RESEARCH INTERESTS [ Research Statement ]:

1. Data Mining, Machine Learning, and Data Quality

2. Information retrieval, Video databases and multimedia systems

Recent advances in high-performance networking and computer hardware have led to the emergence and proliferation of knowledge extraction from data with considerably large volume. This problem is complicated by the reality that data are physically distributed, contaminated by various imperfections, such as errors, outliers and anomalies. Meanwhile, data modality may vary significantly from market transaction data, sensor network data to video, audio, images and other multimedia data. Knowledge discovery and data mining from those data is an exciting scientific discipline since it requires us to integrate and advance the knowledge produced in multiple areas, including database systems, statistics, machine learning, and multimedia databases. My current research has been focused on the following three themes: (1) knowledge discovery across multiple information sources; (2) knowledge extraction in the presence of data imperfections; and (3) knowledge extraction from multimedia data. :

1. Knowledge discovery across multiple information sources

1. Discovering Relational Patterns across Multiple Databases

2. Knowledge Extraction in the Presence of Data Imperfections

1. Data quality and integrity assessment and noise impact analysis

2. Identifying mislabeled training examples for effective learning (including cost-sensitive noise handling)

3. Identifying and localization errors (noise) which are introduced to the attributes

4. Cost-constrained data acquisition for data quality enhancement

5. Multi-Classifier Systems for effective mining from noisy data streams

3. Knowledge Extraction from Multimedia Data

1. Video content structure mining for efficient database management and access

2. Video data mining from the association perspective

4. Content-based Video/Image Analysis, Summarization, Access and Database Management

1. Content-based Information Retrieval

2. Video Analysis and Processing

3. Medical images (video) analysis and processing

Please contact me, if you have any concerns regarding the system demonstrations, publications, and detailed experimental results.