Support Vector Machine–Based Prediction System for a FootballMatch Result

Journal Title: IOSR Journals (IOSR Journal of Computer Engineering) - Year 2015, Vol 17, Issue 3

Abstract

 Abstract: Different techniques have been used to develop result prediction systems. In particular, footballmatch result prediction systems have been developed with techniques such as artificial neural networks, naïveBayesian system, k-nearest neighbor algorithms (k-nn), and others. The choice of any technique depends on theapplication domain as well as the feature sets. The priority of a system developer or designer in most cases is toobtain a high prediction accuracy. The objective of this study is to investigate the performance of a SupportVector Machine (SVM) with respect to the prediction of football matches. Gaussian combination kernel type isused to generate 79 support vectors at 100000 iterations. 16 example football match results (data sets) weretrained to predict 15 matches. The findings showed 53.3% prediction accuracy, which is relatively low. Untilproven otherwise by other studies, an SVM-based system (as devised here) is not good enough in thisapplication domain.

Authors and Affiliations

Chinwe Peace Igiri

Keywords

Related Articles

 An in-building multi-server cloud system based on shortestPath algorithm depending on the distance and measuredSignal strength

 Abstract: In this paper we are proposing a system that will work as a virtual cloud computing environment.Our system will be suitable for an office for example software development firm. Cloud Computing is anemergi...

 Quality Technical, Vocational Education and Training: A Toolfor Self Reliance

 Abstract: This paper has discusses the relevance of a quality TVE training as a tool for self-reliance. Itidentified skill and knowledge as the engine room for economic growth and sustainable livelihood only if eff...

 Roulette Wheel Selection Model based on Virtual Machine Weight for Load Balancing in Cloud Computing

 Abstract: The increasing use of cloud computing resources have placed a large workload on the data centers. Solving the load balancing problem among all virtual machines (VMs) increases performance, throughput and...

Security Issues and Attacks in Wireless Sensor Network

Remote sensor arranges is a standout amongst the most developing innovation for detecting and playing out the distinctive undertakings. Such systems are valuable in numerous fields, for example, crises, wellbeing observi...

 Comparative Analysis of Different Types of Full Adder Circuits

 The Full Adder circuit is an important component in application such as Digital Signal Processing (DSP) architecture, microprocessor, and microcontroller and data processing units. This paper discusses the ev...

Download PDF file
  • EP ID EP111323
  • DOI -
  • Views 97
  • Downloads 0

How To Cite

Chinwe Peace Igiri (2015).  Support Vector Machine–Based Prediction System for a FootballMatch Result. IOSR Journals (IOSR Journal of Computer Engineering), 17(3), 21-26. https://europub.co.uk/articles/-A-111323