Many students, teachers and researchers view thousands of websites a day as they scour for information. It takes up hours and even days to find the information they need, and most webpages might not even yield the desireed information. Entering a webpage's information to retrieve a summary would be extremely time-saving, removing trouble of reading and comprehending information that they might not need.
An open-source extractive algorithm which can summarise and extract key bits of information from a given paragraph and output its summary. This project hinges on the one key library which is the Natural Language Toolkit. The package helps to manipulate the paragraphs and sentences to let us calculate the weighted and unweighted frequencies of words and sentences alike.
This doesn't end here though, I'm planning to take this project up by
- Adding a web interface so people can access it and summarise their paragraphs
- Enhancing the algorithm to make it abstractive
- Using transformers and libraries to make summaries more dynamic
- Using Google APIs to allow users to enter search queries and get webpages to scrape and summarise