Computational Estimate Visualisation and Evaluation of Agent Classified Rules Learning System

Abstract

Abstract!Student modelling and agent classified rules learning as applied in the development of the intelligent Preassessment System has been presented in [10], [11]. In this paper, we now demystify the theory behind the development of the pre-assessment system followed by some computational experimentation and graph visualisation of the agent classified rules learning algorithm in the estimation and prediction of classified rules. In addition, we present some preliminary results of the pre-assessment system evaluation. From the results, it is gathered that the system has performed according to its design specification.

A System of Serial Computation for Classified Rules Prediction in Non-Regular Ontology Trees

“…In practical systems this implies: on or off, good or bad, working or notWorking, found or notFound, present or notPresent, etc. The investigation of relational-concept ontology in [16], [17] for the pre-assessment of students from node-to-node in a regular ontology: ontology with equal number of leaf-nodes across all parent nodes (Fig.1) used the boolean parameter T = 2 torepresent either a pass or a fail for measurement of pre-assessment outcomes. The parameters T and leaf-nodes N of a given parent class node then formed the basis of agent classification learning.…”

“…To estimate and predict the number of classified induction learning rules R, [16], [17] as stated earlier projected the Initialisation algorithm R = CT N + 1 eq.1…”

“…Based on the message received from the sending agent to a receiving agent (i.e. the classifier agent, see [16], [17], the classifier agent can then classify the environment (students) into one of the selective category according to number of nodes the students are pre-assessed on. Figure 3 is a real model of a non-regular ontology in the domain of SQL learning structure in a teaching-learning environment.…”

“…Objects or network structures that are regular take uniform dimensions. But not all processes or models are regular.While the work of [16], [17] projected the Initialisation algorithm for regular ontology number of rules prediction, this paper projects a polynomial for the non-regular ontology counterpart. The polynomial equation predicts the number of classified induction rules in a multiagent system(MAS) for the classification of students for appropriate learning material(s) after pre-assessment about their SQL learning.…”

“…Now, within a given belief network or ontology structure, how can the total amount of classified induction rules for both positive (true) and negative (false) training examples be estimated and predicted? In [16], [17] directed edges represented object properties [19] of some semantic predicate and knowledge representation between nodes in the ontology. The edges as directed path for student learning through to the leaf-node terminals where students' learning are pre-assessed and classified based on two possible state (boolean) parameters.…”