This new REU site will engage undergraduates in research in information retrieval with a focus on diverse data. The world is becoming increasingly digitalized. Digital data appear in diverse forms: textual, multimedia, georeferenced, and structured, for example. Retrieving information from large scale and diverse data poses unprecedented challenges. New paradigms need to be explored and next-generation computer scientists need to be prepared with advances in this field. This REU site provides opportunities for ten undergraduate students each summer to directly participate in research that propels Internet innovation in information retrieval.

This site emphasizes the participation of a diverse group of students, in particular, women, minorities, and students from community colleges. The research activities are structured using the Affinity Research Group (ARG) model, a cooperative approach for involving students with diverse backgrounds. The key elements of this REU experience are: 1) students are given the opportunities to work on research problems with practical applications; 2) students work in groups in a scholarly community that consists of mentors and practitioners who serve as role models and become part of students' virtual support group; 3) students participate in a series of field trips and workshops to bridge their transition to graduate programs; 4) students are closely mentored on posters and research reports preparation that culminates in a public oral presentation at Texas State at the end of each summer.

Intellectual Merit: All proposed projects are based on strong research bases derived from the synergistic research expertise from four faculty members in the Texas State Computer Science Department. Each proposed REU project will advance the state of the art in information retrieval. The tools developed by students will enable the community at large to leverage massive amounts of data on the Web for new knowledge discovery and contribute towards national cyberinfrastructure development. Ten weeks of in-depth summer research will increase retention of computer science students, improve their career perspectives, engage them to participate in the next frontier of Internet revolution, and motivate them to enter graduate computer science programs.

Broader Impacts: Because Texas State is ranked 13th among producers of Hispanic baccalaureate graduates in the nation, an REU site at Texas State has the potential to provide this audience with quality research experience. Furthermore, encouraging women and minorities to pursue advanced degrees in computer science will build a pipeline of future role models. The network of communities (faculty, students, computer scientists from industry) developed from this REU site will form a valuable support group for undergraduate students to talk and learn more about research as they prepare to enter graduate studies, and could develop into a collaborating relationship between Texas State and the participating students' institutions.

Project Report

The primary objective of REUIR (OCI-106243) was to motivate a diverse body of undergraduate students (female, minority, non-traditional, first generation college students) in the exploration of research processes in computer science and to inspire them to undertake graduate studies in computer science and/or careers in the IT field. The core aspects covered in this 9-week summer programs included new paradigms and architectures for information retrieval and Web search, multimedia information retrieval, spatial and temporal data retrieval, and discovery of Web data sources. The site official web site can be found in http://reu.cs.txstate.edu REUIR supported 32 student participants in the last four years. We are able to support two more students over the budgeted number due to the generous housing support provided by Texas State University. Among the participants, 5 were females, 5 were non-traditional students, one was American Indian, 5 were Hispanics or African Americans, and 8 were first generation college students. Overall, around 50% of the students were from the diverse group that we are targeting. Three students have obtained fellowships and started their graduate schools at University of Texas at San Antonio, University of Nebraska at Lincoln and Michigan University at Ann Arbor respectively. Five more students are planning to go to graduate schools encouraged by the REU experiences. Five students attended international conferences and presented their papers or posters. One student changed to Computer Science major as a result of participating in this program. The rest are still pursuing CS majors or working in the IT industry. The following is the list of papers published with REU students as co-authors over the last four years (mentors names are omitted here): "Geospatial Algorithms on GPU" by Gabriel A. DeLuna and Schiller D. Hill in the 19th ACM SIGSPATIAL International Symposium on Advances in Geographic Information Systems, Chicago, IL, USA, November 2011. "Discovery and Cataloging of Deep Web Sources" by C. Hicks and M. Scheffer in IEEE International Workshop on Data Integration and Mining (DIM-2012), Las Vegas, USA, July 2012. "Large Scale Similar Song Retrieval Using Beat-Aligned Chroma Patch Codebook with Location Verification" by J. Cabrera in International Conference on Signal Processing and Multimedia Applications (SIGMAP), Rome, Italy, July 2012. "A Subjective Logic Framework for Multi-Document Summarization" by Reed Coke, accepted to the 24th International Conference on Computational Linguistics (COLING'12), Mumbai, India, December 2012. "ServiceXplorer: A Similarity-based Web Service Search Engine", by S. Julian, accepted to ACM Conference on Information Retrieval (SIGIR 2014 Demo track), Gold Coast, Australia, July 2014 "SmartMart: IoT-based In-store Mapping for Mobile Devices", by Dylan Hicks, Kevin Mannix, Hannah Bowles in the First International Workshop on Internet of Things (C-IoT 2013), Austin, Texas, October 2013 "Social-PPM: Social-Aware Personal Process Management", by H. Memmi, J. Torres, The Second Asia-Pacific Conference on Business Process Management", APBPM 2014, Brisbane, Australia. The following data sets and systems were made available to the public as a result of REUIR: A synonymous image search dataset, a contour-based image retrieval dataset, and a proposal-based image testing dataset (http://mvlab.cs.txstate.edu) A mass-collaborative clustering Web search engine ClusteringWiki (http://dmlab.cs.txstate.edu/clusteringwiki) A user-centric Web search result organization system Rants (http://dmlab.cs.txstate.edu/rants) A Web service similarity search engine: ServiceXplorer (http://eil.cs.txstate.edu/ServiceXplorer) The program had trained undergraduate students in: Advanced technology development Background materials on information retrieval, machine learning, web crawling and web search Research processes and research ethics Research presentation and writing skills Poster preparation and what to expect during the poster judging session Applying to graduate schools and applying for NSF Graduate Fellowship How to give interesting research talks via faculty research speaker Life as a researcher via industry visits and what is involved in IT career. We engaged Dr. Heather Thiry, director of Golden Evaluation & Policy Research to formally evaluate our program. She administrated an online survey to all students in week 8. At the last week of the program, she visited our site in person to conduct a face-to-face focused group interview session with all REU students as well as attending students’ poster presentations. She also conducted an online survey regarding faculty mentors' perception of student learning for all mentors at the end of the program. In her summary, she wrote, ``in conclusion, all REUIR students at Texas State University reported multiple learning gains from their research experiences, including increased confidence in computer science, a greater understanding of the technical research process, technical skills, and teamwork skills. Students were also highly satisfied with other REU program elements, such as mentoring, housing, social activities, financial stipends, and orientation to research.''

Agency
National Science Foundation (NSF)
Institute
Division of Advanced CyberInfrastructure (ACI)
Type
Standard Grant (Standard)
Application #
1062439
Program Officer
Sushil K Prasad
Project Start
Project End
Budget Start
2011-03-01
Budget End
2015-02-28
Support Year
Fiscal Year
2010
Total Cost
$359,827
Indirect Cost
Name
Texas State University - San Marcos
Department
Type
DUNS #
City
San Marcos
State
TX
Country
United States
Zip Code
78666