Welcome to the home page of the Knowledge Information & Data Management Laboratory (KINDML), a research centre at Sirindhorn International Institute of Technology, Thammasat University. KIND is dedicated to the basic and applied researches in the areas of Natural Language Processing, Data Mining, Knowledge Management, and Machine Learning.
Knowledge Information & Data Management Laboratory founded in 1999 at SIIT Rangsit Campus.
Natural Language Processing
- Robust NLP and Linguistic Knowledge AcquisitionWhile NLP systems are gradually becoming accepted by a wider range of people both in academic and business area, many difficult problems are still unsolved. One of the important problems is how to improve robustness and adaptiveness in NLP system, especially how to analyze and interpret various phrases and sentences which are ungrammatical (also called ill-formed inputs). A user-friendly system should be robust and flexible in that it can analyze any well-formed and ill-formed input efficiently. The system should also be adaptive to deal with phrases/sentences including unseen construction and vocabulary, for instance learning some new grammar rules. Currently, we are focusing on both rule-based and corpus-based approaches to cope with ill-formed inputs and, when needed, to acquire novel linguistic knowledge. On the increase of very large electronic corpora, statistics obtained from such corpora are a useful clue for this problem.
- Text Interpretation: Information Retrieval, Categorization and Information ExtractionIn the past, most online information stored in databases or spreadsheets. At the present time, the majority of online information is text-based, e.g., e-mail, news, journal articles, reports, books, encyclopedias. These information sources are worth but there is too much information available, and not enough time to sort through it. Text interpretation techniques are helpful for categorizing, filtering and extracting information from text. Three types of text interpretation are information retrieval, categorization, and information extraction. We are interested in developing of efficient methods to various tasks of text interpretation.
Knowledge Science and Engineering
- Knowledge Data Discovery in DatabaseKnowledge Data Discovery (KDD) is a rapidly growing interdisciplinary field that merges together databases, statistics, machine learning and other AI technologies in order to extract useful knowledge from a large-scaled collection of data. The problems in this field are of two general categories: (1) prediction and (2) knowledge discovery. Knowledge discovery is a stage prior to prediction, where information is insufficient for prediction, such as clustering, association rules, text mining and so on. Our study aims at finding and implementing efficient, robust and scalable methods in real-world situation where databases are complex, voluminous, noisy and non-stationary. Some interesting applications include computer-aided education (CAI), decision support systems, and management information systems.
- Intelligent Decision Support SystemsIn business, government, and other organizations, decision making plays an important part in determining the landscape of tomorrow’s world. Computer systems that assist decision-making process are called decision support systems (DSSs). Intelligent decision support systems (IDSSs) are DSSs that make use of techniques emerging from the field of artificial intelligence (AI). Our research focuses on studying new techniques in both (1) model-driven support systems, which are based on strong theory or model, and (2) data-driven support systems, which are based on database technologies and statistical methods