Medical Health Posts Summarization Using Lesk Algorithm
Published: 2019
Author(s) Name: Vinod L. Mane, Ashwini Abhale and Sumit Khandelwal |
Author(s) Affiliation: Assistant Professor, IT Dept., DYPCOE Akurdi, Pune, Maharashtra, India.
Locked
Subscribed
Available for All
Abstract
Today’s world is all about information, mostly online. With the growth of internet many communication technologies emerged quickly as important information sources, such as emails, forums, social networking sites, etc. It is time and space consuming to handle such large amount of data. Text summarization is technique by which important portion of text are obtained. Traditionally these important portions are selected based on frequency of keywords, position of sentence, style of writing word, keywords in title, etc. Extractive text summarization is produced by concatenating several sentences taken exactly as they appear in the text. Sentences are selected based on some scoring techniques. In our approach, we are using simplified Lesk algorithm with some modification. Our approach is applicable to medical health posts. In this, sentences having important information from medical perspective are arranged in decreasing order of their weights. Based on given input percentage, relevant number of sentences is given as summary. We compared results with human expert summary. The proposed approach gives promising results.
Keywords: Health posts, Lesk algorithm, Summarization, UMLS.
View PDF