A Summarizer for Tamil Language Using Centroid Approach

2. Literature Survey

The paper focus on summarizing Persian language, which the authors have termed as AZOM. The proposed approach combines statistical and conceptual property of text in regards of document structure that extracts summary of text. AZOM is capable of summarizing unstructured documents. The result presented is also much superior than common structured text summarizers and much better than other Persian text summarizers (Zamanifar & Kashefi, 2011).

Rapid growth of Internet has led to the emergence of large of amount of data. India has diversified languages and hence forth summarization technique differs and so as generated summaries. The authors have proposed a system for detection and removal of deadwood (meant as word or phrase that can be omitted without any loss) for Punjabi language. The steps involves sentence segmentation and removal of stop words which is considered to be the preprocessing task. The second task is assigning weights taking into account five different features to the textafter tokenization. Sentence with higher scores are eliminated and finally deadwood is eliminated and removed from summary.