Hindi Text Summarization Using the Improved Dingo Optimization

Document summarization has become an area of considerable interest across various fields, with approaches generally falling into two categories: extractive summarization and abstractive summarization. Extractive summarization generates a summary by selecting important sentences or phrases from the original text, ensuring that the key ideas are retained without altering the wording. In contrast, abstractive summarization generates sentences that succinctly represent the central concepts of the text more concisely and coherently. Both approaches face notable challenges, including errors in sentence selection, limited content coverage, and redundancy in generated summaries. To overcome these limitations, we introduce an innovative approach that utilizes the Improved Dingo Optimization (IDO) algorithm for the summarization of Hindi text. Before applying the IDO algorithm the input text is pre-processed, and feature extraction is done focusing on the coverage and redundancy features, two critical aspects for effective summarization. The IDO algorithm is then applied to optimize these features, improving the overall quality of the generated summaries. The proposed work was tested on the FIRE 2011 Hindi dataset, and its performance was compared with existing summarization techniques. The experimental results highlight notable improvements in accuracy and summary quality, highlighting the efficiency of the proposed method.