Expert system for crop disease based on graph pattern matching: a proposal

Sambony, Emmanuel Gerardo Lasso; Corrales, Juan Carlos

doi:10.22395/rium.v15n29a5

Cited by 4 publications

(1 citation statement)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The problem topicality is substantiated by the reviews [1] describing the current research and developments in the fi eld of expert systems. Nowadays, the expert systems are used by business [2], medicine [3], agriculture [4], education [5], defense industry [6] and by other engineering applications and by many other spheres.…”

mentioning

confidence: 99%

Non-relational approach to developing knowledge bases of expert system prototype

Hnatushenko¹,

Hnatushenko²,

Dorosh³

et al. 2022

Nauk. visn. nat. hirn. univ.

View full text Add to dashboard Cite

Purpose. Use of a non-relational database management system is proposed while developing a database of a prototype of expert system with using a semantic model of the knowledge. Methodology. The study compares traditional relational approach with the proposed non-relational one in terms of the formation of certain queries. The following indices are used to compare efficiency of two management systems for the databases: particular query set (in MySQL and Cypher languages); runtime for the specified record size (i.e. their processing speed); ease of understanding: and software support of the queries. Findings. It has been identified that the graph model is a more expedient solution in the process of designing semantic networks and their development where complex hierarchical relationships between objects have to be stored and processed. Architecture of the graph database has been applied in terms of the specific example. A prototype of an expert system has been developed to demonstrate the capabilities of the created system of logical inference. The classifier of sciences was chosen as an example in the subject area. Originality. A prototype of the expert system, using the proposed non-relational approach, has been designed involving modern service-oriented architecture (SOA). The abovementioned helped separate the database from the inference engine and the user interface, facilitate perception as well as update and code debugging. Service-oriented architecture makes the system more flexible and robust. Practical value. The developed software is meant to develop both simple expert systems and medium-complex ones.

show abstract

mentioning

confidence: 99%