, to relax the user, to entertain the user, to fresh the mood of the user. A person wants to listen