=Paper=
{{Paper
|id=None
|storemode=property
|title=Product Reputation Model: An Opinion Mining Based Approach
|pdfUrl=https://ceur-ws.org/Vol-917/SDAD2012_2_Abdel-Hafez.pdf
|volume=Vol-917
|dblpUrl=https://dblp.org/rec/conf/pkdd/Abdel-HafezXT12
}}
==Product Reputation Model: An Opinion Mining Based Approach==
https://ceur-ws.org/Vol-917/SDAD2012_2_Abdel-Hafez.pdf
Product Reputation Model: An Opinion Mining Based
Approach
Ahmad Abdel-Hafez, Yue Xu, Dian Tjondronegoro
School of Electrical Engineering and Computer Science
Queensland University of Technology
Brisbane, Australia
ahmad.abdelhafez@student.qut.edu.au, {yue.xu,dian}@ qut.edu.au
Abstract. Product rating systems are very popular on the web, and users are in-
creasingly depending on the overall product ratings provided by websites to
make purchase decisions or to compare various products. Currently most of
these systems directly depend on users’ ratings and aggregate the ratings using
simple aggregating methods such as mean or median [1]. In fact, many websites
also allow users to express their opinions in the form of textual product reviews.
In this paper, we propose a new product reputation model that uses opinion
mining techniques in order to extract sentiments about product’s features, and
then provide a method to generate a more realistic reputation value for every
feature of the product and the product itself. We considered the strength of the
opinion rather than its orientation only. We do not treat all product features
equally when we calculate the overall product reputation, as some features are
more important to customers than others, and consequently have more impact
on customers buying decisions. Our method provides helpful details about the
product features for customers rather than only representing reputation as a
number only.
Keywords: reputation model, opinion mining, features impact, opinion strength
1 Introduction
Many websites nowadays provide a rating system for products, which is used by
customers to rate available products according to their own experience. Reputation
systems provide methods for collecting and aggregating users’ ratings to calculate the
overall reputation for products, users, or services [2]. This final rate is very important,
as it represents the electronic ‘word of mouth’ that customers build their trust in a
product on. On the other hand, most websites allow customers to add textual reviews
to explain more about their opinion to the product. These reviews are available for
customers to read, to the best of our knowledge, they are not analyzed and counted in
the product overall reputation. Many reputation models have been proposed, but most
of them concentrated on user’s reputation in C2C (Consumer to Consumer) websites
such as eBay.com, while service and product reputation has received less attention.
Besides, most of the literature about product reputation models neglected users’ re-
adfa, p. 1, 2011.
© Springer-Verlag Berlin Heidelberg 2011
16
views and counted users’ ratings only. Therefore, their reputation systems did not
provide any summaries and details about the weakness and strength points in the
product.
In this work we will provide a reputation model for products using sentiment anal-
ysis methods. The proposed model generates reputation for a specific product depend-
ing on the textual reviews provided by users rather than depending on their ratings
because users’ ratings do not reveal an actual reflection for the products’ features, and
they do not provide details for customers about features reputation and about “why”
the reputation is high or low. For example, a strict user might give three stars for the
product although he does not have a clear negative opinion about the product. On the
other hand a more generous customer might have a couple of negative opinions about
the product but still give four stars. Additionally, textual reviews can be used to pro-
vide summaries about product features reputation in addition to the aggregated value
for the product reputation, which can make the reputation system more meaningful
rather than being just a number. We calculate features impact by counting how many
times every feature is mentioned explicitly in the text reviews, assuming that features
that are mentioned more by users are more important for them.
In the rest of this paper, we will demonstrate couple of existing product reputation
model in the section II, and in the following sections we will explain equations we use
to calculate the reputation value for a product. We will also provide diagrams to show
the difference between the results of our reputation calculation method and the regular
average method used by most websites to represent the overall product reputation.
2 Related Work
2.1 Reputation Models
Reputation models have been studied intensively by many researchers in the last
decade, many of these researches concentrated on user’s reputation and some of them
have discussed product reputations. One of the most basic works on ratings aggrega-
tion analyzed robustness of different aggregators, in particular the mean, weighted
mean, median and mode, and proposed that using median or mode is more efficient
than using mean [1]. Cho et al. [3] proposed a more sophisticated model, they calcu-
lated user reputation and used it in order to calculate weights for different ratings.
Moreover, they assumed that some users tend to give higher ratings than others,
hence, they calculated rating tendency for users and deducted it from user rating.
They used the user’s accurate prediction and the degree of his activity to define his
level of expertise, and then they used this value to represent user’s reputation. This
method might not be an accurate way to give different weights for ratings, because a
user’s reputation should not reduce the weight of his opinion about a product. On the
other hand, another promising work introduced by Leberknight et al. [4], discussed
the volatility of online ratings, where authors aimed to reflect the current trend of
users’ ratings, they used weighted average where old ratings have less weight than
current ones. They introduced a metric called Average Rating Volatility (ARV) that
17
captured the extent of fluctuation present in the ratings, and then they used it to calcu-
late discounting factor, which is used in weighting older ratings.
2.2 Opinion Mining
Many literatures have focused on extracting useful information from the huge
amount of available users' opinions in the internet. Opinion mining was used in many
different domains. Business Intelligence is the most popular one, where many studies
concentrated on mining customers' reviews for better market understanding [5]. Re-
searchers focused on the sentiment analysis part and represented product reputation as
a simple count of positive and negative sentiments [6] [7]. Turney [8], Pang et al. [9],
and Kamps et al. [10] provided different methods to determine the orientation of a
word as positive or negative. In contrast, Hu & Liu [6] proposed a set of techniques
for mining and summarizing product reviews to provide a feature based summary of
customer reviews, they searched for frequent noun and noun phrases as candidate
features. While Popescu et al. [11] identified parts and features of a product depend-
ing on finding relation between noun words and the product class using PMI algo-
rithm [8]. Morinaga et al. [12] were one of the first researchers to introduce a general
framework for collecting and analyzing users’ reviews in order to find the overall
product reputation. They used two dimensional positioning Maps, which contained
the extracted opinion phrases and associate products with them. The distance between
opinion-phrases and products represents closeness. Their proposed method does not
mine product features [6], which might be crucial element in the product reputation
analysis. In contrast, Hashimoto & Shirota [13] depended on buzz marketing sites to
provide a framework for reputation analysis considering product’s features. They
attempted to discover the topic of each review as initial step, and then they deter-
mined important topics depending on the contribution rate of each topic and the polar-
ity of the messages. Finally, the results are visualized for users. However, the effec-
tiveness of their framework has not been evaluated, and the visualization method used
to represent the results has not been perfected. Moreover, they neglected topics with
lower contributions which might affect the overall product reputation.
To the best of our knowledge none of the previous work has proposed a convenient
method to calculate product reputation, depending on the outcome of mining users’
reviews. Most of the available methods represent the reputation as a simple count or
average of positive and negative opinions in the reviews. While the convenient
represented models depended on users’ ratings rather than users’ textual reviews.
3 The Proposed Approach
3.1 Definition
A product can be described by a set of features representing its characteristics.
Some of the features may be more specific or more general than others. For example,
for a specific mobile phone product, the “Mobile Camera” is considered as a general
feature, while Resolution, Optical Zoom, Flash Light, Video Recording are more
18
specific features of Mobile Camera. In this paper, we define product features as a
hierarchy.
Definition 1 (Feature hierarchy): A feature hierarchy consists of a set of features
and their relationships, denoted as FH # !F , L" , F is a set of features where
F # ! f1 , f 2 ,..., f n "and L is a set of relations. In the feature hierarchy, the relationship
between a pair of features is the sub-feature relationship. For f i , f j $ F , if f j is a
sub-feature of f i , then ( f i , f j ) $ L , which means, f j is more specific than f i . The
root of the hierarchy represents the product itself, and the first level children are the
generic features. In this paper, we assume that the feature hierarchy is available.
Definition 2 (User’s Review): R is a set of reviews where R # !r1 , r2 ,..., rm " .
Every review consists of a number of opinions about different features, denoted as
%ri $ R ri # {( f i1 , oi1 , s i1 ),..., ( f in , oin , s in )} . oij is the orientation of the opin-
ion; oij $ !Pos, Neg , Neu" , which represents positive, negative, and neutral respec-
tively. si is the strength of the opinion, si $ !1,2,3" , where 1 represents “weak opin-
ion”, 2 for “moderate”, and 3 for “strong opinion”.
In this paper, we assume that the product features and the opinion orientation and
strength to the features in each product review have been determined by using exist-
ing opinion mining techniques. The proposed reputation model will generate product
reputation based on the opinion orientation information, i.e., this information is avail-
able, and is the input to the reputation model. There are different methods that can be
used to extract this information [14] [15].
For a specific feature f j , the set of negative reviews are denoted as
R neg
j !
# ri oij # Neg " , R neg
j # +*r1neg , r2neg ,..., r Rnegneg (' ,
) j &
where %ri neg $ R neg
j ,
neg
ri # {( f i1neg , oineg neg neg
1 , si1 ),..., ( f ij , oijneg , sijneg ),...}
pos
The same definitions also apply for the set of positive reviews R j . The neutral
orientation reflects the lack of opinion about the specific feature and consequently
will not be considered in the reputation model.
Our proposed product reputation model consists of three stages:
, Feature Reputation: the reputation of every feature is calculated based on the
frequencies of positive and negative opinions about the features and its sub
features.
, Features’ Impact: feature impact is used to give a different weight for every
feature depending on the number of opinions available in users’ reviews
about this feature.
, Product Reputation: the final product reputation is the aggregation of fea-
tures’ reputations.
19
In the following sections we will describe them in details.
3.2 Feature Reputation
The basic idea of the proposed model is to generate the reputation of a product
based on the reputation of the product’s features. The reputation of each feature is
generated based on the opinion orientation and strength of its sub features. For a
feature f i , the reputation of f i will be the aggregation of the positive and negative
opinions weights for all of its sub-features f j , where ( f i , f j ) $ L as mentioned in
Definition 1. This section will discuss how to derive feature reputation based on sub
features’ opinion information.
Negative Opinion Weight
In this part we suggest a formula to give more weights for frequent negative opi-
nions about a specific feature. By “frequent”, we mean that the negative opinion about
a feature has occurred in many reviews. Frequent negative opinions may indicate a
real drawback in the product, where there is a larger probability that a customer will
have the same problem if he buys this product. Thus, when more reviews share a neg-
ative opinion about the same feature, the risk of facing the same problem becomes
higher. These kinds of problems must appear in the reputation model in order to re-
flect a true evaluation for the product in use, and to draw user’s attention so that he
can look for more details and have a rational decision about buying the product.
Therefore, we suggest giving these types of negative opinions more weight to draw
the user’s attention to problems in the products. If we have some negative opinions
about different sub-features f j for the feature f i , we do not consider them as frequent
for the feature f i . For example, if we have negative opinions about a mobile phone
camera as follows “Low video recording quality”, “The flash light give a very harsh
light”, and “No zoom available”, these negative opinions about the camera cannot be
considered frequent in terms of “camera” because they are about different sub-
features (video recording, flash light, and zoom) of the generic feature “camera”.
Equation (1) is used to calculate the negative weights for each feature f j .
R neg
j
3 i 5 ! 41 0
N j # - 11 sijneg 5 4 1.. (1)
i #1 2 ! /
N j : is the weight for negative opinions of feature f j .
R neg
j : is the number of reviews that contains negative opinions about the feature
fj.
20
sijneg : is the strength of negative opinion in review (i) about the feature (j).
! : is a positive integer that is used to define the interval of weight increment for
the subsequent opinions, where
1 (2)
Interval #
!
The value of ! is subject to change, higher ! values will furnish higher feature
reputation values, and that is because the Interval value in (2) will be less, which
indicates fewer increments in weights for frequent negative opinions. We use"6 ! # 3
7; which indicates that the weights for frequent opinions will match with the series in
(3), as it appears in the series we keep the value of the opinion strength si intact, and
we add Interval to increase the weight.
R neg
j 41
Nj # s neg
1j 5 (s neg
2j 5 0.33) 5 ... 5 ( s R neg j 5
neg
) (3)
j 3
For a feature f i which has sub features {f1, f2, …, fk}, Equation (4) is proposed to
calculate the overall weight for negative opinions about the generic feature f i , which
is the sum of the weights of all its sub-features calculated using Equation (1).
3 k 0
WN i # 11 - N j .. 5 N i (4)
2 j #1 /
WN i : is the weight of all negative opinions about a generic feature f i in the hie-
rarchy FH.
k : is the number of sub-features of feature f i .
N i : represents the weight of negative opinions about the generic feature f i itself
and not about one of its sub-features. It is calculated using Equation (1).
Positive Opinion Weight
For the positive opinions, we propose to calculate the positive weight for a feature
f j by adding opinion strength values si given in Equation (5). If the feature has sub
features {f1, f2, …, fk}, the overall weight for positive opinions about the generic fea-
ture f i , is the sum of the positive weights of all its sub-features plus the positive
weight of itself, as showed in Equation (6) below:
21
R jpos
Pj # - sijpos (5)
i #1
3 k 0
WPi # 11 - Pj .. 5 Pi (6)
2 j #1 /
Pj : is the weight for positive opinions of feature f j .
WPi : is the weight of all positive opinions about a generic feature f i in the hie-
rarchy FH.
Pi : represents the weight of positive opinions about the generic feature f i itself and
not about one of its sub-features. It is calculated using Equation (5).
Calculating Feature Reputation
In this paper, we propose to calculate the reputation of a feature based on its over-
all positive and negative weights as showed in Equation (7), which represents the
percentage of positive opinion weights to the total weights of both positive and nega-
tive opinions.
WPi (7)
FREPi # 8 100
WPi 5 WN i
An example is given in Table 1 to demonstrate the proposed method. In the table,
for simplicity, each feature listed on the left most column has three sub features;
NOf1, NOf2, and NOf3 are the number of reviews which contain negative opinions to
the corresponding sub features; N1, N2, and N3 are the negative weight of correspond-
ing sub features; NOFi and WNOFi are the number of reviews containing negative
opinions about feature Fi and its negative weight respectively. It also shows the total
number of positive reviews (PO), the total number of negative reviews (NO), overall
weight for positive (WPi) and negative (WNi) opinions, and the aggregation (FREPi)
using the proposed method and the (PPR) which is the percentage of positive reviews
among all reviews without considering the strength of opinion and it can be calculated
using Equation (7), where (WPi=PO) and (WNi=NO). (Note: the strength of each
opinion was not provided in the table).
The example shows the detailed calculations for both positive and negative opi-
nions weights. In the last two columns we can see the differences between the feature
reputation value using our method (FREPi) and the simple average method (AVG).
Our method results in lower reputation in all cases, this is logical as we give more
weight for negative opinions. For example, the total number of negative opinions
(NO) for both F2, and F7 are the same which is equal to 21. Nevertheless, the overall
weight for negative opinions (N2) for F2 is 63.33 and for F7 is 53.00, which is totally
different. This difference is due to; first, the large frequency for the second sub-
feature (NOf2 # 11) for F2, second, higher values for opinions’ strength (which was
not provided in the table). Fig. 1 shows the relation between (FREPi) and (AVG)
22
where the difference between the two values is the most when the percentage of nega-
tive opinions to positive ones is higher. And this complies with our purpose of giving
negative opinions more weight.
3.3 Feature Impact
Depending on the fact that product features are not equally important to customers,
we will calculate feature’s impact, which is a value that reflects a feature’s influence
between users. Some of the features are essential for a product to work, but they do
not inspire customers to buy the product, as they become consistent over time. On the
other hand, some hot features, that are improved continually or new features have
high influence on customers to be more interested in the product. Thus, these features
should have more impact on the product overall reputation. Features impact will be
used to give different weights for every feature in the final product reputation aggre-
gation formula. We suggest that features that frequently occurred in users’ reviews
have more impact than other features. Let Mj denote the number of reviews that have
opinion about this feature, whether positive or negative, the impact of a feature f j ,
denoted as Ij, is defined in Equation (8) below:
M j # R neg
j 5 R jpos
Mj
Ij # (8)
M Max
MMax: is the largest value of Mj for all features.
All feature impacts will be given values between 0 and 1; 1 for the feature that was
mentioned the most in the users’ reviews, and thus has the most influence on users.
Table 1. An example showing the calculation of feature reputation
Features NOf1 N1 NOf2 N2 NOf3 N3 NOFi WNOFi PO WPi NO WNi FREPi PPR
F1 2 4.33 4 9 1 3 5 15.33 110 266 12 37.67 87.60 90.16
F2 3 6 11 37.33 3 9 4 11 87 170 21 63.33 72.86 80.56
F3 10 39.00 9 26 7 22 0 0 215 425 26 87.00 83.01 89.21
F4 7 19 6 15 3 7 2 4.33 366 722 18 45.33 94.09 95.31
F5 13 49 8 25.33 2 3.33 1 1 145 283 24 78.67 78.25 85.80
F6 9 30 11 38.33 5 14.33 17 78.33 417 835 42 161.00 83.84 90.85
F7 8 20.33 5 14.33 3 5 5 13.33 329 655 21 53.00 92.51 94.00
F8 12 47 2 6.33 3 6 0 0 273 563 17 59.33 90.47 94.14
23
100.00
80.00
60.00
FREP
40.00
PPR
20.00
0.00
F1 F2 F3 F4 F5 F6 F7 F8
Fig. 1. Feature reputation diagram for the proposed method and the simple average method
3.4 Product Reputation
Many opinions in customers’ reviews targeted the product itself rather than men-
tioning a specific feature in the product, these opinions are also considered in our
model. We propose to calculate the product reputation by integrating the reputation
calculated based on the reviews which are directly about the product and the reputa-
tions of the product’s direct features.
Assume that a product has h direct sub features, FREPj and Ij are the reputation and
the impact of each sub feature, respectively. Let GOP denote the product reputation
calculated using Equation (7) where WNi and WPi are the number of negative and
positive opinions about the product itself in the reviews respectively, and GOP have
the impact of 1. The following equation is proposed to calculate the product’s overall
reputation, where every feature reputation, calculated using Equation (7), is multiplied
by its impact, calculated using Equation (8), in order to give different weights for
features, plus the GOP, and the total is divided by the summation of all features’ im-
pacts plus 1 that represents the GOP impact.
h
- ( FREP 8 I ) 5 GOP
j #1
j j
PR # h
(9)
- I 51
j #1
j
Table 2 shows the results of calculating the overall product reputation using our
model, and the simple average technique. It shows the values of (FREPi) and (PPR),
from Table 1, the (Mj) column indicates how many times this feature and its sub-
features have been explicitly mentioned in the reviews, and (Ij) column is calculated
using (8) where MMax = 459 (the most mentioned feature). It also shows the results of
the product reputation (PR) and the regular (AVG).
24
Table 2. Example Reputation Calculation
Features FREPj PPR Mj Ij FREPi * Ij
F1 87.60 90.16 122 0.27 22.90
F2 72.86 80.56 108 0.24 15.71
F3 83.01 89.21 241 0.53 41.05
F4 94.09 95.31 384 0.84 77.06
F5 78.25 85.80 169 0.37 26.09
F6 83.84 90.85 459 1.00 77.51
F7 92.51 94.00 350 0.76 68.36
F8 90.47 94.14 290 0.63 55.05
GOP 86.31 86.31 528 1.00 86.31
Total - - - 5.63 470.03
AVG - 89.59 - - -
PR 87.00 - - - -
As we mentioned before, our model reveals a final reputation lower than the aver-
age method. One of the strength points in our model is data representation, as we are
able to provide details for customers about every specific feature. Fig. 2 shows the
reputation of every feature, which can be more inspiring for customers than the one
value reputation representation. Furthermore, more detailed information can also be
provided as showed in the example in Fig. 3. For example, if the user is interested in a
specific feature and he wants to see more about it, a second level will show the details
of negative opinions about sub-features and the frequency of each one.
100.00
80.00
60.00
FREP
40.00
PPR
20.00
0.00
F1 F2 F3 F4 F5 F6 F7 F8 GOP PR AVG
Fig. 2. Results of product reputation model including all features and the regular average result
25
Camera
12
8
4
Negative Opinions
0
Resolution Zoom Flash Light Video
Recording
Sub Features
Fig. 3. Example of negative opinions of features at the second level
4 Conclusion
In this paper we have presented a new reputation model for products, our model
used text reviews rather than users’ ratings. We extracted opinions about hierarchy of
features and calculated the frequencies for positive and negative opinions assuming
that frequent negative opinions about features and sub-features should get more
weight in the reputation calculation, as they indicate a problem in a product a custom-
er may face if they buy it. In Addition, we calculated the impact of features, hence
certain features in some products are more inspiring for users, and therefore they are
more important in the reputation model. Our model integrates the strength of opinions
and provides summary about users’ opinions for customers rather than representing
reputation as a number of stars. For future work, the reputation model may be mod-
ified to consider age and validity of reviews, and also detect malicious users’ reviews
which aim to sabotage the reputation of a product.
References
1. F. Garcin, B. Faltings, and R. Jurca, "Aggregating reputation feedback," in International
Conference on Reputation ICORE, 2009, pp. 119-128.
2. P. Resnick, K. Kuwabara, R. Zeckhauser, and E. Friedman, "Reputation systems," Com-
munications of the ACM (CACM), vol. 43, pp. 45-48, 2000.
3. J. Cho, K. Kwon, and Y. Park, "Q-rater: A collaborative reputation system based on source
credibility theory," Expert Systems with Applications, vol. 36, pp. 3751-3760, 2009.
4. S. Leberknight, S. Sen, and M. Chiang, "On the Volatility of Online Ratings: An Empirical
Study " in E-Life: Web-Enabled Convergence of Commerce, Work, and Social Life. vol.
108, ed: Springer Berlin Heidelberg, 2012, pp. 77-86.
5. B. Pang and L. Lee, "Opinion mining and sentiment analysis," Foundations and Trends in
Information Retrieval, vol. 2, pp. 1-135, 2008.
26
6. M. Hu and B. Liu, "Mining and summarizing customer reviews," in ACM International
Conference on Knowledge Discovery and Data Mining (SIGKDD), 2004, pp. 168-17
7. J. Yi and W. Niblack, "Sentiment Mining in WebFountain," in IEEE International Confe-
rence on Data Engineering (ICDE), 2005, pp. 1073-1083.
8. P. D. Turney, "Thumbs up or thumbs down?: Semantic orientation applied to unsupervised
classification of reviews," in Association for Computational Linguistics Conference (ACL),
2002, pp. 417-424.
9. B. Pang, L. Lee, and S. Vaithyanathan, "Thumbs up?: sentiment classification using ma-
chine learning techniques," in Conference on Empirical Methods in Natural Language
Processing (EMNLP), 2002, pp. 79-86.
10. J. Kamps, M. Marx, R. J. Mokken, and M. De Rijke, "Using wordnet to measure semantic
orientations of adjectives," in International Conference on Language Resources and Eval-
uation (LREC) 2004, pp. 1115-1118.
11. A. M. Popescu and O. Etzioni, "Extracting product features and opinions from reviews," in
Conference on Human Language Technology and Empirical Methods in Natural Lan-
guage Processing (HLT/EMNLP), 2005, pp. 339-346.
12. S. Morinaga, K. Yamanishi, K. Tateishi, and T. Fukushima, "Mining product reputations
on the Web," in ACM International Conference on Knowledge Discovery and Data Mining
(SIGKDD), 2002, pp. 341-349.
13. T. Hashimoto and Y. Shirota, "Semantics extraction from social computing: a framework
of reputation analysis on buzz marketing sites," in International Conference on Databases
in Networked Information Systems (DNIS), 2010, pp. 244-255.
14. S. Baccianella, A. Esuli, and F. Sebastiani, "Sentiwordnet 3.0: An enhanced lexical re-
source for sentiment analysis and opinion mining," in International Conference on Lan-
guage Resources and Evaluation (LREC), 2010, pp. 2200-2204.
27