Research Interests
Throughout the
journey of my career, my research interest has mainly focused on all aspects of
Machine Learning and Data Mining and,
recently, with the emphasis on their applications in areas of bio-medical informatics and e-healthcare
under IoT environment. I have been currently leading several research
projects in such area funded by the Korea Research Foundation, the Korea Science
and Engineering Foundation, the ETRI, the KCDC, the Ministry of Food and Drug
Safety and so on. Also, another focus of my interest is to develop
computational approaches for the mining of popular opinions about commercial
products. such as smart-phones, spread over the
interest, blogs, and etc. For this purpose, ther
issues of applying machine learning techniques for text-mining are being
investigated in various aspects.
1.
Bio-Medical Informatics
2. Data
Mining and Machine Learning
3.
Text-Mining / Opinion-Mining
1. Bio-Medical Informatics
The field of
bio-medical informatics concerns about the great use of information technology
to handle a large volume of bio-medical data, such as DNP chip data, SNP
genotyping data, clinical data, and etc., in an
efficient way. So, it deals with the resources, devices, and methods required
to optimize the acquisition, storage, retrieval, and use of information in
health and biomedicine. This research can contribute to the advance in
personalized medicine, new drug development, desease
diagnosis and prognosis.
- Cancer
diagnosis and prognosis system using DNP chips/SNP chips
- Gene
network modeling using gene expression profiles
-
Computational technology for effective healthcare
- Clinical
decision support systems
-
Bio-medical image processing
2. Data Mining and Machine
Learning
The task of data
mining is to extract significant patterns or knowledge hidden under a large
volume of data. It is currently used in a wide range of fields, such as
marketing, fraud detection, and scientific discovery. As a variety of data have
grown in size and complexity, data mining tools that transform data into
business intelligence or other valuable knowledge became of great demand. The
related topics include classification, clustering, neural networks, decision
tree, support vector machines, and etc.
- Predictive
modeling in a variety of applications
- Deep
learning
-
Interpretable machine learning
3. Text-Mining / Opinion-Mining
Text mining, also
frequently referred to as text data mining, refers to the process of extracting
high-quality information from text. High-quality information is typically
derived through the quessing of patterns and trends
by using machine learning or other computational techniques along with natural
language processing. Typical text mining tasks include text categorization,
text clustering, sentiment analysis, document summarization, relation
extraction, and etc.
- Opinion
mining (or sentiment analysis) of mobile phones from blogs or internet
- Gene-gene
relation extraction from bio-literature
-
Text-mining based gene raking methods for biomarker discovery
-
Text-mining techniques for healthcare provider quality determination