Information Retrieval and Visualization for Searching Scientific articles and Patents

Given the rapidly changing face of technology, keeping up with the trends and identifying potential areas to be explored for research or commercialization is a challenging task. Decision makers, research analysts, scholars, research directors all make use of digital collections, use of which is facilitated by search applications developed on top of them. However, search is a humandriven activity and the result of such analysis is largely dependent on the initial inputs that are provided by the expert. Besides, aggregating and assimilating all the information returned by a search engine is no less daunting. In this paper, we propose intelligent methods for presenting search results to help information assimilation. We also present methods for analyzing large collections of documents in an automated way to generate insights that can prove to be useful for analysts. Starting from time-stamped collections of research publications and patent documents, we present several Information retrieval (IR) techniques that can successfully extract and present insights about emerging, popular and receding trends in research along with their current levels of commercialization. We present results of experiments based on research abstracts made available by digital libraries and US patent office.