Estimating query communication cost between user and service provider for location based spatial query processing using novel space transformation compared with point transformation

• Main Authors: Voddu, CharanTeja Reddy, Soundari, A. Gnana
• Format: Conference Proceeding
• Language: English
• Subjects: Accuracy; Datasets; Estimation; Location based services; Query processing; Spatial data
• ISSN: 0094-243X; 1551-7616
• DOI: 10.1063/5.0232807

The main goals of this study are to estimate the communication cost of queries between location-based service providers and consumers and to minimise the number of nodes needed to search for geographical data utilising revolutionary space transformation over point transformation. What to Do and What to Bring: A dataset consisting of published student examination results from different locations of the Indian subcontinent is used to train and evaluate the suggested space transformation model. This dataset is derived from the Indian Ministry of Education website and contains more than 10,000 test results with eleven distinct attributes. In order to ensure that the suggested space transformation model is accurate, it is put through its paces in location-based service query processing and cost estimation for spatial data. There are two categories: the new space transformation method and the old point transformation method. We utilise a sample size of N=10 for the accuracy test. We use 80% G-power while running the t-test analysis and deciding on the sample size. According to the findings, the Point Transformation model can reach an accuracy of, while the Space Transformation method can reach an accuracy of up to 88.024 percent (75.786 percent ). With a p-value of less than 0.05, the novel Space Transformation method statistically differentiates from the Point Transformation method. The study obtained a p-value of 0.001, which is considered statistically significant. In geographical data query processing, the results show that the new Space Transformation method is more accurate (88.024 percent), faster, and more significant than the Point Transformation indexing technique (75.786 percent ).