Over the past 10 years, he has been sole inventor of software applications that support tasks ranging from text classification, social and blog data capture, realtime and historical Twitter data access, machine-learning, and manuscript review management, to the development of better annotation tools for improving optical character recognition. His current research focuses on the use of the five pillars of text analytics in the detection of threats, fear, bots, fake news, document relevance, and novel ideas communicated in public comments, blogs, emails, and many types of social media. He conducts regular workshops at top universities around the world with a particular focus on the collection and analysis of Twitter data.
In collaboration with Mark J. Hoy, he was awarded a patent titled “System and method of classifier ranking for incorporation into enhanced machine learning,” known in the colloquial sense under the Sevice Mark “CoderRank,” which is the core IP. Texifter helps researchers in academia, business, and government to quickly and efficiently find the best available human annotators to label their data.
A public lecture by Dr. Shulman: “Text Analytics for Social Data Using DiscoverText & Sifter” – @SMLabTO Research Methods Workshop