My primary research interests lie in Semantic Web, Knowledge Representation, Computer Networks, Service-Oriented Computing, and Ubiquitous and Pervasive Computing. The goal of my research is to analyze ways to describe data and services meaningfully, along with designing frameworks whereby agents can automatically find information and create different applications geared toward users.
Federal Reserve:
I conduct research in semantic web technologies and text analytics to add structure and meaning to financial data and design frameworks for predictive modeling.
Hofstra University:
I led projects at Hofstra University's Big Data Lab and explored problems related to user-generated data on the web. In particular, I looked at two types of data: structured and unstructured.
Currently, the web is becoming more structured with linked data annotations. Some examples of linked data annotations are: Facebook’s Open Graph Protocol, Schema.org, and WikiData. Although linked data’s prevalence on the web is increasing, it is not yet clear as to how this information is being used. My research sought to answer the following questions: how descriptive should the ontology languages for linked data be; what are the error rates of the linked data markups; and how can we expand the ontologies and correct these errors.
I worked in collaboration with Nuance Communications where we looked at how often various parts of the schema.org ontology (and in the future other types of linked data such as DBPedia and WikiData) are necessary by analyzing how user-generated schema.org annotations are used. We hoped to use this information to produce a specification for the schema.org language based on how schema.org markup is actually used. We used a large Schema.org data set is used from Web Data Commons, the Virtuoso Universal Server, a quad store for linked data for storage and issuing SPARQL queries and the JENA API for ontology processing.
Bell Labs:
Web service mashups are the recent trend on the Internet and allow users to create their personal Web space from different types of sources such as social networking, mapping or photo APIs. The proliferation of Web services creates the problem of performing more sophisticated query matching, such as allowing one to query for a combination of different services in a single search. Additionally, these services are being developed independently and increasing exponentially, as evidenced in the Programmable Web, an API directory offering a listing of over 10,000 Web service APIs. Given these current trends, it is clearly evident that service APIs are generating huge volumes of data for communication and commerce. Thus, an improved search mechanism for services has become a vital necessity.
I have conducted research in the following areas in order to improve context-aware service discovery and composition: 1) service description and discovery: creating an ontology classification and a faceted search interface for APIs in the Programmable Web directory; 2) context-aware search: developing a framework for personalized context-aware search of ontology-based tagged data; 3) service composition: incorporating a rule-based context-aware front-end system that maps user- based rules to service composition templates.
Projects: