Projects: Semantics and Digital Computing Research Group

This project is ongoing project. We interest on how to interpret Thai literature using semantic approach. Natural language processing, computer linguistic and ontology are focused on creating of the tool. The support tool interprets the meaning of the content of Thai literature and that supports the user on learning Thai poem. Interpretation captures the meaning of some knowledge regarding activity, event, time, and emotion of the Thai poem.

Selected publications:
status:
Active
student:
Piyawat Thongkeaw (Ph.D. Thesis)

GIS information is important for e-Health surveillance because such information is required for monitoring, exchanging and shared across health GIS systems. However, interacting with the health GIS system is difficult for the users who may have less knowledge about GIS systems. This paper proposes a semantic query builder that provides a query service and can interact with health GIS systems. The semantic query builder receives the semantic query and translates such queries into a SQL query to allow querying information from databases stored in the health GIS system. It delivers GIS tuple data - the data from the map service is provided by the health GIS system. With the proposed semantic query builder, the users who may have less knowledge of SQL and GIS system are able to specify the query flexibly which meets their requirement.

Selected publications:

An applied ontology: A semantic query builder for health GIS system Theeradol Boonprapasri and Gridaphat Sriharee, International Computer Science and Engineering Conference (ICSEC), Chiangmai, Thailand, 23-26 Nov. 2015, pp.1-6

status:
Finished
student:
Theeradol Boonprapasri (Master Project)

Expertise finding is the process to assess the individual expertise. The expertise may describe some aspects of the skill or knowledge of the researcher. In this paper, multi-aspect expertise is introduced and it is analyzed by research documents’ title. The title is matched by the 2012 ACM computing classification system and the multi-aspect expertise is assigned to the document. The multi-aspect expertise covers broad skill and specific skill. We also present the multiple aspect ranking – a ranking method to rank the researchers who may have expertise with conditions specified in the query using the proposed multi-aspect expertise. The experiment is conducted and the result shows that multi-aspect expertise is practical enough for expertise finding as well as for the ranking process.

Selected publications:

Previous related paper:
A Researcher Expertise Search System using Ontology-Based Data Mining, Ravikarn Punnarut and Gridaphat Sriharee, Proc. 7th Asia-Pacific Conference on Conceptual Modelling (APCCM 2010), Brisbane, Australia

status:
Active
student:

Tagging plays a crucial role in the success of social network and social collaboration. This paper proposes an auto-tagging methodology for articles using Latent Semantic Indexing (LSI) and ontology. The proposed methodology consists of pre-processing and tagging process. In pre-processing process, the LSI vector is created for article classification. The tagging process suggests some ontological tags. An accuracy evaluation of auto-tagging compared with manual-tagging is discussed. The experimental results show that the proposed auto-tagging methodology returns high accuracy and recall.

Selected publications:

Auto-Tagging Articles Using Latent Semantic Indexing and Ontology , Rittipol Rattanapanich, Gridaphat Sriharee, Intelligent Information and Database Systems (ACIIDS 2014), Volume 8397, Lecture Notes in Computer Science, pp 153-162

status:
Finished
student:
Rittipol Rattanapanich (Master Project)