Rukaini Abdullah

An ontology-based reasoning framework for reaction mechanisms simulation

Many chemistry students have difficulty in understanding an organic chemistry subject called reaction mechanisms. Mastering the subject would require the application of chemical intuition and chemical commonsense adequately. This work discusses a novel framework using Qualitative Reasoning (QR) to provide means for learning reaction mechanisms through simulation. The framework consists of a number of functional components. These include substrate recognizer, qualitative model constructor, prediction engine, molecule update routine, explanation generator, and a knowledge base containing essential chemical facts and chemical theories. Chemical processes are represented as qualitative models using Qualitative Process Theory (QPT) ontology. The construction of these models is automated based on a set of QR algorithms. We have tested the framework on the SN1 and the SN2 reaction mechanisms. Representative cases of reaction simulation and causal explanation are also included to demonstrate how these models can serve as a cognitive tool fostering the acquisition of conceptual understanding via qualitative simulation.

Data Mining Approach: Relevance Vector Machine for the Classification of Learning Style Based on Learning Objects

Recent researches indicate that a lot of effort has been done to provide learners with personalized learning objects. Previous studies classified learning object based on the description of the learning style preference itself without considering student preference. In this study, we propose a data mining approach to the classification of learning objects based on learning style while considering student preference use of the learning objects. Relevance Vector Machine (RVM) is used to build a classifier for the classification of learners. For the purpose of comparison, Support Vector Machine (SVM) and Neural Network (NN) were applied. Comparative simulation results indicated that the propose RVM classifier accuracy and computational time complexity is superior to the NN, and SVM classifiers. The classifier proposes in this research can be of help to educators in proposing appropriate learning objects with high level of accuracy within a short period of time. This in turn can significantly improve learners performance in understanding the subject matter.

A Summary Sentence Decomposition Algorithm for Summarizing Strategies Identification

Expert summarizers employ a number of strategies to produce summaries. Teachers need to identify which strategies are used by students to help them improve their summary writing. However, the task is time consuming. This paper reports on our effort to develop an algorithm to identify the summarizing strategies employed by students using summary sentence decomposition. The summarizing strategies used by experts are identified and translated into a set of heuristic rules. A summary sentence decomposition algorithm is then developed based on the heuristic rules. A preliminary test was carried out and the results are discussed.

QRIOM: A QPT-based simulator for composing and reasoning qualitative models for learning organic reactions

The work discusses the application of an artificial intelligence technique called qualitative reasoning (QR) and a process-based ontology in constructing qualitative models for organic reaction simulation. We present a framework architecture that uses the QPT ontology as the knowledge representation scheme to model the behaviors of a number of organic reactions. The main focus of this paper placed on the design of two main components (model constructor and reasoning engine) for a tool abbreviated as QRIOM for predicting and explaining organic reactions. The discussion starts by presenting the workflow of the reasoning process and the automated model construction logic. We then move on to demonstrate how the constructed models can be used to reproduce the behavior of organic reactions. Finally, behavioral explanation manifestation is discussed. The simulator is implemented in bi lingual; Prolog is at the backend supplying data and chemical theories while Java handles all front-end GUI and molecular pattern updating.