Large Web Corpora of High Quality for Indian Languages

Large textual resources are the basis for a variety of applications in the field of corpus linguistics. For most languages spoken by large user groups a comprehensive set of these corpora are constantly generated and exploited. Unfortunately for modern Indian languages there are still shortcomings that interfere with systematic text analysis. This paper describes the Indian part of the Leipzig Corpora Collection which is a provider of freely available resources for more than 200 languages. This project focuses on providing modern text corpora and wordlists via web-based interfaces for the academic community. As an example for the exploitation of these resources it will be shown that they can be used for the visualization of semantic contexts of terms and for language comparison