1. International Journal of Computer Engineering and Technology (IJCET), ISSN 0976-6367(Print),
ISSN 0976 - 6375(Online), Volume 5, Issue 3, March (2014), pp. 50-53 © IAEME
50
REVIEW OF STRATEGIES TO IMPROVE THE EFFECTIVENESS
OF MARKETING AND SALES
Pratima O. Fegade1
, Prof. Dinesh D. Patil2
1
M.E. Student, Shri Sant Gadge Baba College of Engineering & Technology,
Bhusawal, North Maharashtra University, India.
2
H.O.D. Computer Science & Engineering Dept., Shri Sant Gadge Baba College of Engineering &
Technology, Bhusawal, North Maharashtra University, India.
ABSTRACT
Marketing means finding out what people wants, why they want it. Marketing is not
advertising. Marketing determines future needs and also have a strategy to meet that needs. Sales
mean converting enquiry into contract. Using effective marketing strategies it is possible to increase
sales revenue and sell more and more product. Today sales and marketing have struggled to work
well together. Customers are constantly moving and more informed. Main objective of any business
is to become competitive. This paper presents survey on several techniques used by researchers to
improve the effectiveness of marketing and sales.
Keywords: Association rule mining, apriori, data mining, éclat, frequent pattern.
I. INTRODUCTION
Marketing is the activity of finding out what the market wants to buy. Main focus of
marketing is to fulfill customer needs through the services and products. Sales try to fulfill sales
volume objective. Marketing identify customer needs, creating products to fulfill that needs and
provide promotion to advertise said products. Marketing provides a way to reach to the customer and
built long lasting relationship whereas selling is the ultimate result of marketing. Main target of
marketing is the construction of brand identity so that it fulfills customer needs. Sales are driven by
human interaction. Sales are simply the identity to meet the needs at right time. For finding out
customers needs it is necessary to track customer purchasing pattern. Customer purchasing patterns
are identified by examining the pattern or items frequently purchase by the customer. For finding the
items frequently purchase by the customer it is necessary to find frequent item sets from the
INTERNATIONAL JOURNAL OF COMPUTER ENGINEERING &
TECHNOLOGY (IJCET)
ISSN 0976 – 6367(Print)
ISSN 0976 – 6375(Online)
Volume 5, Issue 3, March (2014), pp. 50-53
© IAEME: www.iaeme.com/ijcet.asp
Journal Impact Factor (2014): 8.5328 (Calculated by GISI)
www.jifactor.com
IJCET
© I A E M E

2. International Journal of Computer Engineering and Technology (IJCET), ISSN 0976-6367(Print),
ISSN 0976 - 6375(Online), Volume 5, Issue 3, March (2014), pp. 50-53 © IAEME
51
database. There are various methods for finding frequent items such as Apriori algorithm, frequent
pattern growth algorithm, and éclat algorithm.
Apriori algorithm is the most popular algorithm of association rule mining. In many real life
applications of business and industry association rule mining is used. Apriori algorithm is used for
mining frequent item set. Apriori algorithm operates on database containing transactions for example
apriori algorithm operates on the transactions containing set of items purchase by the customer. Each
transaction contains set of items. Apriori uses bottom up approach for finding frequent items.
In association rule mining set of item sets are given for example set of retail transactions,
each contain set of items purchase by the customer. The algorithm finds subsets which are common
to at least a minimum number C of the item sets. Apriori algorithm uses breadth-first search
approach and also uses tree structure to count candidate item sets efficiently. In Apriori algorithm
candidate item sets of length n are generated from length n-1. According to the downward closure
lemma candidate item set contains all frequent n-1 item sets. After that it scans the transaction
database to determine frequent item sets among the candidate. Apriori algorithm suffers from some
inefficiency. In candidate generation process numbers of subsets are generated.
Frequent pattern growth algorithm find frequent items based on the pattern frequently
occurred. One advantage of frequent pattern growth algorithm over apriori is that it discover
frequently purchase items without generation of candidate item sets. In frequent pattern growth
algorithm compact data structure called FP-tree is constructed. This tree is constructed by making
two passes over the data sets. Then frequent item sets are extracted from FP-tree. Frequent pattern
algorithm reduces number of database scan as compare to apriori algorithm. It preserves complete
information for mining frequent patterns and also reduces irrelevant information.
Eclat algorithm is use for mining frequent item sets. Item sets mining allow us to find
frequent pattern like if consumer buys milk, he also buys bread. In éclat algorithm transaction-id set
intersection is used to compute the support of a candidate item set avoiding the generation of subsets
that doesn’t exist in the prefix tree. In this algorithm initial call uses all single items with their
transaction-id sets. After that in each recursive call the function intersects transaction-id sets verifies
each item sets transaction-id set pair with all other pair. If new candidate is frequent then it is added
to the set.
Thus for finding the items frequently purchase by the customers apriori, frequent pattern
growth and éclat algorithm are available. Main idea for improving sales is to examine the order for
the product that have been purchased together. By making the use of this information organize the
store and to place frequently sold products next to each other or in the same area.
II. LITERATURE SURVEY
In [1] Michael J. Shaw, Chandrasekar Subramaniam, Gek Woo Tan, Michael E. Welge have
proposed a technique to manage the marketing knowledge and support marketing decisions. In this
paper they have presented the model for customer retention which accounts for the dynamics of the
today’s market. Their methodology helps for enhancing the customer relationship management.
Their methodology introduces an efficient database encoding technique a novel tree structure called
PC-Tree and PC-Miner algorithm. This paper shows how data mining can be integrated into a
marketing knowledge management framework [1]. The database encoding technique utilizes Prime
number characteristics and transforms each transaction into a positive integer that has all properties
of its items. The PC_Tree is a simple tree structure but yet powerful to capture whole of transactions
by one database scan. The PC_Miner algorithm traverses the PC_Tree and builds the gcd (greatest
common divisor) set of its nodes to mine maximal frequent item sets. Experiments verify the
efficiency and advantages of the proposed method [1].
Xiaohui Yu, Yang Liu, Jimmy Xiangji and Aijun An [2] performed mining on online reviews
for predicting sales performance. They have conducted a case study in the movie domain. For the

3. International Journal of Computer Engineering and Technology (IJCET), ISSN 0976-6367(Print),
ISSN 0976 - 6375(Online), Volume 5, Issue 3, March (2014), pp. 50-53 © IAEME
52
sentiment factor they proposed sentiment PLSA (S-PLSA), in which they considered review as a
document, generated by number of hidden sentiment factor and capture the complex nature of
sentiment. Through S-PLSA they obtain the information embedded in the review. Based on S-PLSA
they proposed ARSA, an Autoregressive Sentiment-Aware model for sales prediction. After that
quality factor was used, with a focus on predicting a quality of review and present ARSQA, to utilize
sentiments and quality for predicting product sales performance.
In [3] Zong Kaisheng has done study on E-marketing. They investigated the dimension of
online sales promotion and proposed a theoretical model on the relationships between online sales
promotion and impulse buying online via the time and energy spended on web browsing. The finding
of this paper converts important messages to scholars by showing the value of exploratory study as
well as empirical study on online sales promotion and its impact on impulse buying online, and
converts novel messages to managers by showing the relationship between online sales promotion
and impulse buying online via web browsing [3].
In [4] Charanjeet Kaur used apriori algorithm for finding frequent item sets. They make the
survey of most recent research work done on apriori algorithm. Association rule mining is the most
important technique in the field of data mining. Association rule mining predict the occurrence of an
item on the basis of occurrence of other items in the transaction. Association rule mining finding
frequent patterns, associations, correlations, or causal structures among sets of items or objects in
transaction databases, relational databases, and other information repositories [4].
Ando T [5] developed the multivariate time series modeling method in the framework of a
general state space model to measure the effectiveness of marketing activities and baseline sales. The
Bayesian approach via Markov Chain Monte Carlo algorithm was used for estimating model
parameters. To evaluate the goodness of model, the Bayesian predictive information criteria were
used.
In [6] Mishra, Patel, Sarangdevot proposed special thought to check the impact of external
factor on the different sector. Whole sale price index is a reflection of the government policy, it
accounts for all trends in demand and supply and it's truly right when we say it is the index of all the
goods which is necessity of life [6].
In [7] Tao Ji, ChongQing introduced the way to integrate the international packaged software
with local customer software, real time sharing of customer profiles, automated meter reading and
invoicing, task distribution for abnormal activities, and other automated processes. They used
standardized coding and centralized platform for integration of onsite customers and sales and
marketing management.
Akbarian R, TavaKoli M, Fard has done study on mobile marketing. Maintaining the stock of
marketing has been most important for companies. Mobile marketing is one of the modern method of
marketing which is considered as a significant factor of competition in development marketing stock,
attraction and maintenance of customer and improvement of interactions by utilizing technology;
whereas mobile marketing is associated with different sciences, so study of mobile marketing subject
is taken into account attractive and new subjects of world trade, information technology and
economics [8].
IN [9] Jian Liu, Yanqing Wang presented model about internet marketing. Based on the
theories of traditional marketing and Internet marketing, the paper builds the model of Internet
marketing relationship that describes consumer's purchase decision process and firm's Internet
marketing process, and correlation among consumer, firm, and bank and logistics firm [9]. By
making the analysis of these two processes the model reveals intrinsic characteristics and essences of
internet marketing that are all different from traditional marketing. The model provides a researchful
platform for the researchers, and a fundamental basis for further researching Internet marketing [9].
In [10] Yu and Lin A, design and implemented search engine marketing management system.
Search engine plays very important role toward the success of business objectives. Search engine
marketing is divided into two categories paid and non-paid. Paid marketing includes pay per click

4. International Journal of Computer Engineering and Technology (IJCET), ISSN 0976-6367(Print),
ISSN 0976 - 6375(Online), Volume 5, Issue 3, March (2014), pp. 50-53 © IAEME
53
and non-paid marketing includes organic search engine optimization and Link popularity. They
investigated the impact of important dimension of Service- oriented architecture, reach and range,
and to design a web-based Search Engine Marketing Management System (SEMMS), which can be
effectively applied to technical aspects in managing paid and non-paid search engine marketing
required in service-oriented architecture dimension.
III. CONCLUSION
Researchers have used different strategies for improving the effectiveness of marketing and
sales. For increasing sales it is necessary to find out frequently sold items. Frequent items are
generated by applying apriori, frequent pattern growth or éclat algorithm. By making the use of
information obtain from frequent item sets we can take more informed decision about product
placement, pricing and promotion, learn more about customer behavior and find out which products
should be cross sold and also find out which products perform similarly to each other.
REFERENCES
[1] Nadimi Shahraki M.H., N. Mustapha, M. N. B. Sulaiman, and A. B. Mamat, A New Method
for Mining Maximal Frequent Item sets, International IEEE Symposium on Information
Technology, Malaysia, pp.309-312, 2008.
[2] Xiaohui Yu, Yang Liu, Jimmy Xiangji and Aijun An, Mining Online Review, Knowledge and
Data Engineering, volume -24, issue 4, april 2012, pp 720-734.
[3] Zong Kaisheng, Online Sales Promotion and Impulse Buying Online in the E-business age: A
Theoretical Model Approach, IEEE 2nd
international conference on Software Engineering and
Service Science, july 2011, pp. 618-621.
[4] Charanjeet Kaur, Association Rule Mining Algorithm using Apriori Algorithm: A Survey,
International Journal of Advanced Research in Computer Engineering & Technology
(IJARCET), Volume 2, Issue 6, June 2013, pp-2081-84.
[5] Ando T, Bayesian State Space Modeling Approach for Measuring the Effectiveness of
Marketing Activities and Baseline Sales from POS Data, 6th
international conference on Data
Mining, December 2006, pp 21-32.
[6] Mishra N, Whole Sale Price Index, Financial Indicators and its Effect on Market and Different
Sector, 5th
international conference on Computational intelligence and Communication
Networks, September 2013.
[7] Tao Ji, ChongQing, Effective Integration of Onsite Customer and Sales and Marketing
Management by Standardized Coding and Centralized Platform, 20th
international conference
and exhibition on electricity distribution, June 2009, pp 1-4.
[8] Akbarian R, TavaKoli M, Fard, Establishment Framework of Mobile Marketing in Iran, 7th
international conference in developing countries”; with focus on e-security, April 2013, pp. 1-
2.
[9] Jian Liu, Yanqing Wang, Research on Internet Marketing Relationship Model, 9th
international
conference on Fuzzy System and Knowledge Discovery, May 2012, pp. 826-30.
[10] Yu, Lin A, The Design and Implementation of a Search Engine Marketing Management
System Based on Service Oriented Architecture Platform, IEEE international Conference on e-
Business Engineering” October 2007, pp. 513-19.
[11] Mr. Sandip Dhakecha, “A Study on Effectiveness of Cause Related Marketing [CRM] As A Strategic
Philanthropy In Terms of Brand Popularity & Sales” International Journal of Marketing & Human
Resource Management (IJMHRM), Volume 4, Issue 1, 2013, pp. 28 - 39, ISSN Print: 0976 – 6421,
ISSN Online: 0976- 643X.