Week 1 — Emotion Detection
Hello everyone!
The team consists of three people named Şeyma Yılmaz, Mücahit Fındık and Yusuf Emre Genç. In this post, we will introduce our machine learning project which determining what are people’s emotion. Then we will continue to report progress every week throughout the semester. Wish you pleasant reading!
We know that it can sometimes be difficult to understand people’s emotions. However, it is critical to recognize what people have emotions in some situations. For example, imagine you are a speaker in a session. Would not it be perfect to know the feelings of the audience? Or when traveling in your car what about a music list that changes according to your emotions? So how will we observe these emotions?
With an appropriate machine learning algorithm and data set, we can cope with this problem very well. The data set consists of photographs of different people with different emotions. There are 36k images in the dataset, which is enough to train the system very well. Also, we will create classes -fear, happy, sad, disgust, neutral, angry, surprised- for each different emotion using the Convolutional Neural Network algorithm. So how do we detect emotions in an image?
CNN is a very effective mechanism used for image recognition. With the algorithm, we can identify low-level features such as curves and edges. In other words, we will examine the prominent areas of the face, such as the eyes, nose, and mouth. Then, we will use the information to predict emotion.
So, that’s all for now! We will give more details about the algorithm will be used in the next post. Let’s meet again next week!
