Presentation is loading. Please wait.

Presentation is loading. Please wait.

How Opinions are Received by Online Communities A Case Study on Amazon.com Helpfulness Votes Cristian Danescu-Niculescu-Mizil 1, Gueorgi Kossinets 2, Jon.

Similar presentations


Presentation on theme: "How Opinions are Received by Online Communities A Case Study on Amazon.com Helpfulness Votes Cristian Danescu-Niculescu-Mizil 1, Gueorgi Kossinets 2, Jon."— Presentation transcript:

1 How Opinions are Received by Online Communities A Case Study on Amazon.com Helpfulness Votes Cristian Danescu-Niculescu-Mizil 1, Gueorgi Kossinets 2, Jon Kleinberg 1, Lillian Lee 1 1 Dept. of Computer Science, Cornell University, 2 Google Inc. WWW 2009 2009. 07. 30. IDS Lab. Hwang Inbeom

2 Copyright  2009 by CEBT Outline  Users’ evaluation on online reviews: Helpfulness votes Observation of behaviors Making some hypothesis and proving their validity Coming up with a mathematical model explains these behaviors 2

3 Copyright  2009 by CEBT Introduction Opinion What did Y think of X? 3

4 Copyright  2009 by CEBT Introduction Meta-Opinion What did Z think of Y’s opinion of X? 4

5 Copyright  2009 by CEBT The Helpfulness of Reviews  Widely-used web sites include not just reviews, but also evaluations of the helpfulness of the reviews The helpfulness vote – “Was this review helpful to you?” Helpfulness ratio: – “a out of b people found the review itself helpful” 5

6 Copyright  2009 by CEBT Amazon.com Helpfulness Votes Data  4,000,000 reviews about roughly 700,000 books, including average star ratings and helpfulness ratios 6 Average star rating Helpfulness ratio

7 Copyright  2009 by CEBT Definitions of “Helpfulness”  Helpfulness in the narrow sense: “Does this review help you in making a purchase decision?” Liu’s work: annotation and classification of review helpfulness Annotators’ evaluation differed significantly from the helpfulness votes  Helpfulness “in the wild” The way Amazon users evaluate each others’ reviews Intertwined with complex social feedback mechanisms 7

8 Copyright  2009 by CEBT Flow of Presentation HypothesizingVerifyingModeling 8

9 Copyright  2009 by CEBT Flow of Presentation Hypothesizing Conformity Individual-bias Brilliant-but-cruel Quality-only VerifyingModeling 9

10 Copyright  2009 by CEBT Hypotheses: Social Mechanisms underlying  Well-studied hypotheses for how social effects influence group’s reaction to an opinion The conformity hypothesis The individual-bias hypothesis The brilliant-but-cruel hypothesis The quality-only straw-man hypothesis 10

11 Copyright  2009 by CEBT Hypotheses  The conformity hypothesis Review is evaluated as more helpful when its star rating is closer to the consensus star rating – Helpfulness ratio will be the highest of which reviews have star rating equal to overall average  The individual-bias hypothesis When a user considers a review, he or she will rate it more highly if it expresses an opinion that he or she agrees with 11

12 Copyright  2009 by CEBT Hypotheses (contd.)  The brilliant-but-cruel hypothesis Negative reviewers are perceived as more intelligent, competent, and expert than positive reviewers  The Quality-only straw-man hypothesis Helpfulness is being evaluated purely based on the textual content of reviews Non-textual factors are simply correlates of textual quality 12

13 Copyright  2009 by CEBT Flow of Presentation Hypothesizing Verifying Absolute deviation of helpfulness ratio Signed deviation of helpfulness ratio Variance of star rating and helpfulness ratio Making use of plagiarism Modeling 13

14 Copyright  2009 by CEBT Hypotheses Conformity A review is evaluated as more helpful when its star rating is closer to the average star rating Individual-bias A review is evaluated as more helpful when its star rating is closer to evaluator’s opinion Brilliant-but-cruel A review is evaluated as more helpful when its star rating is below to the average star rating Quality-only Only textual information affects helpfulness evaluation 14

15 Copyright  2009 by CEBT Absolute Deviation from Average  Consistent with conformity hypothesis Strong inverse correlation between the median helpfulness ratio and the absolute deviation Reviews with star rating close to the average gets higher helpfulness ratio 15

16 Copyright  2009 by CEBT Hypotheses Conformity A review is evaluated as more helpful when its star rating is closer to the average star rating Individual-bias A review is evaluated as more helpful when its star rating is closer to evaluator’s opinion Brilliant-but-cruel A review is evaluated as more helpful when its star rating is below to the average star rating Quality-only Only textual information affects helpfulness evaluation 16

17 Copyright  2009 by CEBT Signed Deviation from Average  Not consistent with brilliant-but-cruel hypothesis There is tendency towards positivity Black lines should not be sloped that way if it is valid hypothesis 17

18 Copyright  2009 by CEBT Hypotheses Conformity A review is evaluated as more helpful when its star rating is closer to the average star rating Individual-bias A review is evaluated as more helpful when its star rating is closer to evaluator’s opinion Brilliant-but-cruel A review is evaluated as more helpful when its star rating is below to the average star rating Quality-only Only textual information affects helpfulness evaluation 18

19 Copyright  2009 by CEBT Addressing Individual-bias Effects  It is hard to distinguish between the conformity and the individual-bias hypothesis  We need to examine cases in which individual people’s opinions do not come from exactly the same distribution Cases in which there is high variance in star ratings Otherwise conformity and individual-bias are indistinguishable – Everyone has same opinion 19

20 Copyright  2009 by CEBT Variance of Star Rating and Helpfulness Ratio 20 Helpfulness ratio is the highest with reviews of which rating is slightly- above the average Two-humped camel plots: local minimum around average Helpfulness ratio is the highest when star ratings of reviews have average value

21 Copyright  2009 by CEBT Hypotheses Conformity A review is evaluated as more helpful when its star rating is closer to the average star rating Individual-bias A review is evaluated as more helpful when its star rating is closer to evaluator’s opinion Brilliant-but-cruel A review is evaluated as more helpful when its star rating is below to the average star rating Quality-only Only textual information affects helpfulness evaluation 21

22 Copyright  2009 by CEBT Plagiarism  Making use of plagiarism is effective way to control for the effect of review text  Definition of plagiarized pair(s) of reviews Two or more reviews of different products With near-complete textual overlap 22

23 Copyright  2009 by CEBT An Example Skull-splitting headache guaranteed! If you enjoy thumping, skull splitting migraine headache, then Sing N Learn is for you. As a longtime language instructor, I agree with the attempt and effort that this series makes, but it is the execution that ultimately weakens Sing N Learn Chinese. To be sure, there are much, much better ways to learn Chinese. In fact, I would recommend this title only as a last resort and after you’ve thoroughly exhausted traditional ways to learn Chinese … Migraine Headache at No Extra Charge If you enjoy a thumping, skull splitting migraine headache, then the Sing N Learn series is for you. As a longtime language instructor, I agree with the effort that this series makes, but it is the execution that ultimately weakens Sing N Learn series. To be sure, there are much, much better ways to learn a foreign language. In fact, I would recommend this title only as a last resort and after you’ve thoroughly exhausted traditional ways to learn Korean … 23

24 Copyright  2009 by CEBT Plagiarism (contd.)  Plagiarized reviews Almost(not exact) same text – More possibly, same text could be considered as spam reviews Different non-textual information  If the quality-only straw man hypothesis holds, helpfulness ratios of documents in each pair should be the same  Possible other methods Human annotation – Could be subjective Classification using machine learning methods – We cannot guarantee the accuracies of algorithms 24

25 Copyright  2009 by CEBT Experiments with Plagiarism  Text quality is not the only explanatory factor Statistically significant difference between the helpfulness ratios of plagiarized pairs 25 The plagiarized reviews with deviation 1 is significantly more helpful than those with deviation 1.5

26 Copyright  2009 by CEBT Hypotheses Conformity A review is evaluated as more helpful when its star rating is closer to the average star rating Individual-bias A review is evaluated as more helpful when its star rating is closer to evaluator’s opinion Brilliant-but-cruel A review is evaluated as more helpful when its star rating is below to the average star rating Quality-only Only textual information affects helpfulness evaluation 26

27 Copyright  2009 by CEBT Flow of Presentation HypothesizingVerifying Modeling Based on individual bias and mixtures of distributions 27

28 Copyright  2009 by CEBT Authors’ Model  Based on individual bias and mixtures of distributions  Two distributions: one for positive, one for negative evaluators Balance between positive and negative evaluators: Controversy level: – Density function of helpfulness ratios of positive evaluators – Gaussian distribution of which average is -centered – Density function of helpfulness ratios of negative evaluators – Gaussian distribution of which average is -centered 28

29 Copyright  2009 by CEBT Validity of the Model  Empirical observation and model generated 29

30 Copyright  2009 by CEBT Conclusion  A review’s perceived helpfulness depends not just on its content, but also the relation of its score to other scores  The dependence of the score is consistent with a simple and natural model of individual-bias in the presence of a mixture of opinion distributions  Directions for further research Variations in the effect can be used to form hypotheses about differences in the collective behaviors of the underlying populations It would be interesting to consider social feedback mechanisms that might be capable of modifying the effects authors observed here Considering possible outcomes of design problem for systems enabling the expression and dissemination of opinions 30

31 Copyright  2009 by CEBT Discussions  So, how can we use this? In which cases would this information be helpful? Available information is very limited – Star ratings – Helpfulness ratios  Conclusion is rather trivial Does not present new discoveries 31


Download ppt "How Opinions are Received by Online Communities A Case Study on Amazon.com Helpfulness Votes Cristian Danescu-Niculescu-Mizil 1, Gueorgi Kossinets 2, Jon."

Similar presentations


Ads by Google