Quantifying Summary Effectiveness with Text Similarity Methods

DOI

No DOI

Published In

Volume 4, Issue 6 (2024)

Published Online

November-December 2024

Pages

04-06

Article Type

Research Article

Check for updates

No DOI available for cross-reference

Text summary evaluation could be broadly categorized into two types: extrinsic evaluation and intrinsic evaluation. Extrinsic evaluation focuses on the impact of summarization on other tasks, while intrinsic evaluation determines the summary quality on the basis of comparison between the automatically generated summary and the human generated summary. There are intrinsic evaluation methods to check the summarization system by itself available in the literature by comparing it with human made summary (Saiyed & Prithi 2017). When impact of the summaries is to be measured extrinsic evaluation plays a prevalent role. There are many tasks such as text classification, information retrieval and text similarity detection for analyzing the impact of text summaries generated. Text similarity plays an important role in text summarization systems and it is a core part of information retrieval and processing systems. This inspired the development of text similarity detection system for comparing the text summaries which serves as a measure for similarity detection. Key Words: Text Similarity, Knowledge-Based Similarity, Graph Databases, Jaccard coefficient