Abstract

Abstract: It isimportant to derive the requisite data from an incredibly high volume of data.The heterogeneity of astronomically large data is difficult to deal with. Avariety of resources are needed for processing and analysis. In the worksuggested there was an extraction of details from Wikipedia text articles usingDeepDive (extraction of relationships between two Nominated individualmentions). We have observed the daunting capabilities of DeepDive forrelationship extraction, listed extraction, and measured the probability thateach variable is valid with statistical learning and inference.Furthermore, we evaluatedfindings by drawing calibration plots of training and data set, taking outputtests in Wikipedia articles to determine its effectiveness.

Keywords—DataMining, DeepDive, Wikipedia, Text extraction, Summarization.