Hair-oriented data model for spatio-temporal data representation. (15th October 2016)

• Record Type: Journal Article
• Authors: Madraky, Abbas; Othman, Zulaiha Ali; Hamdan, Abdul Razak
• Is Part Of: Expert systems with applications. Volume 59(2016)
• Page Start: 119
• Page End: 144
• Publication Date: 2016-10-15
• Subjects: Spatio-temporal data models -- File size reduction -- Query execution time -- Hair-oriented data model -- Nested tables; Expert systems (Computer science) -- Periodicals; Systèmes experts (Informatique) -- Périodiques; Electronic journals; 006.33
• DOI: 10.1016/j.eswa.2016.04.028
• Languages: English
• ISSNs: 0957-4174

Abstract

Highlights: We proposed a bio-inspired data model for spatio-temporal systems. We define formal concepts for data structure and functions in the proposed model. We compare the file size reduction and query execution time to other data structures. We implement the proposed data model and import dataset using the Oracle database management software. Abstract: Having an effective data structure regards to fast data changing is one of the most important demands in spatio-temporal data. Spatio-temporal data have special relationships in regard to spatial and temporal values. Both types of data are complex in terms of their numerous attributes and the changes exhibited over time. A data model that is able to increase the performance of data storage and inquiry responses from a spatio-temporal system is demanded. The structure of the relationships between spatio-temporal data mimics the biological structure of the hair, which has a 'Root' (spatial values) and a 'Shaft' (temporal values) and undergoes growth. This paper aims to show the mathematical formulation of a Hair-Oriented Data Model (HODM) for spatio-temporal data and to demonstrate the model's performance by measuring storage size and query response time. The experiment was conducted by using more than 178, 000 records of climate change spatio-temporal data that were implemented in implemented in an object-relational database using nested tables. The data structure and operations are implemented by SQL statements that are related to the concepts of Object-Relational databases. The performances of file storage and execution query are compared using a tabular and normalized entity relationship model that engages various types of queries. The results show that HODM has a lower storage size and a faster query response time for all studied types of spatio-temporal queries. The significances of the work are elaborated by doing comparison with the generic data models. The experimental results showed that the proposed data model is easier to develop and more efficient.