06-04-2011, 04:15 PM
Predicting Missing Items in Shopping Carts
Abstract:
Existing research in association mining has focused mainly on how to expedite the search for frequently co-occurring groups of items in “shopping cart” type of transactions; less attention has been paid to methods that exploit these “frequent itemsets” for prediction purposes. This paper contributes to the latter task by proposing a technique that uses partial information about the contents of a shopping cart for the prediction of what else the customer is likely to buy. Using the recently proposed data structure of itemset trees (IT-trees), we obtain, in a computationally efficient manner, all rules whose antecedents contain at least one item from the incomplete shopping cart. Then, we combine these rules by another technique called Bayesian decision theory to predict the mutually independent items. Finally we introduce a new algorithm based on the Dempster-Shafer (DS) theory of evidence combination which is combined with above techniques to perform well in prediction process.
Existing System:
The existing system focused only on frequently occurring items for prediction process. It uses itemsets trees data structures and rules combined with Bayesian decision theory to predict the mutually independent items. But this approach does not perform well.
Proposed System:
Our proposed system introduces Dempster-Shafer (DS) theory of evidence combination algorithm. DS theory still grow very fast with the average length of the transactions and with the number of distinct items in real world
applications.
List of Modules:
• Item tree generation.
• Rule generation mechanism.
• Bayesian approach.
• DS combination algorithm.
ITEM TREE GENERATION:
This module describes the generation of item sets. These item sets shows the available stock details. All the items are seperated based on their category. Here price,discount,quantity of each items can be maintained by administrator. Each item can be identified by separate item code.
Eg:
Items
Health drinks Sweets
Boost Complan Chocolates Milk sweets
RULE GENERATION MECHANISM:
The proposed rule generation algorithm makes use of the flagged item tree created from the training data set. The algorithm takes an incoming itemset as the input and returns a graph that defines the association rules entailed by the given incoming itemset.
To expedite the rule generation process, we use the Item tree approach that modifies the rule generation algorithm due to two reasons. First, the algorithm addresses a slightly different task, generating all rules of the form. Second, our goal is not to generate all association rules, but, rather, to build a predictor from a set of “effective” association rules.
BAYESIAN APPROACH:
The mathematically “clean” version is known to be computationally expensive in domains where many independent variables are present. Fortunately, this difficulty can be sidestepped by the so called Naive-Bayes principle that assumes that all variables are mutually pairwise conditionally independent.The process of identifying mutually independent items known as Bayesian decision theory.
DS COMBINATION ALGORITHM:
When searching for a way to predict the presence or absence of an item in a partially observed shopping cart, we wanted to use association rules. The question is how to combine the potentially conflicting evidence. One possibility is to rely on the Dempster-Shafer-based Association Rule Mining(DS-ARM) theory of evidence combination.
DS theory assigns to any set, a numeric value called a basic belief assignment (BBA) or mass that quantifies the evidence one has towards the proposition that the given attribute values. This process also provides Body of Evidence(BoE). This BoE describes what are all the evidences to predict the missing items.
