Optimizing Hindi Paragraph Summarization through PageRank Method
Sheshang Degadwala, Dhairya Vyas, Khush Nilaykumar Patel, Mukesh Soni, Pavitar Parkash Singh, Ramya Maranan
Abstract
This research paper presents a novel approach to optimize Hindi paragraph summarization using the PageRank method. With the increasing volume of Hindi language content on the internet, there is a growing need for effective summarization techniques tailored specifically for Hindi text. The proposed method leverages the inherent structure and connectivity of Hindi paragraphs to extract the most important sentences for summarization. The PageRank algorithm is applied to construct a rank representation of the text, with sentences as nodes and their co-occurrence patterns as edges. By calculating PageRank scores iteratively, the algorithm identifies the most central and significant sentences in the paragraph. Linguistic features such as sentence length, position, and content relevance are incorporated to enhance the quality of the summaries. Additionally, the algorithm dynamically adjusts sentence importance based on their relations with other sentences rank. Experimental results on a comprehensive dataset demonstrate that the PageRank-based approach outperforms existing summarization techniques in terms of informativeness and coherence, as evidenced by higher similarity score. The proposed method offers an effective solution for optimizing Hindi paragraph summarization, facilitating the management and comprehension of Hindi text in various applications.