International Journal of Advanced Research in Computer Engineering & Technology (IJARCET)
Volume 7, Issue 6, June 2018, ISSN: 2278 – 1323
544
All Rights Reserved © 2018 IJARCET
Abstract- Online trading even in the retail sector is gaining
more and more popularity with the widespread use of
Internet. The main advantage of online shopping is the
convenience of finding the right kind of products with the
required quality of service, on-time delivery, shipping
charges, in addition to the right price, at the click of the
mouse. However, the problem with such online shopping is
the bewildering array of options, which are available on the
Internet. This causes customers to go by the rating of the
online stores made by the users. These ratings may not be the
true indicators of the online stores.This leads to erroneous
selection of the online stores.In this paper, a novel method is
proposed which works in two stages: In the first stage, it
identifies and extracts the values of the overall rating and the
presence of those features in the full review. The second stage
focuses on establishing the co-relation between overall rating
of the online store and opinion given by the customer in the
full review. The paper also reveals that, there is no similarity
between overall rating and full reviews posted by the
customers. The authors are of the strong opinion that the
customer’s decision of selecting the right online store has to
be based on summary of the full review and overall rating of
only those reviews which are correlated.
Keywords: Online Shopping, Reviews, Rating values
Extractor, Co-Relation.
I. Introduction
Online trading is one of such sector that is gaining
more and more popularity with the widespread use of
Internet. Online shopping offers many advantages in terms
of choice, access to goods and services. A considerable
number of studies have been conducted on the effects of
comments on the web. People’s appraisal has high
confidence to express their behaviors and aspects. Online
shopping allows customers to openly buy things or services
from a seller over the Internet using a web browser.
Customers find a merchandise of interest by visiting
website of the retailer directly or by searching
amongstother vendors by means of a shopping search
engine, which will display the same product's availability
and price at different e-retailers. As of 2016, customers can
shop online using a range of different computers and
devices, including desktop computers, laptops, tablet
computers and smartphones.
An Internet presentation management company held a
survey [1] on over 1400 customers across 11 countries in
Middle-East, North America, Asia and Europe and the
results of survey are given below:
Online retailers must increase the speed of website.
Online retailers should ease customers’ fear of
security.
These worries majorly distress the decisions of nearly two
thirds of the customers.
Fig 1: Year-wise Online shopping growth
Thus, the appraisals on the web are significant
information for customer making their decision. Through
the concept of web, people could share their opinion on the
web freely Many companies develop their business around
this concept. Surfer can obtain lots of useful information
on the web with these reviews added by user rapidly [2].
We get product information only from the manufacturer or
our friends who experienced it in the past time, but through
the experienced people share their comment, we could
know a product more clearly and do better decision. Thus,
it is very essential to know which online store is the best
one where the customer can buy a product [3]. In order to
make this decision the customer has to go through the
opinions on the different online stores.
There are two review formats by which the customer
can give his review
Format(1) – Overall Rating:The reviewer is asked to give
his opinion in the form of rating, in the scale of 1-5 on
some important features of the online stores. The example
of overall rating is shown in the sample web page of Fig.2.
Format (2) – Full review:The reviewer is asked to express
his opinion in the form of sentences. The format of the full
review is shown in the sample web page of Fig2.
A Review Based Web Mining for Online Store
Selection
Sanjeev P. Kaulgud
Department of Computer Science and Engineering, Presidency University, Bengaluru
sanjeev.kaulgud@gmail.com