Team Members

Lara Ghoniem

Team Leader

Mohammed Yasser

Team Member

Salma Helmy

Team Member

Omar Wael

Team Member

Supervisors

Dr. Taraggy Ghanim

Assistant Professor

Eng. Mahmoud Heidar

Teaching Assistant

Abstract

Nowadays, a company’s reputation is one of the most important factors that determines its success. To help companies stay informed about what people are saying about them, we propose a system that enables clients to monitor various media platforms, including television, radio, podcasts, social media, blogs, and articles. By entering relevant keywords, clients can track brand mentions on a live dashboard, which offers data analysis and visualization features, including sentiment analysis, topic modeling, and text summarization. The system integrates several Application Programming Interfaces (APIs) and tools, such as the Bing Search API, Google Web Speech API, and Mazajak API, and also leverages Artificial Intelligence and Natural Language Processing techniques, including Support Vector Machine (SVM), Convolutional Neural Network-Long Short-Term Memory (CNN-LSTM), BERT2BERT, and BERTopic. With this innovative tool, companies can proactively manage their reputation and gain valuable PR insights, thereby increasing their chances of success.

System Objectives

• To provide a reliable Arabic media listening tool that meets user expectations.

• To provide monitoring of multiple sources, including television, radio, podcasts, social media, blogs, and articles.

• To provide a user-friendly interface that is easy to navigate.

• To provide sentiment analysis to help users understand customers’ feelings towards their brand.

• To provide a text summarization feature to provide users with a more context of mentions.

• To provide comprehensive analytics displayed on a dashboard on the brand listening tool website.

• To provide the ability to export analytics into daily and monthly reports.

System Scope

  • The system will be a website that provides a brand listening tool in Arabic.
  • Text and audio retrieved by the tool will be summarized.
  • Sentiment analysis is implemented to indicate the number of positive and negative mentions.
  • The data will be processed to perform spam detection, Franco-Arabic translation, and other data processing techniques to improve the quality of the retrieved data.
  • A live dashboard will be provided to display mentions found.
  • The tool will generate links for the mentions.
  • Daily and monthly reports will be produced.

Documents and Presentations

Proposal

You will find here the documents and presentation for our proposal.

Document

Presentation

SRS

You will find here the documents and presentation for our SRS.

Document

presentation

SDD

You will find here the documents and presentation for our SDD.

Document

presentation

Thesis

You will find here the documents and presentation for our Thesis

Document

Presentation

Accomplishments

Publications