Challenge background
Pakistan is home to a plethora of low-resource languages. Mental health issues are a pertinent issue people in Pakistan are facing these days, however, due to the stigmas associated with this topic, it remains largely ignored. This project aims to tackle this issue, by building a first of its kind urdu based mental health chatbot. This project will be a vibrant addition to the work promoting the development of low-resource languages, and will also address the issue of mental health issues by building a compassionate NLP-based model that can hold personalized textual conversations with its users.
The problem
We will need to gather an urdu based dataset that can be done by using newspaper sources. We will further use NLP techniques to build the model. Furthermore, we will build and deploy a dashboard to display the results.
Goal of the project
1. Gather Urdu based data.
2. Develop a chatbot model.
3. Develop visualizations.
4. Deploy the app.
Project timeline
- 1
Week 1
Data Collection (pre-week 1 even).
Data Pre-Processing.
- 2
Week 2
Building model
- 3
Week 3
Building model and visualization
- 4
Week 4
Building and deploying the app.
- 5
Week 5
Preparing final documents and results.
What you'll learn
1. Collection of Data. 2. Pre-processing of Data. 3. Modelling. 4. Model deployment into a possible API. 5. Visualisation.