Last updated: December 01, 2012

National Library of Medicine
Communications Engineering Branch/MSC 3824
Bldg. 38A, Room 10S1015H
8600 Rockville Pike
Bethesda, MD 20894 USA

(301) 435-3245

Dr. Xiaoli Zhang joined the Communications Engineering Branch of National Library of Medicine in February 2008. She received the M.E. degree in Applied Mathematics and the Ph.D. degree in Electrical Engineering both from Rensselaer Polytechnic Institute in 2007, and the B.E. degree in Electrical Engineering from Hefei University of Technology, China, in 1993. Her research interests include pattern recognition, text mining, information retrieval, document analysis and understanding.

Current Projects

Publisher Data Review System (PDRS): this project is to automate the production of citations by extracting bibliographic data (author, title, abstract, etc.) from Web-based online medical articles. I have developed modules to help extract metadata such as grant support, investigator names and references from HTML-formatted articles using machine learning algorithms.

Automatically Creating OLDMEDLINE Records for NLM (ACORN): this project is to automatically create OLDMEDLINE records from the scanned Quarterly Cumulative Index Medicus (QCIM) documents. I am responsible for the GUI design for ACORN which includes two parts: Records Preparation (Window-based application) and Records Reconciliation (Web-based application). Records Preparation is to process scanned document images, detect multiple records of a same citation and utilize the cross-reference information to correct OCR errors in duplicate records, based on which an OLDMEDLINE record is created automatically. Records Reconciliation is to further correct errors and create final OLDMEDLINE records through user-interaction.