Skip to main content

Discovering Influence of Yelp Reviews Using Hawkes Point Processes

  • 701 Accesses

Part of the Lecture Notes in Networks and Systems book series (LNNS,volume 296)

Abstract

With the development of technology, social media and online forums are becoming popular platforms for people to share opinions and information. A major question is how much influence these have on other users’ behavior. In this paper, we focused on Yelp, an online platform for customers to share information of their visiting experiences on restaurants, to explore the possible relationships between past reviews and future reviews of a restaurant through multiple aspects such as star-ratings, user features and sentiment features. By using the lasso regression model with review features processed through Hawkes Process Model and B-Spline basis functions as the modeling of restaurant basic performance, average star-ratings, low star-ratings and sentiment features of past reviews have been found to have significant influence on future reviews. Due to the limited dataset, we performed simulation on restaurants’ reviews using Multinomial Logistic Regression and re-built the model. A verification process has been performed eventually using Logistical Regression. The simulation and the verification results have been found to support the prior findings which indicate that influence between past and future reviews does exist, and can be revealed on multiple aspects.

Keywords

  • Yelp reviews
  • Hawkes Point Process
  • Data mining
  • Sentiment features

This is a preview of subscription content, access via your institution.

Buying options

Chapter
USD   29.95
Price excludes VAT (USA)
  • DOI: 10.1007/978-3-030-82199-9_7
  • Chapter length: 21 pages
  • Instant PDF download
  • Readable on all devices
  • Own it forever
  • Exclusive offer for individuals only
  • Tax calculation will be finalised during checkout
eBook
USD   219.00
Price excludes VAT (USA)
  • ISBN: 978-3-030-82199-9
  • Instant PDF download
  • Readable on all devices
  • Own it forever
  • Exclusive offer for individuals only
  • Tax calculation will be finalised during checkout
Softcover Book
USD   279.99
Price excludes VAT (USA)
Fig. 1.
Fig. 2.
Fig. 3.

References

  1. Pentina, I., Bailey, A.A., Zhang, L.: Exploring effects of source similarity, message valence, and receiver regulatory focus on yelp review persuasiveness and purchase intentions. J. Mark. Commun. 24(2), 125–145 (2018)

    CrossRef  Google Scholar 

  2. Parikh, A., Behnke, C., Vorvoreanu, M., Almanza, B., Nelson, D.: Motives for reading and articulating user-generated restaurant reviews on Yelp. com. J. Hosp. Tour. Technol. (2014)

    Google Scholar 

  3. Luca, M.: Reviews, reputation, and revenue: The case of Yelp. com. Com (March 15, 2016). Harvard Business School NOM Unit Working Paper (12-016) (2016)

    Google Scholar 

  4. Vinson, D.W., Dale, R., Jones, M.N.: Decision contamination in the wild: sequential dependencies in online review ratings. Behav. Res. Methods 51(4), 1477–1484 (2018). https://doi.org/10.3758/s13428-018-1175-8

    CrossRef  Google Scholar 

  5. Hu, N., Liu, L., Zhang, J.J.: Do online reviews affect product sales? the role of reviewer characteristics and temporal effects. Inf. Technol. Manag. 9(3), 201–214 (2008)

    CrossRef  Google Scholar 

  6. Asghar, N.: Yelp dataset challenge: review rating prediction. arXiv preprint arXiv:1605.05362 (2016)

  7. Xu, Y., Wu, X., Wang, Q.: Sentiment Analysis of Yelp’s Ratings Based on Text Reviews (2014)

    Google Scholar 

  8. Kc, S., Mukherjee, A.: On the temporal dynamics of opinion spamming: case studies on yelp. In: Proceedings of the 25th International Conference on World Wide Web, pp. 369-379 (2016)

    Google Scholar 

  9. Rahimi, S., Andris, C., Liu, X.: Using yelp to find romance in the city: a case of restaurants in four cities. In: Proceedings of the 3rd ACM SIGSPATIAL Workshop on Smart Cities and Urban Analytics, pp. 1-8 (2017)

    Google Scholar 

  10. Hawkes, A.G.: Spectra of some self-exciting and mutually exciting point processes. Biometrika 58(1), 83–90 (1971)

    MathSciNet  CrossRef  Google Scholar 

  11. Freed, A.M.: Earthquake triggering by static, dynamic, and postseismic stress transfer. Ann. Rev. Earth Planet. Sci. 33, 335–367 (2005)

    CrossRef  Google Scholar 

  12. Mishra, S., Rizoiu, M. A., Xie, L.: Feature driven and point process approaches for popularity prediction. In: Proceedings of the 25th ACM International on Conference on Information and Knowledge Management, pp. 1069-1078 (2016)

    Google Scholar 

  13. Pinto, J.C.L., Chahed, T., Altman, E.: Trend detection in social networks using Hawkes processes. In: Proceedings of the 2015 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining 2015, pp. 1441–1448 (2015)

    Google Scholar 

  14. Wang, Y., Du, N., Trivedi, R., Song, L.: Coevolutionary latent feature processes for continuous-time user-item interactions (2016)

    Google Scholar 

  15. Porter, M.: Multivariate hawkes point process models for social systems. In: Proceedings of the 62nd World Statistics Congress of the International Statistical Institute (2017)

    Google Scholar 

  16. Yelp open dataset. https://www.yelp.com/dataset/, Accessed 20 Feb 2020

  17. Schoenberg, I.J.: Contributions to the problem of approximation of equidistant data by analytic functions. In: IJ Schoenberg Selected Papers, Birkhäuser, Boston, MA, pp. 3–57 (1988)

    Google Scholar 

  18. TextBlob: Simplified Text Preprocessing. https://textblob.readthedocs.io/en/dev/, Accessed 23 Jan 2021

Download references

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Yichen Jiang .

Editor information

Editors and Affiliations

8 Appendix

8 Appendix

1.1 8.1 B-Spline Basis Function

The basic framework of B-Spline curve has been created by Schoenberg on 1946 [17], and has been developed to adjust different application such as modeling of 3-D geometry shape or interpolation of fluctuating data points for smoothing purpose. A k-order B-Spline curve is composed by a set of linear-combined control points \(P_i\) and B-Spline basis functions denoted as \(N_{i,k}(t)\), and each control point is associated with a basis function in a recurrence relation such that:

$$\begin{aligned} N_{i,k}(t) = N_{i,k-1}(t)\frac{t-t_i}{t_{i+k-1}-t_i}+N_{i+1,k-1}(t)\frac{t_{i+k}-t}{t_{i+k}-t_{i+1}} \end{aligned}$$
(14)
$$N_{i,1} = {\left\{ \begin{array}{ll} 1&{} if \quad t_i \le t \le t_{i+1}\\ 0&{} otherwise \end{array}\right. }$$

The shape of B-Spline basis function is determined by the knot vector:

$$\begin{aligned} T = (t_0, t_1,...,t_{k-1},t_k,t_{k+1},...,t_{n-1}, t_n, t_{n+1},..., t_{n+k}) \end{aligned}$$
(15)

The number of elements of the knot vector is defined by the sum of number of control points and the order of the B-Spline curve (n+k+1).

Rights and permissions

Reprints and Permissions

Copyright information

© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG

About this paper

Verify currency and authenticity via CrossMark

Cite this paper

Jiang, Y., Porter, M. (2022). Discovering Influence of Yelp Reviews Using Hawkes Point Processes. In: Arai, K. (eds) Intelligent Systems and Applications. IntelliSys 2021. Lecture Notes in Networks and Systems, vol 296. Springer, Cham. https://doi.org/10.1007/978-3-030-82199-9_7

Download citation