Siva Kumar Cheekula

Master's in Computer Science / siva@knoesis.org

About

I am Siva Kumar Cheekula, a graduate research assistant at Kno.e.sis Center, Wright State University. I am pursuing Master's in Computer Science, advised by Dr. Amit P Sheth - LexisNexis Ohio Eminent Scholar, and Director of Kno.e.sis, an Ohio Center of Excellence in Knowledge Enabled Computing. As a research assistant, I've contributed to research efforts involving Machine Learning, Natural Language Processing and Semantic Web technologies.

Resume

Education

Masters in Computer Science

Wright State University, USA

Graduating in April 2016

Bachelors of Technology in Computer Science

Jawaharlal Nehru Technological University, India

May 2009

Experience

Research Assistant

Kno.e.sis Center, Wright State University

Jan 2014 ~ Current

Research Intern

ezDI, LLC

Summer 2014

Teaching Assistant

Wright State University

Fall 2013

Test Engineer

Bwin.party Digital Entertainment

Nov. 2011 - Aug. 2013

Programmer Analyst

Cognizant Technology Solutions

Mar. 2010 - Oct. 2011

Projects

Projects

Crowd Sourced Taxonomy for Entity Recommendations

Entity recommendation systems are enormously popular on the Web. These systems harness manually crafted taxonomies for improving performance of recommendations. However, manually creating a taxonomy is time, labor intensive process and might require timely updates. This challenge can be addressed by utilizing crowd sourced taxonomies. In this project, I've explored a taxonomy extracted from crowd sourced knowledge base - Wikipedia - for entity recommendations. Investigated and evaluated an entity recommendation algorithm utilizing crowd sourced knowledge base and addressed the challenges in utilizing a crowd sourced taxonomy as background knowledge base.

Entity Recommendations using Hierarchical Knowledge base: An adaptation of spreading activation theory is implemented on the hierarchical category structure of Wikipedia for identifying user interests. By implementing a reverse spreading mechanism, I've explored approaches for recommending new entities to users. Evaluation on Movielens dataset has shown the hierarchical knowledge bases can be utilized for personalized recommendations. [paper]

Characterizing Concepts in Taxonomy for Entity Recommendations (Master's Thesis): Although taxonomies were utilized as background knowledge bases, their potential was not completely explored. I have proposed three prominent characteristics of concepts in a taxonomy that can impact entity recommendations. Also, investigated several approaches for measuring each characteristic and evaluated their impact on entity recommendations. The evaluation on two diverse data-sets suggests the characteristics have significant impact on recommendations performance.

Oct 2014 ~ Current

Hazards SEES

Infrastructure systems are a cornerstone of civilization. Damage to infrastructure from natural disasters such as an earthquake (e.g. Haiti, Japan), a hurricane (e.g. Katrina, Sandy) or a flood (e.g. Kashmir floods) can lead to significant economic loss and societal suffering. Human coordination and information exchange are at the center of damage control. This project seeks to radically reform decision support systems for managing rapidly changing disaster situations by the integrated exploitation of social, physical and hazard modeling capabilities. [wiki]

Aug 2015 ~ Dec 2015

Automated Clinical Document Improvement

Quality of clinical documentation is important for patients as well as hospitals, and government agencies. Beyond patient care, documentation is critical for financial assessment, fraud prevention and clinical research. Assuring the quality of documents has been manually performed by CDI (Clinical Document Improvement) personnel. The manual analysis of clinical charts is time consuming and requires expert knowledge for assessment. We investigated several approaches to automatically identify clinical charts that may have incomplete information, so as to prioritize documents for review. Our approach utilizes domain's background knowledge base as well as machine learning techniques to identify documents with incomplete information.

May 2014 ~ Aug 2014

Publications

Publications

Siva Kumar Cheekula, Pavan Kapanipathi, Derek Doran, Amit Sheth. Entity Recommendations Using Hierarchical Knowledge Bases. 4th International Workshop on Knowledge Discovery and Data Mining Meets Linked Open Data, 2015

Siva Kumar Cheekula, Pavan Kapanipathi, Amit Sheth. Characterizing Concepts in Taxonomy for Entity Recommendations. Technical report. [work in progress]

Pavan Kapanipathi, Siva Kumar Cheekula, Prateek Jain, Chitra Venkataramani, Amit Sheth, Derek Doran. Hierarchical Knowledge Bases to Identify User Interests on Social Media. [work in Progress]