Alia M

Internship

Developers Research

Jul 2025 - Aug 2025 • 1 mo

During the phase of the internship, I worked on 6 AI & machine learning projects: 1. Exploring & Visualizing Iris Dataset Loaded Iris dataset using pandas  Performed data inspection (.head(), .info(), .describe())  Created scatter plots, histograms, and box plots using matplotlib & seaborn  Identified feature relationships, distributions, and outliers 2.Heart Disease Prediction  Loaded Heart Disease UCI dataset from Kaggle  Cleaned data & handled missing values  Performed EDA to analyze health trends  Trained Logistic Regression & Decision Tree models  Evaluated with Accuracy, ROC curve & Confusion Matrix 3.General Health Query Chatbot  Built chatbot using Hugging Face LLM with prompt engineering.  Added filters for safe responses to health queries.  Deployed simple Streamlit app for user interaction. 4.End-to-End ML Pipeline for Telecom Churn Prediction  Built a Flask web app with an interactive form for real-time predictions  Performed data preprocessing and trained a Decision Tree model  Deployed the model for inference through the Flask app 5.News Topic Classifier (BERT on AG News)  Fine-tuned the bert-base-uncased model on the AG News dataset (4 classes: World, Sports, Business, Sci/Tech)  Evaluated model using Accuracy and weighted F1 metrics  Built interactive demo apps using both Streamlit and Gradio for real-time topic classification 6.Context-Aware Chatbot (Retrieval-Augmented Generation Demo)  Built a Streamlit app with LangChain–powered RAG using local text documents, FAISS indexing, and a Hugging Face embedding model  Implemented retrieval logic (demo_rag.py) to fetch context-relevant info from .txt files  Developed interactive Streamlit UI (app_streamlit.py) for user queries with index-based answer retrieval