Tutorial

SNLP 2007 Tutorial Programs (Tentative)

Objective

Tutorials in SNLP 2007 will specially focus on educating Thai students and researchers who just start, have been involved, or would like to extend their knowledge in the area of Natural language processing (NLP). Therefore, tutorials have been carefully designed to cover some popular sub-areas. In order to accelerate and encourage research, introductions as well as practical uses of NLP tools will be given in each particular class. All tutorials by Thai instructors will be lectured in the Thai language.

Topics

1. HMM and Its Applications

Hidden Markov Model (HMM) is currently a popular tool that have been applied in a wide range of pattern recognition problems such as speech recognition, handwriting recognition, gesture recognition, singing scoring and bioinformatics. The HMM is a statistical model, in which the system being modeled is assumed to be a Markov process with unknown parameters, and the challenge is to determine the hidden parameters from the observable parameters. The extracted model parameters can then be used to perform further analysis. This tutorial aims to introduce the HMM algorithm, practice using the well-known Hidden Markov Toolkit (HTK), and demonstrate utilizations of the HMM. Examples are given in areas of speech recognition, speech/non-speech detection, speaker segmentation, noise classification, and voice-event detection.

2. Wordnet and Its Applications

WordNet, an electronic lexical database, is considered to be the most important resource available to researchers in computational linguistics,text analysis, and many related areas. Its design is inspired by current psycholinguistic and computational theories of human lexical memory. English nouns, verbs, adjectives, and adverbs are organized into synonym sets, each representing one underlying lexicalized concept. Different relations link the synonym sets. In this tutorial, the design of WordNet and the theoretical motivations behind from linguistics viewpoint will be illustrated. Some Wordnet research and related application will be explained.

3. Named-Entity Recognition

Named entity recognition (NER) is one of the most fundamental tasks in information extraction (IE) and NLP. The main function of NER is to locate and classify elements in text into predefined categories such as the names of persons, organizations, locations, expressions of times, quantities and monetary values. Research on NER for English and other well-known Asian languages such as Chinese and Japanese have been extensively performed and various experimental results using different approaches were published in numerous conference proceedings and journals. However, NER for Thai language is still very limited. One of the reasons is due to the lack of background knowledge in this task. Therefore, this tutorial aims to give some background theory in NER including the discussion of several machine learning techniques including Support Vector Machines (SVMs) and Conditional Random Field (CRF). The tutorial will also survey some of the software tools which could be applied for NER task.

4. Workbench Ontology Acquisition and Its Applications

Knowledge management systems help to share the knowledge among different communities using shared agreed terminologies with ontological knowledge in multilingual aspects. The system, compared to a traditional knowledge management system can make use of knowledge contained in terminology system such as thesauri to construct semantic, multilingual and domain-specific ontology based knowledge server. The AGROVOC Concept Server Workbench is such a concept-based system, which allows the representation of more semantics such as specific relationships between concepts and relationships between their multilingual lexicalization. This tutorial aims to give an overview of the workbench for collaborative ontological knowledge construction and maintenance and its applicability in the field of food and agriculture