AI-Final-Project

Final project for Artificial Intelligence course.

🎲 How does the process look like?

🧩 Implementation flow of the algorithm:

1. 🗂️ Data Collection -> 50 videos that come from the TikTok app
2. 📊 Preprocessing Data -> convert raw data obtained through the Data Collection process into data/information that will be used as parameters or information
3. 📚 Feature Learning -> the program learns the features contained in each data frame that has been obtained in the previous process
   1. Convolution -> make use of 'filter' using 'random' way
   2. ReLU (Rectivfied Linear Unit) -> transform data in the Convolution process with a zero value to the pixel value
   3. Pooling -> reduce the spatial size of a feature in the convolution process to speed up the computational process by using a down-sampling operation
4. 📦 Classification -> classify the results of the Pooling on the Feature Learning process based on several parameters
   1. Flattening -> changing the frames generated from Feature Learning to be collected into one linear vector
   2. Fully Connecting -> reshaping the activation map into a vector so that it can be used as input
   3. Softmax -> the activation stage used for output
   4. Model -> classify each frame

🎯 Project brief: Detecting potential mental health disorder from video using the CNN (Convolutional Neural Network) Method

🔗 Dataset Link

🔧 Tech stack: Python

See the code here!

📄 Looking for a full explanation of this project? Download the project report here.

🧞‍♂️ Team 14

1. 12S20017 - Lile Manalu
2. 12S20034 - Daniel Limbong
3. 12S20048 - Jevania