# Sales Prediction by Integrating the Heat and Sentiments of Product Dimensions

^{1}

^{2}

^{*}

## Abstract

**:**

## 1. Introduction

## 2. Literature Review

#### 2.1. eWOM’s Effect on Sales

#### 2.2. eWOM-Based and GSI-Based Sales Prediction

## 3. Materials and Methods

#### 3.1. Research Framework

**H1**.

#### 3.2. Data and Variables

#### 3.2.1. Data Collection

#### 3.2.2. Dynamic Topic Analysis

^{th}dimension of the $i$th movie on day $t$. ${\vartheta}_{i,t}$ can be calculated as follows:

^{th}dimension word at the $i$

^{th}time (location) in document d for one movie. The sentiment of the $k$

^{th}dimension for one movie on the $t$

^{th}day can then be formulated as follows:

^{th}dimension. Figure 5 shows the average sentiments of the dimension plot for 122 movies. Using Figure 4 and Figure 5, we can easily monitor consumer feedbacks (heat and sentiments) on product dimensions over time.

#### 3.3. Predictive Model

#### 3.3.1. Autoregressive Model

#### 3.3.2. ARO Model

^{th}online information variable on day $t$. We determined $p\mathrm{and}q$ by comparing model accuracy when using different values of $p\mathrm{and}q$. ${\phi}_{i}$ and ${\rho}_{i,j}$ are parameters that need to be estimated. The parameter q specifies the lags of the preceding days of the online information variables; J indicates the number of these variables. The ARO model uses preceding sales, Google Trends, the eWOM variables and other predictors in Table 4 to predict current and future sales.

#### 3.3.3. The ARHS Model

## 4. Results

#### 4.1. Performance of the Parameters in the ARHS Model

#### 4.2. Comparison of the Predictive Models

#### 4.3. Robustness of the Predictive Power of the Heat and Sentiments of Dimensions

## 5. Conclusion and Discussion

## Author Contributions

## Funding

## Conflicts of Interest

## References

- Wang, X.; Yu, C.; Wei, Y. Social media peer communication and impacts on purchase intentions: A consumer socialization framework. J. Interact. Mark.
**2012**, 26, 198–208. [Google Scholar] [CrossRef] - Felix, R.; Rauschnabel, P.A.; Hinsch, C. Elements of strategic social media marketing: A holistic framework. J. Bus. Res.
**2017**, 70, 118–126. [Google Scholar] [CrossRef] - Amornpetchkul, T.; Duenyas, I.; Şahin, Ö. Mechanisms to induce buyer forecasting: Do suppliers always benefit from better forecasting? Prod. Oper. Manag.
**2015**, 24, 1724–1749. [Google Scholar] [CrossRef] - Hu, N.; Koh, N.S.; Reddy, S.K. Ratings lead you to the product, reviews help you clinch it? The mediating role of online review sentiments on product sales. Decis. Support Syst.
**2014**, 57, 42–53. [Google Scholar] [CrossRef] - Geva, T.; Oestreicher-Singer, G.; Efron, N.; Shimshoni, Y. Using forum and search data for sales prediction of high-involvement products. MIS Q.
**2017**, 41, 65–82. [Google Scholar] [CrossRef] - Chern, C.C.; Wei, C.P.; Shen, F.Y.; Fan, Y.N. A sales forecasting model for consumer products based on the influence of online word-of-mouth. Inf. Syst. e-Bus. Manag.
**2015**, 13, 445–473. [Google Scholar] [CrossRef] - Ghiassi, M.; Lio, D.; Moon, B. Pre-production forecasting of movie revenues with a dynamic artificial neural network. Expert Syst. Appl.
**2015**, 42, 3176–3193. [Google Scholar] [CrossRef] - See-To, E.W.K.; Ngai, E.W.T. Customer reviews for demand distribution and sales nowcasting: A big data approach. Ann. Oper. Res.
**2018**, 270, 415–431. [Google Scholar] [CrossRef] - Fishbein, M. An investigation of the relationships between beliefs about an object and the attitude toward that object. Hum. Relations
**1963**, 16, 233–239. [Google Scholar] [CrossRef] - Rosenberg, M.J. Cognitive structure and attitudinal affect. J. Abnorm. Psychol.
**1956**, 53, 367–372. [Google Scholar] [CrossRef] - Li, X.; Wu, C.; Mai, F. The effect of online reviews on product sales: A joint sentiment-topic analysis. Inf. Manag.
**2018**. [Google Scholar] [CrossRef] - Liang, T.P.; Li, X.; Yang, C.T.; Wang, M. What in consumer reviews affects the sales of mobile apps: A multifacet sentiment analysis approach. Int. J. Electron. Commer.
**2015**, 20, 236–260. [Google Scholar] [CrossRef] - Siering, M.; Muntermann, J.; Rajagopalan, B. Explaining and predicting online review helpfulness: The role of content and reviewer-related signals. Decis. Support Syst.
**2018**, 108, 1–12. [Google Scholar] [CrossRef] - Yang, D.; Zhang, L.; Wu, Y.; Guo, S.; Zhang, H.; Xiao, L. A sustainability analysis on retailer’s sales effort in a closed-loop supply chain. Sustainability
**2019**, 11, 18. [Google Scholar] [CrossRef] - Floyd, K.; Freling, R.; Alhoqail, S.; Cho, H.Y.; Freling, T. How online product reviews affect retail sales: A meta-analysis. J. Retail.
**2014**, 90, 217–232. [Google Scholar] [CrossRef] - Ghose, A.; Ipeirotis, P.G.; Li, B. Designing ranking systems for hotels on travel search engines by mining user-generated and crowd-sourced content. Mark. Sci.
**2012**, 31, 493–520. [Google Scholar] [CrossRef] - Wang, F.; Liu, X.; Fang, E. User reviews variance, critic reviews variance, and product sales: An exploration of customer breadth and depth effects. J. Retail.
**2015**, 91, 372–389. [Google Scholar] [CrossRef] - Chen, P.-Y.; Dhanasobhon, S.; Smith, M.D. All reviews are not created equal: The disaggregate impact of reviews and reviewers at Amazon.com. 2006. Available online: http://archive.nyu.edu/bitstream/2451/14961/2/USEDBOOK19.pdf (accessed on 20 July 2006).
- Baker, A.M.; Donthu, N.; Kumar, V. Investigating how word-of-mouth conversations about brands influence purchase and retransmission intentions. J. Mark. Res.
**2016**, 53, 225–239. [Google Scholar] [CrossRef] - Gu, B.; Tang, Q.; Whinston, A.B. The influence of online word-of-mouth on long tail formation. Decis. Support Syst.
**2013**, 56, 474–481. [Google Scholar] [CrossRef] - Kostyra, D.S.; Reiner, J.; Natter, M.; Klapper, D. Decomposing the effects of online customer reviews on brand, price, and product attributes. Int. J. Res. Mark.
**2016**, 33, 11–26. [Google Scholar] [CrossRef] - Hafeez, H.A.; Manzoor, A.; Salman, F. Impact of social networking sites on consumer purchase intention: An analysis of restaurants in Karachi. J. Bus. Strateg.
**2017**, 11, 1–20. [Google Scholar] [CrossRef] - Yan, Q.; Wu, S.; Wang, L.; Wu, P.; Chen, H.; Wei, G. E-WOM from e-commerce websites and social media: Which will consumers adopt? Electron. Commer. Res. Appl.
**2016**, 17, 62–73. [Google Scholar] [CrossRef] - Parboteeah, V.; Valacich, J.; Wells, J. Online impulse buying: understanding the interplay between consumer impulsiveness and website quality. J. Assoc. Inf. Syst.
**2011**, 12, 32. [Google Scholar] - Dewan, S.; Ramaprasad, J. Social media, traditional media, and music sales. MIS Q.
**2014**, 38, 101–121. [Google Scholar] [CrossRef] - Allan J., K.; Philip J., K. WOM and social media: Presaging future directions for research and practice. J. Mark. Commun.
**2014**, 20, 5–20. [Google Scholar] - Cui, G.; Lui, H.; Guo, X. The effect of online consumer reviews on new product sales. Int. J. Electron. Commer.
**2012**, 17, 39–57. [Google Scholar] [CrossRef] - Chevalier, J.A.; Mayzlin, D. The effect of word of mouth on sales: Online book reviews. J. Mark. Res.
**2006**, 43, 345–354. [Google Scholar] [CrossRef] - Duan, W.; Gu, B.; Whinston, A.B. Do online reviews matter?—An empirical investigation of panel data. Decis. Support Syst.
**2008**, 45, 1007–1016. [Google Scholar] [CrossRef] - Clemons, E.K.; Gao, G.G.; Hitt, L.M. When online reviews meet the hyperdifferentiation: A study of craft beer industry. J. Manag. Inf. Syst.
**2006**, 23, 149–171. [Google Scholar] [CrossRef] - Chintagunta, P.K.; Gopinath, S.; Venkataraman, S. The effects of online user reviews on movie box-office performance: Accounting for sequential rollout and aggregation across local markets. Mark. Sci.
**2010**, 29, 944–957. [Google Scholar] [CrossRef] - Xu, H. Is More Information Better? An Economic Analysis of Group-Buying Platforms. J. Assoc. Inf. Syst.
**2018**, 19, 1130–1144. [Google Scholar] [CrossRef] - Hansen, F. Consumer choice behavior: An experimental approach. J. Mark. Res.
**1969**, 6, 436–443. [Google Scholar] [CrossRef] - Relling, M.; Schnittka, O.; Sattler, H.; Johnen, M. Each can help or hurt: Negative and positive word of mouth in social network brand communities. Int. J. Res. Mark.
**2016**, 33, 42–58. [Google Scholar] [CrossRef] - Sweeney, J.C.; Soutar, G.N.; Mazzarol, T. Word of mouth: Measuring the power of individual messages. Eur. J. Mark.
**2012**, 46, 237–257. [Google Scholar] [CrossRef] - Lin, Z.; Wang, Q. E-commerce product networks, word-of-mouth convergence, and product sales. J. Assoc. Inf. Syst.
**2018**, 18, 848–871. [Google Scholar] [CrossRef] - Amblee, N.; Bui, T. Harnessing the influence of social proof in online shopping: The effect of electronic word of mouth on sales of digital microproducts. Int. J. Electron. Commer.
**2011**, 16, 91–114. [Google Scholar] [CrossRef] - Liu, Y. Word of mouth for movies: Its dynamics and impact on box office revenue. J. Mark.
**2006**, 70, 74–89. [Google Scholar] [CrossRef] - Zhang, Z.; Li, X.; Chen, Y. Deciphering word-of-mouth in social media: Text-based metrics of consumer reviews. ACM Trans. Manag. Inf. Syst.
**2012**, 3, 1–23. [Google Scholar] [CrossRef] - Chen, Y.; Xie, J. Online consumer review: Word-of-mouth as a new element of marketing communication mix. Manag. Sci.
**2008**, 54, 477–491. [Google Scholar] [CrossRef] - Lee, C.; Jung, M. Predicting movie incomes using search engine query data. In Proceedings of the International Conference on Artificial Intelligence and Pattern Recognition, Kuala Lumpur, Malaysia, 17–19 November 2014; pp. 45–49. [Google Scholar]
- Bughin, J. Google searches and twitter mood: nowcasting telecom sales performance. NETNOMICS Econ. Res. Electron. Netw.
**2015**, 16, 87–105. [Google Scholar] [CrossRef] - Wilkie, W.L.; Pessemier, E.A. Issues in marketing’s use of multi-attribute attitude models. J. Mark. Res.
**1973**, 10, 428. [Google Scholar] [CrossRef] - Tirunillai, S.; Tellis, G.J. Mining marketing meaning from online chatter: Strategic brand analysis of big data using latent Dirichlet allocation. J. Mark. Res.
**2014**, 51, 463–479. [Google Scholar] [CrossRef] - Guo, Y.; Barnes, S.J.; Jia, Q. Mining meaning from online ratings and reviews: Tourist satisfaction analysis using latent Dirichlet allocation. Tour. Manag.
**2017**, 59, 467–483. [Google Scholar] [CrossRef] - Cheung, C.M.K.; Thadani, D.R. The impact of electronic word-of-mouth communication: A literature analysis and integrative model. Decis. Support Syst.
**2012**, 54, 461–470. [Google Scholar] [CrossRef] - Duan, W.; Gu, B.; Whinston, A.B. The dynamics of online word-of-mouth and product sales-An empirical investigation of the movie industry. J. Retail.
**2008**, 84, 233–242. [Google Scholar] [CrossRef] - Hur, M.; Kang, P.; Cho, S. Box-office forecasting based on sentiments of movie reviews and independent subspace method. Inf. Sci.
**2016**, 372, 608–624. [Google Scholar] [CrossRef] - Blei, D.M.; Lafferty, J.D. Dynamic topic models. In Proceedings of the 23rd International Conference on Machine Learning, Pittsburgh, PA, USA, 25–29 June 2006; pp. 113–120. [Google Scholar]
- Guerini, M.; Gatti, L.; Turchi, M. Sentiment analysis: How to derive prior polarities from SentiWordNet. In Proceedings of the 2013 Conference on Empirical Methods in Natural Language Processing, Seattle, WA, USA, 18–21 October 2013; pp. 1259–1269. [Google Scholar]
- Socher, R.; Perelygin, A.; Wu, J.; Chuang, J.; Manning, C.; Ng, A.; Christopher, P. Recursive deep models for semantic compositionality over a sentiment treebank. In Proceedings of the 2013 Conference on Empirical Methods in Natural Language Processing, Seattle, WA, USA, 18–21 October 2013; pp. 1631–1642. [Google Scholar]
- Blei, D.M.; Ng, A.Y.; Jordan, M.I. Latent Dirichlet allocation. J. Mach. Learn. Res.
**2003**, 3, 993–1022. [Google Scholar] - Lash, M.T.; Zhao, K. Early predictions of movie success: The who, what, and when of profitability. J. Manag. Inf. Syst.
**2016**, 33, 874–903. [Google Scholar] [CrossRef] - Schouten, K.; Frasincar, F. Survey on aspect-level sentiment analysis. IEEE Trans. Knowl. Data Eng.
**2016**, 28, 813–830. [Google Scholar] [CrossRef] - Moon, S.; Bergey, P.K.; Iacobucci, D. Dynamic effects among movie ratings, movie revenues, and viewer satisfaction. J. Mark.
**2010**, 74, 108–121. [Google Scholar] [CrossRef] - Kleinbaum, D.; Kupper, L.; Nizam, A.; Rosenberg, E. Applied Regresion Analysis and Other Multivariable Methods; Duxbury Press: Pacific Grove, CA, USA, 2013. [Google Scholar]
- Yu, X.; Liu, Y.; Huang, X.; An, A. Mining online reviews for predicting sales performance: A case study in the movie domain. IEEE Trans. Knowl. Data Eng.
**2012**, 24, 720–734. [Google Scholar] [CrossRef] - Kraft, F.B.; Granbois, D.H.; Summers, J.O. Brand evaluation and brand choice: A longitudinal study. J. Mark. Res.
**1973**, 10, 235–241. [Google Scholar] [CrossRef] - Lehmann, D.R. Television show preference: Application of a choice model. J. Mark. Res.
**1971**, 8, 47. [Google Scholar] [CrossRef]

**Figure 2.**The relationship between online information and the box-office revenues of the movie Gravity. (

**a**) The relationship between Google Trends and box-office revenues; (

**b**) the relationship between the number of reviews and box-office revenues.

**Figure 4.**Average heat of the three dimensions for 122 movies. (

**a**) The average heat of the dimension plot; (

**b**) the average heat of dimension star; (

**c**) the average heat of dimension genre.

**Figure 6.**The effects of the parameters on prediction accuracy. (

**a**) Effects of $p$; (

**b**) effects of $q$; (

**c**) effects of $\gamma $; (

**d**) effects of $\delta $.

**Figure 8.**Comparisons of model accuracy. (

**a**) Comparison of the eWOM model and the GSI model; (

**b**) comparison of the GSI model and the ARHS model; (

**c**) comparison of the eWOM model and the ARHS model; (

**d**) comparison of the ARO model and the ARHS model.

**Figure 10.**Accuracy of models predicting opening-week revenues. (

**a**) MAPE of the three models; (

**b**) RMSE of the three models.

Distributor | Freq. | Genre | Freq. | Release Month | Freq. | MPAA Ratings | Freq. |
---|---|---|---|---|---|---|---|

Warner Bros. | 18 | Drama | 38 | January | 10 | R | 57 |

Lionsgate | 16 | Comedy | 37 | February | 11 | PG-13 | 50 |

Paramount | 12 | Thriller | 14 | March | 12 | PG | 14 |

Weinstein | 10 | Action | 13 | April | 7 | NC-17 | 1 |

Fox | 10 | Sci-Fi | 10 | May | 10 | Total | 122 |

Sony | 9 | Horror | 9 | June | 6 | ||

Universal | 7 | Animation | 8 | July | 7 | ||

Open Road Films | 7 | Crime | 6 | August | 11 | ||

Focus Features | 6 | Fantasy | 5 | September | 11 | ||

Roadside Attractions | 6 | Adventure | 3 | October | 12 | ||

FilmDistrict | 4 | Sports | 2 | November | 11 | ||

Relativity | 4 | Music | 2 | December | 14 | ||

Buena Vista | 4 | Romance | 2 | ||||

CBS Films | 2 | Documentary | 1 | ||||

Bleecker Street | 2 | War | 1 | ||||

TriStar | 2 | ||||||

A24 | 1 | ||||||

Radius-TWC | 1 | ||||||

Rogue Pictures | 1 |

Domestic Gross (Million) | Freq. | Production Budget (Million) | Freq. |
---|---|---|---|

≤ 25 | 40 | ≤25 | 59 |

25–50 | 32 | 25–50 | 30 |

50–75 | 21 | 50–75 | 11 |

75–100 | 10 | 75–100 | 7 |

100–125 | 9 | 100–125 | 3 |

125–150 | 1 | 125–150 | 6 |

150–175 | 3 | 150–175 | 1 |

175–200 | 3 | 175–200 | 5 |

200–225 | 1 | Total | 122 |

225–250 | 1 | ||

250–275 | 1 |

Variable | Description (for Each Movie) | Measure and Data Sources |
---|---|---|

Sales | Daily domestic box-office revenues | Dollars (log-transformation); BoxOfficeMojo.com |

${v}_{t,1}$ | Daily number of reviews | Number (log-transformation); IMDb.com |

${v}_{t,2}$ | Daily valence of reviews | Average of daily ratings (0–10); IMDb.com |

${v}_{t,3}$ | Days from initial release | Number (1–49) |

${v}_{t,4}$ | Whether the day is on the weekend | 1 = the day is on the weekend (Fri, Sat, and Sun), 0 = others |

${v}_{t,5}$ | Daily number of cinemas | Number (log-transformation); BoxOfficeMojo.com |

${v}_{t,6}$ | Daily Google Trends of movie name | Number (0–100); Google.com |

Variable | Mean | Median | Maximum | Minimum | Std. Dev. | Skewness | Kurtosis |
---|---|---|---|---|---|---|---|

Sales | 1,039,207 | 263,875 | 35,167,017 | 10 | 2,204,483.7 | 5.351 | 46.855 |

${v}_{t,1}$ | 22.11526 | 11 | 506 | 0 | 38.754737 | 3.956 | 26.541 |

${v}_{t,2}$ | 3.801627 | 4 | 10 | 0 | 3.6083277 | 0.162 | 1.410 |

${v}_{t,5}$ | 1483.412 | 1195 | 4324 | 1 | 1264.8718 | 0.343 | 1.634 |

${v}_{t,6}$ | 33.49281 | 28 | 100 | 2 | 21.922873 | 1.101 | 3.815 |

plot | weight | plot | weight | plot | weight |
---|---|---|---|---|---|

story | 0.9% | plot | 0.5% | plot | 0.5% |

plot | 0.4% | story | 0.4% | story | 0.4% |

book | 0.4% | book | 0.3% | book | 0.4% |

horror | 0.3% | horror | 0.3% | horror | 0.3% |

dark | 0.2% | dark | 0.3% | dark | 0.2% |

original | 0.3% | original | 0.2% | original | 0.2% |

scary | 0.2% | scary | 0.2% | scary | 0.2% |

real | 0.2% | maze | 0.2% | maze | 0.2% |

pretty | 0.2% | pretty | 0.2% | pretty | 0.2% |

action | 0.2% | love | 0.2% | house | 0.2% |

Syntax Relations | Examples | Word Sentiments |
---|---|---|

Nominal subject | The plot is boring. | Plot: 3.0 |

Adjectival modifier | She is a good actor. | Actor: 3.8612 |

Direct object | I enjoy 3D. | 3D: 3.9782 |

Open clausal complement | I think the actor enjoys acting. | Acting: 3.9782 |

Adverb modifier | Tom performed earnestly. | Perform: 3.5 |

Relative clause modifier | I saw an actor who people dislike. | Actor: 3.5417 |

Variable | Description | Measures |
---|---|---|

${\vartheta}_{t,1}$ | The heat of the dimension plot on day t | Probabilistic |

${\vartheta}_{t,2}$ | The heat of the dimension star on day t | Probabilistic |

${\vartheta}_{t,3}$ | The heat of the dimension genre on day t | Probabilistic |

${\theta}_{t,1}$ | The sentiment of the dimension plot on day t | Numerical value |

${\theta}_{t,2}$ | The sentiment of the dimension star on day t | Numerical value |

${\theta}_{t,3}$ | The sentiment of the dimension genre on day t | Numerical value |

Variable | Mean | Median | Maximum | Minimum | Std. Dev. | Skewness | Kurtosis |
---|---|---|---|---|---|---|---|

${\vartheta}_{t,1}$ | 0.26932 | 0.00971 | 0.9999957 | 1.93 × 10^{−6} | 0.4063216 | 1.082 | 2.284 |

${\vartheta}_{t,2}$ | 0.13574 | 0.00971 | 0.9999957 | 2.16 × 10^{−6} | 0.3078812 | 2.184 | 5.992 |

${\vartheta}_{t,3}$ | 0.59495 | 0.95943 | 0.9999949 | 1.43 × 10^{−6} | 0.4544953 | −0.426 | 1.250 |

${\theta}_{t,1}$ | 3.07341 | 3 | 4.83333 | 0.130435 | 0.3685301 | −3.347 | 28.987 |

${\theta}_{t,2}$ | 3.09445 | 3 | 4.90476 | 0.130435 | 0.3519009 | −2.837 | 28.437 |

${\theta}_{t,3}$ | 3.08610 | 3 | 4.60417 | 0.130435 | 0.299457 | −3.282 | 35.261 |

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Lyu, X.; Jiang, C.; Ding, Y.; Wang, Z.; Liu, Y. Sales Prediction by Integrating the Heat and Sentiments of Product Dimensions. *Sustainability* **2019**, *11*, 913.
https://doi.org/10.3390/su11030913

**AMA Style**

Lyu X, Jiang C, Ding Y, Wang Z, Liu Y. Sales Prediction by Integrating the Heat and Sentiments of Product Dimensions. *Sustainability*. 2019; 11(3):913.
https://doi.org/10.3390/su11030913

**Chicago/Turabian Style**

Lyu, Xiaozhong, Cuiqing Jiang, Yong Ding, Zhao Wang, and Yao Liu. 2019. "Sales Prediction by Integrating the Heat and Sentiments of Product Dimensions" *Sustainability* 11, no. 3: 913.
https://doi.org/10.3390/su11030913