BUSINESS ENTITY SEARCH AGGREGATOR (BESA): WEB ACCESS PATTERNS FOR SUPPLY AND DEMAND CHAIN
Loading...
Date
2010-12
Authors
SEE, KHOO LEAN
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
The importance of search engines becomes more critical as more business
entities create a presence on the Web. Aiming to improve the relevancy of the
search results, BESA Sdn. Bhd., an upcoming startup company is researching and
designing an aggregator that returns related business entities by product type and
recommends a possible supply and demand chain for the product type concerned.
This provides the users with a possible community of the suppliers and customers
which may lead to a better visualization of the potential business opportunities
available. The targeted users are the SMEs in Penang initially, and gradually the
SMEs in Malaysia. To increase the user awareness of this new aggregator, the
company ·Will leverage on the power of internet advertising, viral marketing and
traditional advertising. ; The main revenue of BE_S~ comes from advertisements
placed by various business entities on the aggregator. BESA will only start to
generate a profit of RM85,495 during the second year, with a better prospect of a
total profit ofRM573,285, at the end ofthe third year.
The objective of this research is to present a possible method to establish the
relationship between a product type and its supply and demand chain. A case study
based on the simulated data of possible web access patterns by users has been
examined to establish the possible supply and demand chain relationship for the
LED industry. Part of the research methodology requires the data collection and
preprocessing of the server logs to obtain individual user web access patterns. Web
content mining is then applied to the web access patterns to extract the frequent
keywords. The frequent keywords exhibiting similar characieristics will be
clustered before applying the association rules based on the Apriori algorithm to
discover the relationship between the keywords to form the frequent itemsets. The
xiv
frequent itemsets are then analyzed and assigned to a product type for establishing a
possible supply and demand chain relationship.
The proposed method has been designed and evaluated using simulated user
web access patterns and by comparing it to the suppliers' and buyers' features in
Alibaba.com, a direct competitor of BESA. The results showed promising
improvements, with BESA having a simpler user interface and is able to recommend
the supply and demand chain to the users without the need to search for the chain
separately.
Description
Keywords
BUSINESS ENTITY SEARCH AGGREGATOR (BESA)