Вы находитесь на странице: 1из 9

PROJECT SYNOPSIS

ON

(MG-CHARM algorithm)

SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR


THE AWARD OF DEGREE OF

Bachelor of Technology
In
Computer Science &Engineering/ Information Technology
Axis Institute of Technology & Management, Kanpur

Submitted to
MS. SHAIL DUBEY

Submitted by
Nupur verma (1171913015)
Poornima singh(1171913018)
Arun kumar (11719130)
Sarvesh kumar (1171913022)
Anurag saxena (1171913006)

AXIS INSTITUTE OF TECHNOLOGY & MANAGEMENT


HATIPUR,ROOMA KANPUR

INTRODUCTION

The applications of data mining technology are being used in a huge variety of fields. The
data mining applications include the analysis of customer transaction records, web site logs,
credit card purchase information, call records, and many more. The interesting results of data
mining can provide useful information such as customer behavior for business managers and
researchers. One of the most important application data mining applications is association rule
mining.
For association rule mining we need to focus on the methods of finding frequent itemsets in
the database. The Apriori was the first approach to focus on this issue. Apriori finds frequent
items by scanning the database to check the frequencies of candidate itemsets, where as
generated by merging frequent subitemsets. However, Apriori based algorithms have undergone
bottlenecks because they have too many candidate itemsets.
CHARM is another method to find frequent itemsets which is based on Apriori algorithm.
The results of the CHARM give too many candidate itemsets. For this reason we proposed an
algorithm called MG-CHARM which reduces candidate generation and time too. So this is called
fast algorithm for mining minimal generators of frequent closed itemsets.
In our application we use this algorithm for Market Basket Analysis. Market basket Analysis is
a modeling technique based upon the theory that if you buy a certain group of items, you are
more (or less) likely to buy another group of items. For example, people who buy bread and
cheese may also buy butter, jam, milk, and honey. This is a subsystem used to analyze &
improve any organization. We are developing the application of Shopping Mall Transaction
System to implement MG-CHARM algorithm

OBJECTIVE
The applications of data mining technology are being used in a huge variety of fields. The data
mining applications include the analysis of customer transaction records, web site logs, credit
card purchase information, call records, and many more. The interesting results of data mining
can provide useful information such as customer behavior for business managers and
researchers. One of the most important application data mining applications is association rule
mining.
For association rule mining we need to focus on the methods of finding frequent itemsets in
the database. The Apriori was the first approach to focus on this issue. Apriori finds frequent
items by scanning the database to check the frequencies of candidate itemsets, where as
generated by merging frequent subitemsets. However, Apriori based algorithms have undergone
bottlenecks because they have too many candidate itemsets.
CHARM is another method to find frequent itemsets which is based on Apriori algorithm.
The results of the CHARM give too many candidate itemsets. For this reason we proposed an
algorithm called MG-CHARM which reduces candidate generation and time too. So this is called
fast algorithm for mining minimal generators of frequent closed itemsets.
In our application we use this algorithm for Market Basket Analysis. Market basket
Analysis is a modeling technique based upon the theory that if you buy a certain group of items,
you are more (or less) likely to buy another group of items. For example, people who buy bread
and cheese may also buy butter, jam, milk, and honey. This is a subsystem used to analyze &
improve any organization. We are developing the application of Shopping Mall Transaction
System to implement MG-CHARM algorithm.

ABSTRACT

The techniques of data mining have been used widely in various applications. One of the
most important applications of data mining is association rule mining. At present, almost all
algorithms for association rule mining are based on Apriori algorithm. The MG-CHARM
algorithm is a fast algorithm based on the CHARM algorithm which reduces the time for finding
associations between the item sets. The prior method used level-wise mining method to find all
mGs that correspond to each closed item set.
The numbers of frequent closed item sets (FCIs) are usually fewer than number of frequent
item sets (FIs). However, it is necessary to find Minimal Generators (mGs) for mining
association rules from them. The finding mGs approaches based on generating candidate loose
timelines when the number of frequent closed item sets are large.
In this paper, we propose a Minimal Generators for Closed Hierarchical Association Rule
Mining (abbreviated as MG-CHARM) based on CHARM which reduces the time for finding
support between the item sets.
MG-CHARM reduces the complexity of Apriori based association rule mining techniques.

SCOPE OF PROJECT
Although Market Basket Analysis conjures up pictures of shopping carts, and supermarket
shoppers, it is important to realize that there are many other areas in which it can be applied.
These include:

Analysis of credit card purchases.

Analysis of telephone calling patterns.

Identification of fraudulent medical insurance claims.

Analysis of telecom service purchases.

EXISTING SYSTEM
In MG Charm Algorithm based Market Basket Analysis Project, we have proposed MGCHARM algorithm for mining minimal generators of frequent closed item sets. The bottleneck
of CHARM is that the number of frequent items are large and it takes more time. To solve this
problem we decreased the iterations and used new comparison methodFor this we implemented a
Shopping Mall Transaction System. This system maintains customer database that are regular,
employee database who are working in the mall, and the transaction database which stores the
customer and item relations. By this system we used to know the customer interests,
and behaviour. We arrange the items according to the customers interest. This makes flexible to
the customer to select items and we can increase the productivity of the mall.

Limitations of Existing System


At present, almost all algorithms for mining Minimal Generators of FCI are based on
Apriori algorithm. In the first method, the authors extended from Apriori algorithm
to find mGs. Firstly, it found candidates that are mGs, and then defined their closures
to find out frequent closed itemsets.
The second method was proposed by Zaki. Firstly, he found all frequent closed
itemsets using CHARM algorithm. Then he used level-wise method to find out all
mGs that correspond to each closed itemset. Both of these methods have
disadvantage in large size of frequent items since the number of considered
candidates is large.

TECHNICAL REQUIREMENTS

Hardware requirements
The minimum hardware requirement specifications for developing this project are as
follows:
Processor

Standard processor with a speed of 1.6GHz

RAM

256MB RAM or higher

Hard Disk

20GB or more

Monitor

Standard color monitor

Keyboard

Standard keyboard

Mouse

Standard mouse

Software Requirements
The software requirement specifications for developing this project are as follows:

Operating System

Windows XP

Presentation Layer

Java, Servlets, JSP

Web Server

Apache Tomcat 6.0

Database

Oracle 10g XE

IDE

Eclipse 3.3

Database Layer

JDBC

Documentation Tool

MS Office

Rational Rose

UML Tools

PROPOSED METHODOLOGY
Proposed System
Both of these existing methods have disadvantage in large size of frequent items since the
numbers of considered candidates are large. The proposed system overcomes the disadvantages
of existing systems. Hence, we propose a fast algorithm for mining mGs based on CHARM. To
implement this algorithm first we are developing the application of supermarket transactions
system. We apply MG-CHARM algorithm on the database of supermarket daily transactions. It
gives most frequent item sets which increases the productivity provides better maintenance.

Benefits of Proposed System


The data mining applications include the analysis of customer transaction records, web
site logs, credit card purchase information, call records, and many more.
CHARM is another method to find frequent itemsets which is based on Apriori
algorithm.

TIME FRAME REQUIRED FOR VARIOUS STAGES OF PROJECT


IMPLEMENTATION
Sr. No.
1.
2.
3.
4.
5.

PHASES
Synopsis
System Design
Coding
Implementation
Testing & Maintenance

TIME DURATION
1 week
2 week
8 week
2 week
During the running process

Вам также может понравиться