Information extraction from wikipedia articles using DeepDive

Mining the required information from an enormous amount of data is a critical task. It is arduous to deal with the heterogeneity of astronomically immense data. Withal processing and analyzing it requires a number of resources. In the proposed work, information extraction (relation extraction between two Named entity mentions) from Wikipedia text articles using DeepDive has been carried out. We have experimented Deep-Dive's challenging capabilities for relation extraction, mention extraction and have calculated the probabilities of every variable being true via statistical learning and inference. Further we have tested results by plotting calibration plots on training data set and on whole data set by considering performance measures to check its efficiency on Wikipedia articles.