State of the Art in Multilingual Text Retrieval Accessing Parallel Web Pages

1588 Words Jan 26th, 2018 6 Pages
Multilingual search engines typically consist of a crawler which traverses the web, retrieves the required web page in the desire languages. It provides front end user interface, which can be used for selecting language for query submission. The way the query is fired leads into two types of search Cross-language information retrieval and Multi-language information retrieval. In cross-language retrieval the user query is machine translated into multiple language queries automatically as per user selection and then fired. In multi-language retrieval the user has to provide queries in multiple languages to fetch the web documents in different languages. Also in some search engine the web page is machine translated and forwarded to the user. NLP is still in the expansion phase and has to make advances in it, research is going on all over world. Experiment was done with top rated multilingual search engine to access parallel pages but could not find it. When billions of parallel web documents are present on the web in different languages why not explore that? A alternative to that can be searching through parallel pair finder.

Keywords: NLP,MT Machine Translation, CLIR Cross-language, information retrieval, Multilingual.

1. Introduction
The total number of languages in the world is between 5,000 and 10,000. In a country like India people speak about 57 different…

More about State of the Art in Multilingual Text Retrieval Accessing Parallel Web Pages

Open Document