This repo includes the code used as part of my final year Software Engineering honours thesis on detecting symptoms of depression in clinical interview transcripts using deep neural networks. The study was centered around the DAIC-WOZ depression dataset of transcripts and audio recordings from 189 clinical interviews with depressed and non-depressed individuals. I have included my full thesis in this repo.

I am not allowed to release the dataset that was used for the project, and the code is a bit fragmented with a whole lot of functions I used to get different results for my research. However, if you are interested to run it on your local machine, you certainly can!

1. First of all, you'll need Python 3.6.

2. Next, you'll have to clone the repository.

git clone https://github.com/sudhirmandarapu/deepdepdetect

3. Set up a python venv for this project.

pip install virtualenv

virtualenv venv

source venv/bin/activate

3. Install the python requirements.

pip install -r requirements.txt

You can now use any of the modules in this project!

• TensorFlow - ML platform
• Keras - Easy to use ML library