Know-Ont Based Ontology Modeling Approach for Skill Knowledge Extraction

Abstract

Ontology is an important tool for organizing information into categorized data for a semantic web search engine in order to create an ontology which can support both the collection and the presentation of skill knowledge. This study used a Know-Ont based ontology modeling approach (KOOM) on a case study of a basketball shooting technique to create semi-automatic ontology engineering, which is expected to help facilitate better understanding and interpretation of the multidimensional table in a knowledge engineering process and reduce the need for expert support. The efficiency evaluation of skill knowledge extraction framework based on KOOM shows the following values: Precision =0.72, Recall =0.71, Accuracy = 0.94 and F-Measure =0 .17 , which are considered to be efficient.Index Terms-Ontology, ontology modeling, semantic web, knowledge engineering.

Technique of Semantic Unambiguity for a Concept Selection of Terms in Focused Contexts with Reinforcement Learning Integration

“…Therefore, a context is a crucial part to minimize the different meanings of each word for more accurate interpretation; for example, "Guard" in a context of a profession and "Guard" in a context of a player position in sports. One of the things that can be used to set the context of the words is the vocabulary set used in one sentence [2] or words that can be seen in the same document, table or context [5]. The related works of how to interpret the vocabulary by its context are as follows:…”

“…3) Synset Detection [5] is the selection process assigning the synset suitable with the focused context to each term. The Reinforcement Learning technique [10] is used to select synset from the vocabulary list on the same multi-dimension data table, or from under the same column.…”

“…After a while of conducting this research, the researcher came across a problem regarding the categorization of terms for ontology structuring [5] because the concept used by the system misrepresented the meaning of terms in the focused context for determining the semantic similarities scores; therefore, generating many incorrect results. However, from the study, it was found that the technique for concept selection suitable for each term has led to more accuracy in categorization.…”

“…This research further studied the synset detection process by [5] to increase the efficiency of synset detection process. The problem was encountered when the evaluated terms had only few related concepts.…”

“…Therefore, the final result was unsuitable for the focused context. Problem solving process development [5] was improved to create the Technique of Semantic Unambiguity for a Concept Selection of Terms in Focused Contexts to truly identify the meaning of each word under the same context and categorize these words to create semi-automatic ontology [5] by using Precision, Recall and F-Measure to compare the efficiency with the synset detection process. This paper was divided into section as follows: Section 2 presented knowledge involving with this research, demonstrating the importance, the concepts and the guideline for the development and the improvement of synset detection process to be more effective.…”