K. J. Somaiya Institute of Engineering & Information Technology, Sion - Mumbai Department of Computer Engineering TEXT & ARTICLE SUMMARISATION APP Amogh Joshi TYCOMPS B - 20 Prathamesh More TYCOMPS B - 33 Soumya Shah TYCOMPS B - 54 Supervisor Prof. Aarti Sahitya.
OUTLINE OF PROJECT. Introduction Problem statement & Project Objective Target audience Deployment Cost Software and Hardware requirement Proposed system / idea Security Features Conclusion References.
INTRODUCTION. These days, content creation has dominated the internet and the social media fields. For most people, it is not only taxing but also inconvenient and repetitive to manually go through and peruse such content. Many a times, content creators as well as news websites have a lot of incomplete, redundant and repetitive content put up. Article and Text Summariser aims to not only help alleviate some of the burden from curators, but also keep track of important information that has been put on news websites while discarding the redundant bits. The text summariser will also aim to accurately summarise the information discussed in an article or an input text, and display the key points in short and easy to understand paragraphs..
PROJECT OBJECTIVE. Accurately Design Text Summariser Design User Friendly and Interactive user interface.
TARGET AUDIENCE. Corporate Managers Teachers, Institute Faculty Students, with little time Researchers Content Writers News Curators.
CHALLENGES AND COST. Summarisation Accuracy - The text summarisation should be accurate, covering important points. Summarisation Discarding Accuracy - The unimportant information should be rightly discarded, or reduced. Cost - As most of the software will be built-upon existing bases, using freeware and other open-source libraries, the cost of the project will be negligible..
SOFTWARE & HARDWARE REQUIREMENTS. Hardware Requirements Intel i3 or Equivalent Internet Connection Web Browser e.g. Google Chrome Software Requirements Python 3 BART Flask.
SECURITY FEATURES. Standard Security measures will be implemented. Text input will not be shared with other users Since the output text will not be stored, there is no way it can be accessed in an unauthorised way Summarised text will be secured, and not shared publicly.
PROPOSED SYSTEM. A web application ( Web App ) give the users a summarized text from text input. Input Box : The input will be entered into an input box, similar to many word processing websites. Text summarizer : The text input will be summarized, with important points displayed while unimportant text is left out. Display Box : The summary text will be displayed in a designated area..
CONCLUSION. The Application will help users lessen the burden of news curators and news curation websites, which aim to deliver news to their end-users in a small bite format. It will also help in summarizing the texts, to only keep the highlights, discarding the unimportant details so the data becomes cleaner, easier to read and understand. The App will have an easy-to-use interface, able to be accessed by even non-technical users..
REFERENCES. https://github.com/ohmeow/blurr https://towardsdatascience.com/bert-explained-state-of-the-art-language-model-for-nlp-f8b21a9b6270 https://huggingface.co/.