#🤖 DeveloperHub-AI-ML-internship-Phase-2

📌 Overview

This repository contains my Phase 2 AI/ML Internship tasks focused on Natural Language Processing (NLP) using BERT and Hugging Face Transformers.

The projects demonstrate the practical implementation of transformer-based models for solving real-world NLP problems, including text classification, question answering, and named entity recognition.

---

🎯 Objectives

• Learn Transformer-based NLP models
• Understand BERT architecture and applications
• Implement pre-trained Hugging Face models
• Perform text classification and information extraction
• Build practical AI applications using Python

---

📂 Repository Structure

AI-ML-Internship-Phase2/
│
├── Task1_News_Topic_Classifier_BERT/
│   ├── Task1_News_Topic_Classifier_BERT.ipynb
│   ├── README.md
│   ├── requirements.txt
│   ├── .gitignore
│   └── outputs/
│
├── Task4_Question_Answering_BERT/
│   ├── Task4_Question_Answering_BERT.ipynb
│   ├── README.md
│   ├── requirements.txt
│   ├── .gitignore
│   └── outputs/
│
├── Task5_Named_Entity_Recognition_BERT/
│   ├── Task5_NER_BERT.ipynb
│   ├── README.md
│   ├── requirements.txt
│   ├── .gitignore
│   └── outputs/
│
└── README.md

---

🚀 Projects Included

📰 Task 1 – News Topic Classification using BERT

Description

Built a News Topic Classification model using a pre-trained BERT model from Hugging Face Transformers. The model classifies news articles into multiple categories using the AG News dataset.

Key Features

• BERT-based Text Classification
• AG News Dataset
• Data Preprocessing
• Tokenization
• Model Training
• Performance Evaluation
• Accuracy & Loss Analysis

---

❓ Task 4 – Question Answering using BERT

Description

Developed a Question Answering system using a pre-trained BERT model capable of extracting accurate answers from a given context paragraph.

Key Features

• Context-based Question Answering
• Pre-trained DistilBERT Model
• Interactive User Input
• Multiple QA Examples
• Confidence Score Prediction

---

🏷️ Task 5 – Named Entity Recognition (NER) using BERT

Description

Implemented a Named Entity Recognition system using a pre-trained BERT model to identify entities such as persons, organizations, and locations from text.

Key Features

• Person Detection
• Organization Detection
• Location Detection
• Multiple Examples
• Interactive User Input
• Confidence Scores

---

🛠 Technologies Used

• Python
• Jupyter Notebook
• Hugging Face Transformers
• BERT
• DistilBERT
• PyTorch
• Pandas
• NumPy

---

📸 Outputs

Each project contains an outputs/ folder including:

• Model Loading
• Predictions
• Evaluation Results
• Custom Examples
• Interactive Demonstrations

---

📖 Learning Outcomes

By completing Phase 2, I gained practical experience in:

• Transformer Models
• BERT Architecture
• Natural Language Processing (NLP)
• Text Classification
• Question Answering
• Named Entity Recognition
• Hugging Face Transformers
• PyTorch Pipelines

---

▶️ Getting Started

Clone the repository:

git clone https://github.com/choudharyasim0/AI-ML-Internship-Phase2.git

Move into the project directory:

cd AI-ML-Internship-Phase2

Install dependencies:

pip install -r requirements.txt

Launch Jupyter Notebook:

jupyter notebook

Run the notebook of the desired task.

---

📌 Internship Skills Demonstrated

• Natural Language Processing
• Transformer Models
• BERT
• Hugging Face Transformers
• Deep Learning
• Information Extraction
• AI Model Evaluation
• Python Programming

---

👨‍💻 Author

M Asim Imtiaz

BS Artificial Intelligence
Superior University

Connect with Me

• GitHub: https://github.com/choudharyasim0
• LinkedIn: https://www.linkedin.com/in/m-asim-imtiaz-1161623a4

---

⭐ If you found this repository helpful, don't forget to star it.