Mining Implicit and Explicit Rules for Customer Data Using Natural Language Processing and Apriori Algorithm

Abstract

The Internet is a massive storehouse of organized and amorphous information. It is not an easy task to examine all these information to draw secret public views and emotions. It's extremely complicated to identify connections with large databases. There are several obsolete and incomplete information files in these archives that are not necessary to retrieve the laws. Therefore, the reliability of the organization rules is significantly affected by these irrelevant data and there is a need to pre-process these documents. In a rapidly growing technology world, there are many customers express their views through online, so organizations are highly dependent on the user’s opinion. Natural Language Processing is the branch of machine learning which is about analyzing any text and used to handle predictive analysis. It can assists and interprets to profound the sentence structure in its meaning. A transaction is just a set of items that a customer purchases in a basket. To examine the connection between the items sold in a supermarket, Apriori algorithm is used to identify frequent sets of items that are explicitly bought together. Implicit relationship is ignored in Apriori algorithm which can be identified using sentiment analysis. Sentiment Analysis helps to identify the object and topic from the text to which the feeling is guided. This proposed analysis may assists to identify the implicit product in order to improve the sales by providing offers for respective implicit products.