2004

Proceedings of the 1st Annual Sessions, Sri Lanka Association for Artificial Intelligence

16 September 2003- Colombo

Building your Medical Expert System- Do it yourself

D.A.I.P. Fernando, A.S. Karunananda

AbstractLack of appropriate expert system development tools, which are user friendly and which required less technical knowledge, have been a barrier for medical doctors themselves to develop expert systems for their own use. We have developed an expert system shell, which has a more user-friendly graphical user interface, simple inference engine and a rule based knowledge base, which represents the knowledge in a relational database. This tool was implemented using Microsoft Visual Basic and Microsoft Access. We have been able to demonstrate its usefulness by developing real world applications.

PDF(76KB)

 

CD-Builder: A Natural Language-based CASE Tool

Pushpakumara DKH, Sriskanthaverl K, Uthaiyashankar S, Wijethunge KPHC

AbstractThis paper describes CD-Builder, a natural language-based CASE tool, which aims at supporting the Analysis stage of software development in an Object-Oriented framework. The CD-Builder uses Information Extraction technique to analyze the problem statement. This paper discusses the design of CD-Builder, IE
technique and the future enhancements.

PDF(56KB)

 

Using a Hybrid Artificial Intelligence for decision making in medicine

D.A.I.P. Fernando, A.S. Karunananda

Abstract-A medical consultation is a process, which utilizes clinician’s intelligence and expertise. There are intelligent systems, which provide assistance to the clinician. But, most of these systems have been developed using individual Artificial Intelligence (AI) techniques, which address only few tasks of this entire process. This paper describes how a hybrid system can provide a better solution, by combining different AI techniques. The hybrid AI system includes a patient record system, which is used to maintain patient information, integrated with the following modules. Expert system module is used to make diagnoses, genetic algorithm module is used to optimize the process of selecting drug treatment, and artificial neural network module is used to predict the response to the prescribed drug treatments. We have implemented this system, and have already demonstrated the usefulness of some of these modules.

PDF(72KB)

Modeling of Tacit Knowledge

D.S. Kalana Mendis, Asoka S. Karunananda, U. Samarathunga, 

Abstract-Tacit knowledge has always been influential in changing the directions and emphasis of explicit
models of knowledge. All explicit knowledge is rooted in tacit knowledge. Due to these reasons modelling of tacit knowledge is of great interest. A research has been conducted to develop an approach to model tacit knowledge. In this research, we have used Artificial Intelligence technique of fuzzy logic for developing an approach to model tacit knowledge. We have considered domain of “Ayurvedic” medicine as a case study domain with tacit knowledge. Tacit knowledge in Ayurvedic sub-domain of individual classification has been acquired through a questionnaire and analysed to identify the dependencies, which lead to make tacit knowledge in the particular domain. In the first place analysis was done using statistical techniques of principle components and the
results were not compatible with the experiences of Ayurvedic experts. As such, fuzzy logic has been used to further model the Ayurvedic subdomain. The result of the modelling of Ayurvedic domain using fuzzy logic has been compatible with the experiences of the Ayurveic experts. A framework for modelling tacit knowledge has been integrated with an expert system shell thereby enabling the development of expert systems for domains with tacit knowledge. Framework has been successfully applied for several tacit domains.

PDF(103KB)

 

Neural Lab: Workbench for Artificial Neural Networks Experiments and Training

D.A.I.P. Fernando, A.S. Karunananda

Abstract-Most students find the area of artificial neural networks difficult to understand, probably due to
inadequate comprehensible teaching materials. Also most of the existing neural network shell softwares are not very user-friendly for novel users. We have developed an software, which serve three main purposes. Firstly it provides demonstrations of theoretical concepts of artificial neural networks and sample applications with, which users can experiment. Secondly it generates neural network topologies according to user requirements and they are demonstrated graphically and can be trained. Thirdly it shows how newer concepts in neurosciences can be used to enhance the models artificial neural networks

 

PDF(251KB)

An Intelligent Algorithm for Utilizing a Low Cost Camera as an Inexpensive Barcode Reader

R Janapriya, L Kularatne, K Pannipitiya, A Gamakumara, C de Silva and N Wickramarachchi

 AbstractWe have developed a low cost, low resolution camera based al barcode reader, which can extract and decode the barcode sequence on a cluttered background. It is composed of three functions: barcode localization from the raw image, transformation of the localized barcode and decoding the sequence with an intelligent algorithm. The localization method is based on detecting the areas with the maximum density difference in two normal directions. The transformation method, capable of identifying any orientation, is based on the Hough line detection method. The decoding method is based on the peak/valley detection method of the barcode waveform and a consistency checking method. The consistency checking method, a constraint network, employs an artificial intelligence searching method.
PDF(291KB)

 

Using Agent Technology for Knowledge Management

Marino Rajasingham, Asoka S Karunananda

AbstractOrganisations acquire knowledge during their lifetime, through the “know how” and experiences of employees
with regard to actions carried out in the organisation. Most of the time this knowledge is confined to individuals
and no access to majority of the users in the organisation. This leads to re-inventing wheel without using evolving
knowledge in the organisation for meeting the competitive advantages. Therefore, there is a necessity for developing a Knowledge Management Agent, which can deceptively learn from organisational experiences and make such knowledge available to employees. Agent technology has been used to develop knowledge management agent that uses genetic algorithms to support evolving of knowledge. The Knowledge Management Agent is capable of categorically learning from both organisational experiences and external world through the Internet. The Agent has been implemented to emulate processes of proactive and reactive learning. More importantly, the Agent supports the evolution of optimal solutions rather than reproducing from the pre-stored knowledge. This knowledge management agent has been tested and the result appears to be very encouraging

PDF(136KB)