Bilingual data extraction and auto summarisation

Text Extraction plays an important role in finding valuable information. As a result of rapid growth of available multimedia documents on web and increasing requirement for information, identification, indexing and retrieval, several research have been done on text extraction. Here, a methodology to build a system for text extraction has been presented that is designed to provide support to the applications like Natural Language Processing that requires the data to be in specific format. The data from various formats is extracted and is easily exported to a database. So this tool will be highly useful for creation of data repository and data bank. Also, to speed up the extraction from large size files, research was done and so, large size files are split into small size chunks and efficient use of parallel processing is done to provide faster results. Along with that, Bilingual support is provided, i.e., files of both English and Hindi language can be easily extracted, building respective data banks. Keywords are the important terms, sometimes called, index terms that contain some kind of valuable information about the document. Automatic keyword extraction is the task to identify a small set of words, which can be designated as keywords for that document, and thus provide a brief idea about the document. An effective Keyword Extraction technique has been discussed that can be applied to extract keywords effectively, that uniquely identify a document. A summary is a shortened version of a text of the document that describes the main points of the original document in brief. Automatic summarisation is the task of generating a summary of any input document by the computer. This summary can be used instead for reading the whole content/text resulting in particularly interested.