Wager, Hastie, & Efron (2014) 建議利用摺刀法 (jackknife) 估計隨機森林 (random forests) 的信賴區間。本研究利用蒙地卡羅模擬驗證摺刀法估計區間的覆蓋機率。我們發現估計區間的覆蓋機率大致上與理論值相符。接下來我們利用隨機森林估計Amazon.com平板商品評論幫助性 (helpfulness of product reviews) 與探討各商品評論特徵的關係。我們發現在固定特定品牌與商品評分之下，評論評論長度(length)、累積曝光時間(longevity)對於評論幫助性是顯著的，並且有非線性的關係。而評論內容的語意特徵，如:評論兩極性 (polarity)、評論價性強度(valence strength)、評論長度、評論累積曝光時間於線性迴歸模型中為顯著變數。與線性迴歸相較，隨機森林可以考慮特徵與評論幫助性非線性的關係。

Wager, Hastie, &Efron (2014) suggest using the Jackknife method to estimate the confidence interval of random forests. This paper uses Monte Corlo simulation to verify the coverage probability of the confidence interval of random forests. We found that the coverage probability of the confidence interval of random forests is roughly consistent with the theoretical value. We also use random forests to estimate the relationship between helpfulness of reviews and other variables related to product reviews. For comparison, we also included linear regression model. We found that under certain brand and product ratings, the length and the longevity of product reviews are significant for helpfulness of reviews. Compared to linear regression model, random forests can consider the non-linear relationship between features.

1. Bradley Efron. (1992). Jackknife-after-bootstrap standard errors and influence functions, Journal of the Royal Statistical Society. Series B (Methodological), 83–127.
2. Bradley Efron. (2014). Estimation and Accuracy After Model Selection, Journal of the American Statistical Association, 109:507, 991-1007
3. Breiman, L. (1996). Bagging predictors. Machine Learning 24, 123–140.
4. Breiman, L., Friedman, J., Olshen, R. and Stone, C. (1984). Classification and Regression Trees. Wadsworth.
5. Duan, W., Cao, Q., & Gan, Q. (2010). Investigating Determinants of Voting for the "Helpfulness" of Online Consumer Reviews: A Text Mining Approach. AMCIS.
6. eMarketer(2017),Worldwide Retail and Ecommerce Sales: eMarketer's Estimates for 2016–2021.
7. Ghose, Anindya and Ipeirotis, (2011). Panagiotis G., Estimating the Helpfulness and Economic Impact of Product Reviews: Mining Text and Reviewer Characteristics. IEEE Transactions on Knowledge and Data Engineering, 2011.
8. Hu, Nan and Liu, Ling and Zhang, Jie (Jennifer). (2008). Do Online Reviews Affect Product Sales? The Role of Reviewer Characteristics and Temporal Effects. Information Technology & Management, 9, 3, 201-214.
9. Korfiatis, Nikos & E, García-Barriocanal & Sánchez--Alonso, S. (2012). Evaluating content quality and helpfulness of online product reviews: The interplay of review helpfulness vs. review content. Electronic Commerce Research and Applications, 11, 3, 205-217.
10. Leo Breiman. (2001). Random forests. Machine Learning, 45, 1, 5–32.
11. Mudambi, Susan & Schuff, David. (2010). What Makes a Helpful Online Review? A Study of Customer Reviews on Amazon.com.com. MIS Quarterly. 34, 1, 185-200.
12. Pan, Y. and Zhang, J.Q. (2011). Born unequal: A study of the helpfulness of user-generated product reviews. Journal of Retailing, 87, 4, 598-612.
13. Quenouille, M. H. (1949). Approximate tests of correlation in time-series. J. Roy. Statist. Soc. Ser. B 11, 68-84.
14. Wager, Stefan & Hastie, Trevor & Efron, Bradley. (2014). Confidence Intervals for Random Forests: The Jackknife and the Infinitesimal Jackknife. Journal of machine learning research : JMLR, 15, 1625-1651.
15. R Core Team (2018). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL:https://www.R-project.org/.
16. 黃俊堯、柳秉佑 (2016)。消費者線上口碑與評論研究:國內外相關文獻回顧與討論。臺大管理論叢，26(3)，215-256。