Rule Based Modeling of Knowledge Bases

Abstract

It is critical to have a knowledge base model for efficient storage of extracted knowledge. This ensures that the knowledge is stored in a meaningful way to be used for different applications. The efficiency of the knowledge base model depends largely on the rules of construction. Knowledge represented using logico-linguistic techniques and semantic networks lack a consistent rule based knowledge model. The current paper deals with the analysis of text from the knowledge extraction, representation and semantic network phase to formulate rules which would lay foundations of a knowledge model. The developed rules seem to be promising providing a comprehensive coverage of different scenarios. The extensive coverage is an indication that the knowledge model will cater to the entire domain knowledge, thereby laying the foundations of automatic construction of efficient knowledge bases.

Semantic Graph Knowledge Representation for Al-Quran Verses Based on Word Dependencies

“…Therefore, a knowledge representation model based on rules is a good approach to extract knowledge from the data and store it in a knowledge base. The rule-based knowledge base construction should cover various scenarios in term of sentence structures as well as its meanings [14].…”

“…On the other hand, semantic graph representation has shown great potential because it allows the logical semantic approach to add semantic elements of the Quranic verses. Previous works have made effort to establish semantic graph knowledge representation through relations between semantic units in Al-Quran [10], [14], [15]. However, the work relied on phrase structure of parsed Quranic text which only considering the syntactic structure of the sentences.…”