This algorithm is used to identify the pattern of data. It’s basically based on observation of data pattern around a transaction.
If a person goes to a gift shop and purchase a Birthday Card and a gift, it’s likely that he might purchase a Cake, Candles or Candy. So these combinations help predict the possible combination of purchase to the retail shop owner to club or package it as offers to make better margins. This also enables to understand consumer behavior.
When we look at apriori algorithm its essential to understand what is Association rules too. That will help to understand in the right perspective.
Association rule learning is a popular machine learning technique in data mining. It helps to understand relationship between variables in large databases. It’s being primarily implemented in Point of Sale in retail where large transactions are recorded.
Reference links for Begineers:
I like this http://nikhilvithlani.blogspot.in/2012/03/apriori-algorithm-for-data-mining-made.html url very simple and easy to understand for novice or beginners.
Reference links for Researchers and algorithm lovers:
My objective of this post is a pre-cursor to use R and Big Data to use Market Basket analysis to do recommendation in retail point of sale domain or based on billions of e-Commerce transactions. In the upcoming posts we will see how we leverage this algorithm and do appropriate analysis on a point of sale data. Keep watching this space.