International Journal of Computer Applications (0975 – 8887) Volume 165 – No.11, May 2017 29 A Survey of Text Summarization Techniques for Indian Regional Languages Sheetal Shimpikar PG Student Department of Computer Engineering-PCE, Mumbai University, India Sharvari Govilkar H.O.D Department of Computer Engineering-PCE, Mumbai University, India ABSTRACT Summarization is the process of reducing a text document with a computer program in order to create a summary that retains the most important points of the original document. Technologies that can make a coherent summary take into account variables such as length, writing style and syntax. The main idea of summarization is to find a representative subset of the data, which contains the information of the entire set. Text summarization is commonly used to handle summaries of email threads, action items from a meeting and simplifying text by compressing sentences used to manage knowledge and also to help Internet search engines. This paper gives comparative study of various text summarization techniques used for Indian regional languages and also discusses in detail two main Types of Text summarization techniques these are extractive and abstractive. Keywords NLP, text summarization, text summarization techniques, extractive, abstractive, features, Rich Semantic graph, TF-IDF, NLG. 1. INTRODUCTION Text summarization is reducing a text with a computer program in order to create a summary that retains the most important points of original text. The main idea of summarization is to find a representative subset of the data, which contains the information of the entire set. Text summarizations choose the most significant part of text and create coherent summaries that state the main purpose of the given document. Text summarization can be categorised into given approaches. Single document summarizer means the summary is extracted from a single document, multiple document summarizer means the summary is extracted from a multiple document, generic document summarizer generates summaries containing main topics of document, query based document summarizer generate summaries containing sentences that are related to given queries . Abstractive and extractive summarization methods are used. 2. LITERATURE SURVEY In this section we cite the relevant past literature of research work done in the field of text summarization techniques for Indian languages. Sunitha.C, Dr.A.Jaya and Amal Ganesh worked on Abstractive summarization techniques in Indian languages. Authors have explained Abstractive summarization technique, classified in two approaches such as structure based approach and semantic based approach. There different methods are used in these approaches [1]. Jagadish S Kallimani suggests a solution for abstractive summarization by making use of extractive methodology in Kannada language. Performance accuracy for Literature 70%, for Entertainment 80%, for Sports 76%. The main idea is to generate abstractive summary by gathering key concepts from source document using extractive summary technique [2]. Manjula Subramanian discusses about semantic graph reducing technique to generate abstractive summary with input text in Hindi [3]. Rajina Kabeer used semantic Graph based method which concentrate on summarizing documents in Malayalam [4]. Atif Khan and Naomie Salim have worked on a paper of review on abstractive summarization methods [5]. Barzilay and Mckeown used Tree based techniques for text representation on dependency based representation: DSYNT tree. Content selection is theme intersection algorithm summary generation uses FUF/SURGE language generator [6]. Harabagiu and Lacatusu worked on template based method [7]. Lee and Jian worked on ontology based method, text representation was on Fuzzy ontology [8]. Green backer used multimodal semantic model [9]. Genest and Lapalme used INIT based method. Text representation on abstract representation information item [10]. Aadia Abbas Mohammed Elsied worked on automatic abstraction summarization a systematic literature review on abstractive summarization [11]. Rosna P Harun worked on text summarization methods in Dravidian language contains languages like Malayalam, Tamil, Telugu, Kannada, Kodagu, Badaga, Byari, and Tulu. [12]. Renjith SR have worked on automatic text summarization for Malayalam using sentence extraction [13]. Krish Perumal proposed a language independent sentence extraction based text summarization technique for English and Tamil [14]. Prajitha U proposed an algorithm namely LALITHA: A light weight Malayalam stemmer using suffix stripping method [15].