Association Rule Analysis: Business Intelligence Application Report

This report provides an in-depth analysis of association rules within the context of business intelligence. It begins with an introduction to business intelligence (BI) and its reliance on data mining techniques for collecting, integrating, analyzing, and presenting information to aid decision-making. The report then surveys five articles related to association rules, exploring algorithms like Apriori and FP-Growth, their applications in market basket analysis, and their role in enterprise systems. The articles cover topics such as the use of association rules in evaluating business intelligence for enterprise systems, association rule mining for business intelligence, comparative surveys of association rule mining algorithms, and performance analysis of Apriori and FP-Growth algorithms. The report highlights the importance of association rules in identifying relationships between data elements, especially in customer purchasing behavior and market analysis. The survey also compares different algorithms and techniques used in the data mining process to enhance the business intelligence process. The report concludes with a summary of the findings and references the sources used.

Running head: ASSOCIATION RULE
Association Rule in Business Intelligence
[Name of student]
[Name of University]

1
ASSOCIATION RULE
Abstract
Association rule is one of the significant and most used processes in the field of data mining and
knowledge discovery. Further, the discovery of business intelligence software and techniques
assist in the data analysis and visualization while determining the underlying information. In
addition to that, various association rules and algorithms are used for in the business intelligence
software and procedure for evaluating the relation between different elements in the database.
This paper evaluates the various processes and conducts a survey of five articles related to
association rule. The survey has provided detailed information regarding the different techniques
and processes employed for the association rule and business intelligence.

2
ASSOCIATION RULE
Table of Contents
1. Introduction..................................................................................................................................3
2. Survey..........................................................................................................................................4
2.1. Introduction...........................................................................................................................4
2.2. Literature Survey..................................................................................................................5
2.2.1. Article 1: Association Rule Approach for Evaluation of Business Intelligence for
Enterprise Systems...................................................................................................................5
2.2.2. Article 2: Association Rules Mining for Business Intelligence.....................................7
2.2.3. Article 3: Comparative Survey on Association Rule Mining Algorithms.....................9
2.2.4. Article 4: Performance Analysis of Apriori and FP-Growth Algorithms (Association
Rule Mining)..........................................................................................................................10
2.2.5. Article 5: Association Rule Mining with Apriori and FP - growth Using Weka........12
2.3. Critical Review/ Analysis...................................................................................................13
3. Summary...................................................................................................................................14
Reference.......................................................................................................................................15

3
ASSOCIATION RULE
1. Introduction
The theory of business intelligence (BI) is often referred to as a combination of various
practices, application and technologies bused for collecting, integrating, analyzing and
presenting new information. Zhao and Bhowmick (2015, p.64) showed that the application of
various BI process allows any business organization to analyze, store and access data for the
decision making process. According to Moro, Cortez and Rita (2015, p.1322), content and data
must not be characterized as separate object, but need to me utilized together in a integrated
format for assisting generating more business opportunities. In order to keep track, analyze and
monitor the significant data, organizations used various technologies and software applications.
Kasemsap (2015, p.25), showed that development of the BI software has been developed with
the significant goal for importing, extracting and analyzing the data for revealing the insight of
business information. Larose (2014, p.6), claimed that in today’s world with the advancement of
information technology, data are being produced in significant velocity and volume. The
exponential increment of data has made it difficult for the business organization to select and
identify the crucial information for assisting in business decision making process. Fan, Lau and
Zhao (2015, p.91), showed that data mining procedures, processes and techniques are being
widely used in both scientific and commercial domain for analyzing and extracting huge amount
of data that are both in structured and unstructured or mixed formation. Over the past decade,
various tools, techniques and algorithm have been proposed and developed for mining
information for the process of BI. Giannotti et al., (2013, p. 388) claimed that various software
and applications used for the Business Intelligence often employ the techniques and theories of
data mining and association rules.

4
ASSOCIATION RULE
This particular report aims at evaluating the role of data mining and association rule in
business intelligence. Apart from that, the reports tend to evaluate the various process and
algorithms of association rules are widely used for data mining. Furthermore, based on the
information gained, the study will try to identify the better process or algorithm that would
enhance the business intelligence process with association rule.
2. Survey
2.1. Introduction
Slimani and Lazzez (2014, p. 143) showed that the techniques of association rules are
widely used in the business operations for the discovery of the various products and services
combinations that the customers tends to purchase together. Further, the application of the
association rules ensures the determination and evaluation of the underlying relationship between
the various products and the hierarchy of products used for purchase of products. This particular
chapter aims at evaluating and reviewing five past articles for evaluating the use of association
rules in business intelligence and various processes of techniques used for data mining in
business organizations.

5
ASSOCIATION RULE
2.2. Literature Survey
2.2.1. Article 1: Association Rule Approach for Evaluation of Business Intelligence for
Enterprise Systems
Reference: Rouhani, S., Ghazanfari, M., Jafari, M., & Akhavan, P. (2012). Association
Rule Approach for Evaluation of Business Intelligence for Enterprise Systems. The IUP Journal
Of Computer Sciences, V, No. 2, 1-19. Retrieved from
https://www.researchgate.net/publication/228133328
Paper is About: In this particular article, the author showed that data mining processes
and techniques are extensively used in the business intelligence software for identifying crucial
information related to business operations. In this article, the authors have highlighted the
significant application of association rule algorithms for data mining process in business
organizations. Furthermore, Rouhani et al., (2012, p.8) have provided a process map using the
association rules algorithm for business intelligence. In addition to that, the article has
represented a combined process used with the apriori and business intelligence techniques used
for data evaluation and analysis in business organization. Apart from that traditional process of
using association rule in business intelligence for extracting information has also been reviewed.
Technical Details: Rouhani et al., (2012, p.2) have defined the role of the association rule
in data mining is for detecting the underlying association or relation in large volume of data set
based on defined nominal attribute values. Further it has been identified that the association rule
is used for discovering the relation between different items used from the huge set of transaction.
Further, the author has used Apriori algorithm as example for of association rule in business
processes.

6
ASSOCIATION RULE
Figure 1: Apriori Algorithm
(Source: Rouhani et al., 2012, p.10)
Rouhani et al., (2012, p.4) have further showed that association rule is widely used for
analyzing market analysis for evaluating the pattern of store types, mail order and purchase of
the items in the supermarket. Confidence and support are the two major variable used in
association rules for determining the “interestingness” of the association rule applied for data
mining. Rouhani et al., (2012, p.1) claimed that the association rule is considered if the evaluated
level of confidence and support are greater that pre-defined confidence and support respectively.
Used in association rule in BI: Rouhani et al., (2012, p.5) have stated that in current
business operations and environments, ERP (Enterprise Resource Planning) systems are widely
used for conducting major operations and computerizing the different procedure used by an
organization. ERP systems are often associated with the revising various documents, filing forms
and logging various transactions while registering the events that occur in the business and
functional module of the organization. Rouhani et al., (2012, p.6) showed that he information

7
ASSOCIATION RULE
and data stored and used in the ERP systems can be used for improving the efficiency and
performance of the business processes. Furthermore it has been identified in the paper that
process mining or business process mining is utilized for explaining and analyzing the behavior
of various process and logs. Therefore, the application of various data mining tools, processes
and algorithms would help in improving the data analysis and evaluation process. Further,
Rouhani et al., (2012, p.4) showed that the process of data mining helps in process of knowledge
discovery and decision making process. In addition to that, the increased use of data mining
application and tools are used for providing tremendous support for improving the quantity
measurement in business operations.
2.2.2. Article 2: Association Rules Mining for Business Intelligence
Reference: Jha, R. (2014). Association Rules Mining for Business
Intelligence. International Journal Of Scientific And Research Publications, 4(5), 1-5. Retrieved
from http://www.ijsrp.org/
Paper is About: This particular paper evaluated the process and techniques used in
business intelligence and data mining. The author has identified a relation between the data
mining and business intelligence used in business operations (Jha, 2014). The paper identified
the essential stages and phases of business while determining the needs of data analysis and
evaluation for the decision making procedure of organization. Furthermore, the paper has aimed
at determining the various challenges and requirement of data mining followed during the
business intelligence process (Jha, 2014). Apart from that, the author has classified the various

8
ASSOCIATION RULE
techniques used in the process of data mining while reviewing and analyzing the association
rules and algorithms that are being used in the business process.
Technical Details: In this paper, data mining has been defined as an issue dependent on
the application. Further, Jha, (2014) claimed that various application required different data
mining techniques and rule for evaluating the detailed information. Jha, (2014) showed that
mining association rule is the major rule that has been widely used by various applications. Two
vital algorithm have been identified in the article, namely Naïve algorithm and Apriori
Algorithm. Further, the author has presented the association of two methods used in the
transaction of the items and determined the support and predictability of the association rules
obtained from both the algorithm.
Used in association rule in BI: Jha, (2014) illustrated that huge data are being generated
by various electronic devices and application of barcodes in the departmental and business has
increased the amount and velocity of data generation. Further, the author has provided example
by stating and on a daily basis, Wal-Mart generated 20 million data based on transaction of good.
The application of the association rule for the mining of information assist the analysis in gaining
information that makes sense to the business processes (Jha, 2014). The application of the
association rule allows the determining and identifying the habit of the customer in purchasing
items from the store. Apart from basket analysis, association rule has significant application in
classification, e-commerce, finances, web mining, customer segmentation and marketing.

9
ASSOCIATION RULE
2.2.3. Article 3: Comparative Survey on Association Rule Mining Algorithms
References: Girotra, M., Nagpal, K., Minocha, S., & Sharma, N. 2013. Comparative
Survey on Association Rule Mining Algorithms. International Journal Of Computer
Applications (0975 – 8887), 84(10), 18-22. Retrieved from http://www.ijcaonline.org/
Paper is About: Girotra et al., (2013, p.20) have identified the popularity of association
rule in the process of data mining among the business organization and marketers. The author
has showed that association rule in data mining and discovery creates various subset of the large
item set for determining the existing relationship between the items. In this paper, the author has
identified that the association rule is divided into two significant parts. The initial phase
determines the most frequent sets used among the large volume of unstructured or structured data
(Girotra et al., 2013, p.18). While the second phase is used to determine the association rule or
relationship between them. The paper has evaluated the different algorithm and process that are
used for the association rule identification in knowledge discovery.
Technical Details: The authors have identified eight significant algorithms that are
frequently used in the association rules. The algorithm identified includes FP- Growth, Recursive
Elimination, Eclat, Apriori hybrid, AprioriTID, Apriori, SETM and AIS. Various characteristic
and functions of the different algorithms has been identified in the article. In addition to that, the
authors have developed a detailed comparison of the various algorithms based on their features
and functions (Girotra et al., 2013, p.20). From the detailed comparison and analysis it has been
identified that use of Apriori Algorithm is most useful in case of closed set of items while Eclat
is most advantageous in case of free set of items. The author showed that Apriori Algorithm,
showed more efficiency and advantageous in performance and operations considering the other
entire algorithm identified. Furthermore, it has been illustrated that the technique and process of

10
ASSOCIATION RULE
Recursive Elimination were better considering all the features of Apriori but poorer when
compared to Eclat algorithm used.
Used in association rule in BI: The theory of association rule has been developed for
identifying the most frequent item set purchased and often considered as the market basket
analysis pattern or process. Girotra et al., (2013, p.21) claimed that the business executives and
the professionals often uses the theory and technique of association rue for developing the layout
plan and to place the services and items placed together for increasing the purchase rate and
efficiency. Furthermore, the author claimed in today’s business world, where technology and
hand held devices are widely used, data are generated at high rate. The detailed information
related to business are essential for developing various strategies related to business analysis
(Girotra et al., 2013, p.22). The application of business intelligence helps in extracting raw
information and provides structure understandable data that are useful for decision making
process in business. Therefore, the application of association rule and data mining technique
would help the analyst in performing the data evaluation in easier and much efficient manner.
2.2.4. Article 4: Performance Analysis of Apriori and FP-Growth Algorithms (Association
Rule Mining)
Reference: Bala, A., Shuaibu, M., Lawal, Z., & Zakari, R. 2016. Performance Analysis
of Apriori and FP-Growth Algorithms (Association Rule Mining). Alhassan Bala Et Al,
Int.J.Computer Technology & Applications, 7 (2), 279-293. Retrieved from http://www.ijcta.com
Paper is About: In this paper, the author has identified that association rule and data
mining techniques are widely used in various data analysis and mining process all over various
industries (Bala et al., 2016, p.283). The selected paper aims at defining the association rule and

11
ASSOCIATION RULE
the concept of data mining used for data analysis and knowledge discovery in various business
organizations. The author has identified the two significant algorithm of association rule namely
FP Growth and Apriori for comparison and analysis of the difference of performance on the data
identification and analysis (Bala et al., 2016, p.291). In addition to that, the author has used
WEKA software for determining and evaluating the effectiveness of the performance of the
association rule.
Technical Details: While analyzing the apriori algorithm, the author has identified the
procedure and techniques used for determining the association rule. The author has showed
apriori runs on four significant steps. The initial step includes the generation of the element
dataset of most frequently used items (Bala et al., 2016, p.286). The step is followed by join step
process for identifying the number of candidate in the frequent step and self join with each other.
The necxt step is followed by the pruning of the frequent dataset. In the last step, the data set that
are least used are pruned from the association rule. Bala et al., (2016, p.283) showed that
application of apriori algorithm uses more space results in the increased database. Similarly, FP
Growth analysis has been analyzed for the evaluation of the performance. It has been identified
that FP growth uses two phase process, first for developing the dataset and second phase for
traversal of the FP tree.
Used in association rule in BI: the association rule in the process of business intelligence
can be utilized for digging out the relevant information regarding various process and technique
used in business processes. The author has demonstrated that the applications of association rule
in the business intelligence process allows in extracting hidden and underlying information from
huge dataset and network for determining the potential association between the neglected items
and predict the trends that can be used in decision making procedure (Bala et al., 2016, p.289).

12
ASSOCIATION RULE
2.2.5. Article 5: Association Rule Mining with Apriori and FP - growth Using Weka
Reference: Mishra, A., Pani, D., & Ratha, D. 2015. Association Rule Mining with
Apriori and FP - growth Using Weka, 2837 -2845.
Paper is About: Mishra, Pani and Ratha (2015, p.2841) have showed association rule is
one of the significant technique for mining significant information. In this paper, the author has
determined and evaluated the significant technique and process used for the two major
association rule including Apriori and FP Growth (Mishra, Pani and Ratha 2015, p.2838). The
paper determines the effectiveness and performance of the two major algorithms. The author has
used WEKA tools for determining the efficiency of the algorithm. Based on the analyzing and
association rule generation in WEKA application,
Technical Details: The paper showed various clustering technique used in the association
rule including Model-based methods, Grid-based methods, Density based methods, Hierarchical
Agglomerative and Partitioning Methods. Mishra, Pani and Ratha (2015, p.2840) author showed
that Apriori Algorithm uses large dataset that are kept in lexicographic order. Furthermore, it has
been identified that in Apriori method, the item set are detected on the basis of various pass and
transaction while eliminating the item assets that are less frequently used. On the other hand, FP
Growth algorithm is characterized with prefix tree structure, fragment pattern growth and
frequent pattern for developing the FP tree.
Used in association rule in BI: The author has claimed that association rule is one of the
revolutionary technology and procedure used for determining the data evaluation and pattern.
The process of association rule in data mining allows in determining the correlation between the
different patterns in huge volume of relational database. Therefore, the author claimed that the

13
ASSOCIATION RULE
association rule are used in the business intelligence on data warehouse for managing the
multidimensional data and accessing information for professional use and decision making
procedure of the business organization (Mishra, Pani and Ratha 2015, p.2839). The data mining
technique provides visualization, analysis, recognition and warehousing of huge information for
enhancing the performance and efficiency.
2.3. Critical Review/ Analysis
The above five journal articles have demonstrated and evaluated the various algorithms
and processes used in the data mining and association rule for identifying the underlying relation
between huge dataset Zhao and Bhowmick (2015, p.64). The detailed literature survey of the
various papers has showed that association algorithm and apriori algorithm are widely used in
various process and techniques. The literature survey has demonstrated the significant and
performance analysis of the various association algorithm and observed that Apriori algorithm
has maximum amount of performance efficiency when considered between various algorithms
with closed dataset. Further, it has been identified that all the authors have included the
application of Apriori Algorithm in their analysis and evaluation process Moro, Cortez and Rita
(2015, p.1322). The application of the Apriori algorithm can be considered standard for data
mining and evaluation for large dataset. In addition to that, from the data evaluation of the
survey, it has been observed that authors tends to use WEKA software tools for evaluation and
determining the efficiency of the algorithm while developing the association rule between the
algorithm. In addition to that, the application of association algorithm for data mining, evaluation
and visualization provides detailed information about the underlying process and items that can
be used for the decision making process of the business. Therefore, it can be claimed that
association rule for data mining is a subset of the business intelligence process used in various

14
ASSOCIATION RULE
industries. Furthermore, Slimani and Lazzez (2014, p. 143) claimed that based on the needs and
requirement of the business intelligence software and the observation, various algorithm are
employed for determining the existing and underlying trends and pattern in the information.
3. Summary
The above survey of past literature has able to provide detailed information about the
various association algorithm used in the data mining process. Various business intelligence
software and systems including ERP, CRM and CSM used data mining algorithms, tools and
techniques for taking business intelligence. The discovery of association rule and algorithm have
increased and improved the efficiency of the knowledge discovery in business intelligence
process. The detailed survey focused on the application of association rules and techniques for
data mining process. The survey was able to provide detailed information regarding the different
procedure and need of association rule in the business intelligence. The detailed survey showed
that the apriori algorithm provides efficient result in the process mining and evaluation of
detailed information. Therefore, the above evaluation was successful in determining the
significant process, technique and procedure that are frequently used in association algorithm. In
addition to that, the survey has focused and showed light in the application of association rule for
the data intelligence and fact finding technique used in the business intelligence procedure of
various business organizations.

15
ASSOCIATION RULE
Reference
Bala, A., Shuaibu, M., Lawal, Z., & Zakari, R. 2016. Performance Analysis of Apriori and FP-
Growth Algorithms (Association Rule Mining). Alhassan Bala Et Al, Int.J.Computer Technology
& Applications, 7 (2), 279-293. Retrieved from http://www.ijcta.com
Fan, S., Lau, R. Y., & Zhao, J. L. 2015. Demystifying big data analytics for business intelligence
through the lens of marketing mix. Big Data Research, 2(1), 28-32.
Giannotti, F., Lakshmanan, L. V., Monreale, A., Pedreschi, D., & Wang, H. 2013. Privacy-
preserving mining of association rules from outsourced transaction databases. IEEE Systems
Journal, 7(3), 385-395.
Girotra, M., Nagpal, K., Minocha, S., & Sharma, N. 2013. Comparative Survey on Association
Rule Mining Algorithms. International Journal Of Computer Applications (0975 –
8887), 84(10), 18-22. Retrieved from http://www.ijcaonline.org/
Jha, R. 2014. Association Rules Mining for Business Intelligence. International Journal Of
Scientific And Research Publications, 4(5), 1-5. Retrieved from http://www.ijsrp.org/
Kasemsap, K. 2015. The role of data mining for business intelligence in knowledge
management. Integration of data mining in business intelligence systems, 12-33.
Larose, D. T. 2014. Discovering knowledge in data: an introduction to data mining. John Wiley
& Sons.
Mishra, A., Pani, D., & Ratha, D. 2015. Association Rule Mining with Apriori and FP - growth
Using Weka, 2837 -2845.

16
ASSOCIATION RULE
Moro, S., Cortez, P., & Rita, P. 2015. Business intelligence in banking: A literature analysis from
2002 to 2013 using text mining and latent Dirichlet allocation. Expert Systems with
Applications, 42(3), 1314-1324.
Rouhani, S., Ghazanfari, M., Jafari, M., & Akhavan, P. 2012. Association Rule Approach for
Evaluation of Business Intelligence for Enterprise Systems. The IUP Journal Of Computer
Sciences, V, No. 2, 1-19. Retrieved from https://www.researchgate.net/publication/228133328
Slimani, T., & Lazzez, A. 2014. Efficient analysis of pattern and association rule mining
approaches. arXiv preprint arXiv:1402.2892.
Zhao, Y., & Bhowmick, S. S. 2015. Association Rule Mining with R. A Survey Nanyang
Technological University, Singapore.