End to End Data Science Project

Final Project

• This is an individual project.
• Presentations will be held on the final day of the course.
• Ensure all labs and projects are submitted by the end of the bootcamp.

Introduction

Welcome to your final project! The goal of this project is to apply your data science skills learned throughout the bootcamp by completing an end-to-end analysis. The project will demonstrate your proficiency in skills required for your career path and help you develop a portfolio to showcase to potential recruiters.

We recommend exploring a topic and dataset that personally interests you to make the project more engaging and rewarding. Your project can focus on:

• Generative AI 
• Machine Learning
• Deep Learning
• Computer vision

Project Overview

1- Data Collection:

• Select a business problem to address.
• Locate and gather necessary data.

2. Data Preparation: Organize and clean your data

Any technique you use to prepare your data depends on the type of data you have and the requirements of your chosen model. Data preparation is a crucial step to ensure your model performs well. Here are some common examples of tasks you might perform:

• Handle outliers and missing values.
• Perform type casting and feature selection.
• Convert categorical data to numerical.
• Apply statistical methods to explore and understand data distributions, correlations, and variances.

3.Exploratory Data Analysis (EDA):

• Analyze variables, patterns, and correlations.
• Generate insights that can guide your narrative.
• Apply statistical techniques and visualization to examine feature relationships.
•  Use Tableau to create a dashboard or some plots.
• Optional SQL Component

4. Machine Learning and Deep Learning: 

• Experiment with various models and hyperparameters.
• feature engineering and preprocessing, model selection and evaluation
• Define evaluation metrics clearly.
• Justification for the best model selection.

5. Gen AI

• Integrate some of these technologies into your pipeline, such as LLMs, RAG, AI agents, text-to-speech, speech-to-text, or VLMs.
• You can develop an AI-powered chatbot that assists users in a practical domain (e.g., healthcare, education, customer service, or legal advice). Make it multimodal or voice-enabled.

Helpful Resources:

• Hugging Face
• LangChain

Data Sources:

• Kaggle
• UCI Machine Learning Repository
• PorData
• And many more

Mandatory Requirements:

• A clear explanation of your dataset and project goals.

• Documented project implementation, highlighting key decisions, challenges faced, and how they were addressed at each stage.

• Insights from exploratory data analysis.

• Well-explained visualizations showcasing key findings.

• A GitHub repository containing:

• Well-documented Python code.

• Optional SQL database

• A README explaining the project goals, methodology, and results.

• requirements.txt

• Project Planning:

  • Use Trello, Kanban... for tracking progress.
  • Create a repository and commit/push frequently.

• A presentation (15-20 minutes) explaining your findings.

Nice-to-Have:

• App to showcase 
• Transform the project into a usable product
• Create a repository and commit/push frequently.

Deliverables

• GitHub repository.
• Final presentation slides summarizing project work and insights.