NLP-Powered GUI Application

This project is a desktop GUI application that combines four essential Natural Language Processing (NLP) tasks into a single, userfriendly

tool:

Sentiment Analysis

Named Entity Recognition (NER)

Machine Translation (English → Arabic)

Text Summarization

Built with Tkinter for the interface and powered by Machine Learning and Deep Learning models, the application enables users to

analyze, translate, and summarize text in real-time.

Functionalities

1. Sentiment Analysis

Classifies input text as positive, negative, or neutral

Built using Logistic Regression, Naïve Bayes, and SVM

Best-performing model selected and integrated into the GUI

2. Named Entity Recognition (NER)

Detects people, locations, organizations, and other entities in text

Developed using LSTM and BiLSTM models

Preprocessing included text cleaning, padding, and numerical encoding

3. Machine Translation (English → Arabic)

Implemented using Encoder-Decoder LSTM architecture

Training leveraged teacher forcing, tested with unseen text for real-time translation

4. Text Summarization

Automatically generates concise summaries from long documents

Built using a sequence-to-sequence (Seq2Seq) LSTM model

Dataset included paired texts and summaries from news and articles