Professor Joe Tekli gave a research seminar on similarity-based data processing in CIS

5/11/2018

Dr. Joe Tekli, Assistant Professor in Computer Engineering at Lebanese American University and Visiting Fulbright Scholar in the CIS Department gave a research seminar on Similarity-based Semi-structured and Semantic Data Processing and Applications.

Dr. Joe Tekli, Assistant Professor in Computer Engineering at Lebanese American University and Visiting Fulbright Scholar in the CIS Department gave a research seminar on Similarity-based Semi-structured and Semantic Data Processing and Applications. 

In recent years, the growing amount of semi-structured data on the Web (e.g., XML and NoSQL data) has underlined the need to manage their interoperability, i.e., the ability to automatically interpret information in Web documents exchanged between different sources, necessary for efficient information management and search applications. In this context, the main theme of our research efforts in the past few years has been the study of semi-structured data similarity and its applications for data integration, extraction, search, retrieval, and related functions. Dr. Tekli and his group have addressed XML similarity at three layers: data layer (document comparison), type layer (grammar comparison), and in between the data and type layers (document/grammar comparison). They have developed dedicated semantic analysis and disambiguation methods to assign well-defined meaning to XML elements and attributes, as well as semantic-aware querying solutions to search structured and NoSQL attribute-value items. Various research projects have branched out from the latter, ranging over RSS merging, SOAP multicasting, SVG annotation, RDF normalization, as well as multimedia event detection and linkage, among others. 

Back to top of page