an e-commerce environment where millions of transactions take place between the
providers and users, a need for establishment of validity of the service
provided arises. A customer feedback system has been provided by the
marketplace operators in order to fulfill such need. But the feedback generated
may not be always relied upon. The feedback may positively or negatively affect
its sales, instead of showcasing the actual genuineness of the product or
service, in customer’s point of view. Our work proposes an enhancement to
traditional feedback system by introducing an Trust Reputation System (TRS)
which helps filtering out the valid customers using a set of algorithms,
thereby creating a trust degree for the user.
Keywords- component; Trust Reputation
System, Opinion Mining, Sentiment Analysis.
The consumers in the online market
face the problem of filtering out the best products from a list of variety of
options.There are various marketplace operators who provide feedback system to
help customer identify quality products, by reviewing the customer opinion and
accordingly choose the product. Most of the consumers buy products based on
This either negative or positively affect the sale of the products. Also,
this paves a way for spammers for
decreasing the sale of the product . To eliminate this, the paper focuses on
enhancing the feedback system by introducing the concept of trustworthiness
.This can be done through Trust Reputation System. TRS are programs that allow users to rate each other. Using such methods
can help decrease the number of
spammers, thereby potentially increasing the amount of genuine reviews. The
advantage of such reviews is it helps
determining the genuiness of the product.
II. RELATED WORK
Sentiment analysis has been studied in wide area of domain
such as movie review, teaching review, product review, e-learning, hotel review
and many more. Most scholars focused to quantitative data analysis. However,
some studies have been done on qualitative data using sentiment analysis, we
found six works that mentioned the idea of using opinion mining and sentiment
analysis in education.
Algorithms such as Naive Bayes, k-means and Support Vector
Machine are used in opinion classification. The paper also focuses on the truth
reputation system. There exists several truth reputation system architectures
having different algorithms to calculate the reputation score related to the
Many authors 1,2,3,4,5,6 have proposed in their
work several TRS architectures with different algorithms to calculate
reputation score related to the product. Also, a few academic work on Truth
reputation system has been devoted to the inclusion of the semantic analysis of
feedbacks in the calculation of the trust score of the product and specially
the trust degree of the user. Even in studies attempting to provide more
complex reputation methods, some issues are still not taken into consideration,
such as the credibility of referees, the update of the trust degree of the user
at any intervention, the age of the rating and the feedback or the concordance
between the given rating which is a scalar value and the textual feedback
associated to it. In contrast to the mentioned TRS, our proposed design
overcomes these issues and makes use of
an algorithm which includes analysis of textual feedbacks in order to
calculate the trust degree of the user giving the feedback and a
trustful reputation score for the product.
The consumers in the online market
face the problem of filtering out the best reviews or feedback for purchase of
the products. We try to eliminate the problem by listing out the best reviews
so that it becomes easy for the customers to decide on a product by analysing other
consumer experiences, by allowing them to post their reviews. Consumers dealing
with the online market might sometimes buy substandard products. Though the
e-commerce company provide facilities like return and exchange of products, the
process becomes a tedious task sometimes. The project aims to provide the
costumers an opportunity to select the desired products based on the rating of
the item they wish or plan on to buy, which has been evaluated on the basis of
rating and reviews contributed by the consumers with the help of a Truth
Reputation System (TRS).
IV. FEATURES OF THE PROJECT
The Opinion Mining of our project will be based on Sentiment
analysis algorithms & methods and also on Truth Reputation System algorithm
.Trust Reputation Systems (TRS) will provide the necessary information to
support relying parties in taking the right decision in any electronic
transaction. In fact, as security providers in e-services, TRS have to
faithfully calculate the most trustworthy score for a targeted product or
service. Thus, TRS must rely on a robust architecture and suitable algorithms
that are able to select, store, generate and classify scores and feedbacks.
V. PROPOSED WORK
In the proposed architecture, for each user who wants to
leave a rating (appreciation) and a feedback (semantic review), we analyse the
customers attitude towards a number of short and selected feedbacks and stored
by product in the knowledge base. This user’s review is going to be reached by
any other user. Then, we suppose that we have a path relaying all the users
(the nodes). As a result, we need to know the trust degree of the user and
determine the trust degree of the feedback.4
Trust Reputation System Design
The customer starts by giving a rating and a textual
feedback about a specific product. When they click on submit, in order to
validate the given information, we are going to redirect the user to another
interface showing this message for example: “please give us your opinion about
the following feedbacks before validating the information you gave below:” In
this interface we will find chosen feedbacks from the database from different
types. Those feedbacks can be fabricated in order to summarize numerous users
feedbacks stored in the database. The generated feedbacks can be stored in
another knowledge base. So as much as we add feedbacks in the ordinary
database, we will fill the knowledge database with prefabricated feedbacks
using text mining algorithms and tools. However, some users can give already
summarized feedbacks that can directly be included in the knowledge database.
Indeed, there are many text mining and data mining algorithms and tools that
could search the most appropriate feedbacks that are first of all related to
the product and that can recapitulate and summarize most of each type of the
Actually, before sending the customers feedback and
appreciation about the product to the trust reputation system, we have to
verify the concordance between them in order to avoid and eliminate
contradiction or malicious programs attacking our system. In the redirected
interface, we will display several feedbacks from different types. However, the
user can specify the number of feedbacks to be liked or disliked. Of course, we
can also specify the minimum and the maximum number of feedbacks to be
displayed by the user.
In fact, we are trying through this redirection to detect
and analyse the user intention behind his intervention on the e- commerce
application. Hence, we examine and evaluate his intention using other pre
fabricated feedbacks with different types. Of course, we have already the
trustworthiness of each feedback. Consequently, we use our reputation algorithm
studied in section 4.2 in order to generate the user trust degree which plays
the role of a coefficient and then rectify his appreciation according to his
trust degree and generates the score of the feedback. Indeed, each feedback has
trustworthiness in a threshold -5,5. The closest is the trustworthiness to 5,
the most trustworthy the feedback is. The closest is the trustworthiness to -5,
the very untrustworthy is the feedback. If the feedback is trustworthy its
score would be included in 0,5 else it would be included in -5,0.4
B. TRS algorithm
Reputation algorithm used in this TRS is using semantic
feedbacks analysis in order to generate a trustful reputation score for the
product.Actually, we have 3 types of feedbacks:
** Positive feedbacks: represent opinions that expressing a
positive point of view about the product. Those ameliorative opinions contain a
positive content concerning the product. Then, the adjective positive is
referring to the nature of the content of the feedbacks not its
trustworthiness. However, each feedback whatever is its type can have either a
positive trustworthiness or a negative trustworthiness. Either positive
trustworthiness or negative one, it is gradual: it has degrees as float in a
threshold of -5.5.
**Negative feedbacks: represent opinions talking negatively
about the product. Logically, the users giving such opinions are not satisfied
of the commented product. This feedback could be telling the truth or a part
from the truth or could be far from the truth. That’s why, each feedback has
its trustworthiness represented by a float number between -5 and5.
**mitigated feedbacks: represent feedbacks that are talking
positively about some aspects of the product and negatively about other aspects.
They are also characterised by trustworthiness included in -5.5.
**contradictious feedbacks: represent feedbacks with a
contradictious content for example a feedback where the user is not talking
about the specified product but another one or he/she is affirming that the
camera of a mobile phone is great and later in the same opinion is saying that
the camera is very bad. In fact, we have to start by detecting the
contradictious feedbacks. Then we are in need of a semantic analysis algorithm
and tool that can detect the contradiction in a specific content related to a
product. We can personalize the analysis according to the product. For
instance, if the user says that “the swimming pool of the hotel which does
not afford one is not clean”, the
algorithm must be able to detect this great contradiction. We can give to the
algorithm for each product as an input the property of the algorithm; if there
is no similarity we can consider it as a contradiction. But the agreement
includes the meaning of course. Because if the customer writes that the
negative thing about this hotel is that there is no swimming pool. He is telling the truth then obviously the presence
of an absent property in a feedback doesn’t mean that there is a contradiction.
Actually, before sending the customers feedback and appreciation about the
product to the trust reputation system, we have to verify the concordance and
the alliance between them so we don’t have contradiction.
After verifying the concordance between the appreciation and
the textual feedback we are going to redirect the user to the selection of
prefabricated feedbacks. Then the user is going to click on like or dislike
according to each feedback. The event of click will be managed in order to get
some information needed in the calculus of the trust degree of the user. The
function uses as a parameter the id of the feedback in order to get from
Knowledge base its
trustworthiness. We need to get also the previous trust
degree of the user if he has been already engaged in a transaction or he has
used the application for rating purpose. The user choices either “like” or
“dislike” is an important parameter to determine his trustworthiness.4.
Intially, the user gives a rating and a textual feedback
feedback about the purchased product.
Then we validate the information provided through an interface.In fact, in this
interface we will find chosen feedbacks from the database from different types.
The feedbacks can be used to summarize numerous users feedbacks stored in the
database. The generated feedbacks can be stored in other knowledge base. So as
much as we add feedbacks in the ordinary database, we will fill the knowledge
database with prefabricated feedbacks using text mining algorithms and tools.
However, some users can give already summarized feedbacks that can directly be
included in the knowledge base.Actually, before sending the user?s feedback and
appreciation about the product to the trust reputation system, we have to
verify the concordance and the alliance between them so we don’t have
Test for measuring the contardiction
in the feedback.
Pseudo-code to verify the
concordance between the rating and
the textual feedback:
concordance =Test_ concordance
(int appreciation, string feedback) ;
//redirection to the feedbacks
URL (url_page); // we thank the
user for his intervention and we put
him temporally in a //blacklist
After measuring the concordance the feedback is sent to
Trust Reputation System for further processing.At the final stage we get only
filtered feedback.Hence only genuine feedback about the product are generated.
Lack of information regarding
particular products leads to wrong selection of product which in turn leads
to huge holes in pockets of the
customers. Thus we aim to provide the accurate and true reviews about the
particular products which will help customers in picking up the right
product. We attempt to calculate the
trust degree of the user according to his subjective choice either “like” or
“dislike” and according to the feedback.Those results such as trust weight and
scores help users making a decision about purchasing or not a product from an
e-commerce application. However those scores are not always truthful. Then,
they can falsify the weight and the ratings. Semantic feedbacks are more
meaningful than single scores.
The consumers dealing with our website would be able to access precise data and
reviews of the consumers feedback and use it intelligently for product selection and for buying of it as
well . This software would be useful for any similar e-commerce business
dealing with problems regarding the issues of trustworthiness of reviews.The
provision of visual representation can be used by customers to buy genuine
products. On some extent it would also help the marketplace operators and
vendors to filter out their potential customers. In today’s time data is said
to be the biggest asset for any company or organization. Thus, it is of immense
importance to analyses the data and get some results out of it.
We sincerely thank to our guide Mrs.
Purvi Sankhe, our HOD Dr. Rajesh S. Bansode, our Dean Dr. Kamal Shah and our
principle Dr. B. K. Mishra for his/her guidance and support for carrying out
our project work.
1 The Analysis and Prediction of
Customer Review Rating Using Opinion
Mining Wararat Songpan Department of Computer Science, Faculty of Science, Khon
Khon Kaen, Thailand.
2 A. Jøsang R. Hayward Simon Pope:
Trust Network Analysis with Subjective Logic. Proceedings of the Second
International Conference on Emerging Security Information, Systems and
Technologies (SECURWARE 2008), Cap Esterel, France, August 2008.
3 Fereshteh Ghazizadeh Ehsaei, Ab.
Razak Che Hussin: Acceptance of Feedbacks in Reputation Systems: The Role of
Online Social Interactions Information Management and Business Review Vol. 4,
No. 7, pp. 391-401, July 2012 (ISSN 2220-3796).
4 A New Reputation Algorithm for
Evaluating Trustworthiness in E-Commerce Context Hasnae RAHIMI1, Hanan EL
BAKKALI2 Information Security Research Team (ISeRT) Université Mohammed V-
5Co-Extracting Opinion Targets and
OpinionWords from Online Reviews Based on the WordAlignment Model
Kang Liu, Liheng Xu, and Jun Zhao
6 A. Gutowska and A. Sloane:
Modelling the B2C Marketplace: Evaluation of a Reputation Metric for
e-commerce. Proceedings of Web Information Systems and Technologies – WEBIST ,
pp. 212-226, 2009.