Authors: Ashish Kumar; Roheet Bhatnagar; Sumit Srivastava
Addresses: Department of Computer Science and Engineering, Manipal University Jaipur, Rajasthan, India ' Department of Computer Science and Engineering, Manipal University Jaipur, Rajasthan, India ' Department of Information Technology, Manipal University Jaipur, Rajasthan, India
Abstract: The amount of data available and information flow over the past few decades has grown manifold and will only increase exponentially. The ability to harvest and manipulate information from this data has become a crucial activity for effective and faster development. Multiple algorithms and approaches have been developed in order to harvest information from this data. These algorithms have different approaches and therefore result in varied outputs in terms of performance and interpretation. Due to their functionality, different algorithms perform differently on different datasets. In order to compare the effectiveness of these algorithms, they are run on different datasets under a given set of fixed restrictions, e.g., hardware platform, etc. This paper is an in depth analysis of different algorithms based on trivial classifier algorithm, kNN and the newly developed ARSkNN. The algorithms were executed on three different datasets and analysis was done by evaluating their performance taking into consideration the accuracy percentage and execution time as performance measures.
Keywords: classification; nearest neighbours; ARSkNN; similarity measure.
International Journal of Computer Aided Engineering and Technology, 2021 Vol.15 No.1, pp.1 - 11
Received: 20 Feb 2018
Accepted: 24 Oct 2018
Published online: 12 May 2021 *