Extracting The User’s Perception About Aadhaar From Print Media Articles

With the advent of technology, agencies have started storing identity data of the users in distributed repositories for improved targeting and ease of the implementation of the commercial and governmental policies. Social networking portal, public registers, social security portals, various ID’s and e-commerce portal are among the popular agencies, who are using these disintegrated yet identifiable digital identities for individuals. This study aims to investigate the user’s perceptions about these stored identities. An experimental test-bed of this research is Aadhaar, the Indian identity repository where media and public reaction has been widely different.As observed in literature, texts generated in un-controlled environment such as Twitter, Facebook etc. are noisy, unstructured and short,it is imperative for us to conduct a pilot on controlled data such as a corpus of newspaper articles. English language articles published in the `The Hindu’ and the ‘ProQuest’ are used as the primary data to extract the view’s in terms of topic, though Internet or the online texts are not all-inclusive. Machine learning tools, such as topic modeling and sentiment analysis are employed to extract the perceptions. The impact of the articles of the aforesaid media houses are also analyzed in terms of neutral, positive and negative polarity