Current Research in Artificial Intelligence

Brief note
talk to graduate students of the department of computer science, Faculty of Science, Kasetsart University, 28 August 2010.

Traditionally, AI has been a collection of various "heuristic" ways to solve problems by computer. Most problems are the problems that require human intelligence to solve:

1. Game playing
2. Natural language
3. Planning
4. Knowledge representation
5. Robotics: image understanding etc.

AI also brings advanced in computer science due to its nature of being forefront in using computing techniques to solve problems. Such contributions are:

Multi-user systems
Interactive computer systems
Advanced computer languages
Modern User interface
Internet

Presently, AI research has been "diversified" into many subfields:

Nutural computing: Neural-networks, Evolutionary computation, Fuzzy...
Logic: verification of hardware and software
Machine Learning: computation learning, data mining ...
Agent: autonomous computing, large scale interaction ...

Basis of AI is built on Logic and Heuristic. Presently, more effort has been put into "scaling-up" the size of problem that AI can be used.

Examples of possible Thai AI research:

Thai text summarisation
Gathering and summarising news from various sources automatically.
Thai speech recognition
Perhaps through mobile phone, there are tremendous opportunity for such services

Prabhas Chongstitvatana