Research

Research Interests

Scholarly Communication for knowledge discovery

Nowadays science is being conducted in a very different way than it was twenty years ago. Vast amount of publicly available knowledge (e.g., datasets, publications, patents, case studies and tools) and the exponential-growing computing power have enabled this change to occur. No scientists can finish reading all the related articles, let alone browsing all the related datasets, patents and tools. The current amount of published knowledge is beyond what a single scientist can consume, and knowledge transfer has been restrained due to the limit of human cognition. New way of conducting science is highly demanded. Here we propose the new concept of entitymetrics which uses entities (i.e., evaluative entities or knowledge entities) in the measurement of impact, knowledge usage, and knowledge transfer to facilitate knowledge discovery. This extends scholarly communication by emphasizing the importance of entities, which are categorized as macro-level entities (e.g., author, journal, article), meso-level entities (e.g., keyword), and micro-level entities (e.g., dataset, method, domain entities). These entities can be analyzed from the temporal perspective to capture dynamic changes or from the spatial dimension to identify geographical differences. Entitymetrics focuses on both knowledge usage and discovery and can be viewed as the next generation of scholarly communication, as it aims to demonstrate how scholarly communication approaches can be applied to knowledge entities and ultimately contribute to knowledge discovery.

Semantic Web for drug discovery

Linked Open Data (LOD) provides linked semantic data for the public. Bio2RDF converts most of the important Biodata into RDF. It opens another research opportunity on how to integrate biomedical data with experimental data and workflows to trace the provenance, semantically mine the bio RDF graphs, and provide provenance aware visualization. Chem2Bio2RDF integrates 25 databases for drug discovery and provides various tools to facilitate search and knowledg discovery.

Social Network Analysis for Research Impact

Social network analysis considers the topology of networks when mining and ranking the network nodes. It offers another important view to measure the impact of scholars in scholarly communications. This research focuses on how to use social network analysis approaches to evaluate research impact, including various centrality measures, network features, PageRank and weighted PageRank, and topic modeling approches. It aims to add temporal features to current PageRank algorithm and identify the potentail convergence of impact measures.

Data integration and mediation in Web2.0

Adding metadata to current Web by tagging so that different blogs, wikis and BBS can be integrated and communicated. The work is based on the integration of different existing popular metadata or social ontologies, such as FOAF (Friend of A Friend), DC (Dublin Core Metadata), SKOS (Simple Knowledge Organization Systems), SIOC (Semantically-Interlinked Online Communities), RSS and so on, to hyperlink different resources for searching and mining.

Semantic Web

Ontology is the backbone of Semantic Web technology. How to generate ontology semi-automatically and create on-site mapping and versioning of various ontologies are critical and interesting research areas.

Ontology Engineering

It focuses on manual-creation of ontology by applying various knowledge acquisition methods (such as interview, self report, laddering, concept sorting, repertory grids and automatic learning techniques), knowledge modeling technologies (modularization, top-level ontologies, spiral knowledge model, etc.) and existing ontology engineering methods (TOVE, Methotology, etc.)

Ontology Generation

Research is focusing on how to use linguistic support to extract part of domain or application ontology semi-automatically. Classes of ontologies can be normalized as key noun phrases. Relation extraction is considered as one of the bottlenecks of the ontology generation. Associated relationship among noun phrases is currently identified as the way to extract relation among these phrases. Future research will be focus on verb, adjective or preposition extraction as relations of ontologies.

Ontology mediation

Ontology meditation is the key part for the whole ontology management structure. Current solutions for ontology mediation still stay at the stage of manually aligning and mapping ontologies with some limited recommendation services. Research is focusing on to identify patterns for ontology mediation. Patterns should be stored in mediation libraries allowing for flexible and easy access and reuse. Ontology mediation library should be set up to manage various mediation patterns. Patterns with some similarity will be clustered together (called patterns cluster) to facilitate the reuse. Personalized view on one mediation pattern can be tailored according to the requirement of specific task or application.