The main of this paper is to analyse the performance of customer behaviour in online shopping by using precision value.A novel Self-organizing map with Random Forest was iterated at different times for predicting accuracy percentage of customer behaviour. The result proved that the Novel Self-organizing map got significant results with 96% accuracy compared to Random Forest with 93% accuracy. Self-organizing mapis a simple and most effective algorithm to build fast machine learning models. The self-organizing map helps in predicting with more accuracy the percentage of customer behaviour.

The main aim of this paper is to analyse the performance of customer behaviour in online shopping by using precision value.A novel Self-organizing map with sample size 10 and k nearest neighbour with sample size 10 was iterated at different times for predicting accuracy percentage of customer behaviour. The minimum power of analysis is fixed as 0.8 and the maximum accepted error is fixed as 0.5. The result proved that the Self-organizing map got significant results with 96% accuracy compared to K Nearest Neighbour with 93% accuracy. The self-organizing map appears to perform significantly better than K-Nearest Neighbour with the value of p=1.000. Self-organizing mapis a simple and most effective algorithm to build fast machine learning models. The self-organizing map helps in predicting with more accuracy the percentage of customer behaviour.