Week 2 — Emotion Detection
Hello everyone! We -Şeyma Yılmaz, Mücahit Fındık, Yusuf Emre Genç- continue to provide information about the progress of our Machine Learning project. The subject we will mention is the algorithm we will use! What algorithm will we use? Why do we use the algorithm? What are its advantages? How does it work? You can be assured that you will find the answer to your questions in the rest of the post. Let’s start!
We will use the Convolution Neural Network (CNN) algorithm in our project, of course. [1] It is a multi-layer neural network with a special architecture. It is used to detect complex features in data. Although it may seem like a strange mix of biology and computer science, CNN is a very effective mechanism for image recognition. It is used in areas such as image recognition and OCR, object detection for self-driving cars, face recognition on social media, image analysis in healthcare.
The CNN algorithm opened a period in machine learning. So it brought many advantages. The most significant advantage of convolutional neural networks is that it does not require any supervision of a human. That means, when you give data, it can train the model by itself. As another advantage, the CNN algorithm is computationally efficient thanks to convolution and pooling operations. Also, the CNN algorithm can train computationally large models by keeping only a small number of parameters.
CNN basically consists of 4 layers. These layers; convolution layer, activation function layer, pooling layer, and normalization layers.
[2] Convolutional Layers: Performs a series of mathematical operations for the entire sub-region in an image to produce a single value in the output attribute map. The activation layer decides whether the neuron will fire or not. So it gives the final value of output. The pooling layer saves us so many features by summarizing the features. Normalization helps to increase the stability of our neural network.
We will give details about the dataset that will be used in the next post. Let’s meet again next week!
