text-summarizer-aweebtaku 1.2.0

A text summarization tool using GloVe embeddings and PageRank algorithm

Text Summarizer

A Python-based text summarization tool that uses GloVe word embeddings and PageRank algorithm to generate extractive summaries of documents.

Features

• Extractive Summarization: Uses sentence similarity and PageRank to identify the most important sentences
• GloVe Embeddings: Leverages pre-trained GloVe word vectors for semantic similarity calculation
• Multiple Input Methods: Support for single documents, CSV files, or interactive creation
• GUI Interface: User-friendly Tkinter-based graphical interface
• Command Line Interface: Scriptable command-line tool for automation
• Batch Processing: Process multiple documents at once

Installation

Prerequisites

• Python 3.8 or higher
• Required packages (automatically installed): pandas, numpy, nltk, scikit-learn, networkx

Install from PyPI

pip install text-summarizer-aweebtaku

Install from Source

1. Clone the repository:

git clone https://github.com/AWeebTaku/Summarizer.git
cd Summarizer

2. Install the package:

pip install -e .

Download GloVe Embeddings

No manual download required! The package will automatically download GloVe embeddings (100d, ~400MB) on first use and cache them in your home directory (~/.text_summarizer/).

If you prefer to use your own GloVe file, you can specify the path:

summarizer = TextSummarizer(glove_path='path/to/your/glove.6B.100d.txt')

Usage

Command Line Interface

# Summarize a CSV file
text-summarizer-aweebtaku --csv-file data/tennis.csv --article-id 1

# Interactive mode
text-summarizer-aweebtaku

Graphical User Interface

# Launch GUI (easiest way)
text-summarizer-aweebtaku --gui

# Or use the dedicated GUI command
text-summarizer-gui

Python API

from text_summarizer import TextSummarizer

# Initialize summarizer (automatic GloVe download)
summarizer = TextSummarizer(num_sentences=3)

# Simple text summarization
text = "Your long text here..."
summary = summarizer.summarize_text(text)
print(summary)

# Advanced usage with DataFrame
import pandas as pd
df = pd.DataFrame([{'article_id': 1, 'article_text': text}])
scored_sentences = summarizer.run_summarization(df)
article_text, summary = summarizer.summarize_article(scored_sentences, 1, df)

Data Format

Input data should be in CSV format with columns:

• article_id: Unique identifier for each document
• article_text: The full text of the document

Example:

article_id,article_text
1,"This is the first article. It contains multiple sentences..."
2,"This is the second article. It also has several sentences..."

Algorithm

The summarization process follows these steps:

1. Sentence Tokenization: Split documents into individual sentences
2. Text Cleaning: Remove punctuation, convert to lowercase, remove stopwords
3. Sentence Vectorization: Convert sentences to vectors using GloVe embeddings
4. Similarity Calculation: Compute cosine similarity between all sentence pairs
5. PageRank Scoring: Apply PageRank algorithm to identify important sentences
6. Summary Extraction: Select top-ranked sentences in original order

Configuration

• glove_path: Path to GloVe embeddings file (default: 'glove.6B.100d.txt/glove.6B.100d.txt')
• num_sentences: Number of sentences in summary (default: 5)

License

This project is licensed under the MIT License - see the LICENSE file for details.

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

Citation

If you use this tool in your research, please cite:

@software{text_summarizer,
  title = {Text Summarizer},
  author = {Your Name},
  url = {https://github.com/AWeebTaku/Summarizer},
  year = {2024}
}