Abstract
Product reviews play a key role in e-commerce platforms. Studies show that many users read product reviews before a purchase and trust them to the same extent as personal recommendations. However, in many cases, the number of reviews per product is large and extracting useful information becomes a challenging task. Several websites have recently added an option to post tips—short, concise, practical, and self-contained pieces of advice about the products. These tips are complementary to the reviews and usually add a new non-trivial insight about the product, beyond its title, attributes, and description. Yet, most if not all major e-commerce platforms lack the notion of a tip as a first-class citizen and customers typically express their advice through other means, such as reviews.
In this work, we propose an extractive method for tip generation from product reviews. We focus on five popular e-commerce domains whose reviews tend to contain useful non-trivial tips that are beneficial for potential customers. We formally define the task of tip extraction in e-commerce by providing the list of tip types, tip timing (before and/or after the purchase), and connection to the surrounding context sentences. To extract the tips, we propose a supervised approach and leverage a publicly available dataset, annotated by human editors, containing 14,000 product reviews. To demonstrate the potential of our approach, we compare different tip generation methods and evaluate them both manually and over the labeled set. Our approach demonstrates particularly high performance for popular products in the Baby, Home Improvement, and Sports & Outdoors domains, with precision of over 95% for the top 3 tips per product. In addition, we evaluate the performance of our methods on previously unseen domains. Finally, we discuss the practical usage of our approach in real-world applications. Concretely, we explain how tips generated from user reviews can be integrated in various use cases within e-commerce platforms and benefit both buyers and sellers.
- [1] . 2013. Opinion fraud detection in online reviews by network effects. In Proc. of ICWSM, Vol. 7.Google Scholar
- [2] . 2022. Automated category tree construction in e-commerce. In Proc. of SIGMOD. 1770–1783.Google Scholar
- [3] . 2009. Multi-facet rating of product reviews. In European Conference on Information Retrieval. Springer, 461–472.Google Scholar
Digital Library
- [4] . 1999. Information retrieval as statistical translation. In Proc. of SIGIR. 222–229.Google Scholar
- [5] . 2017. Enriching word vectors with subword information. TACL 5 (2017), 135–146.Google Scholar
Cross Ref
- [6] . 2018. Local Consumer Review Survey. https://www.brightlocal.com/research/local-consumer-review-survey/.Google Scholar
- [7] . 2012. Folksonomy-based term extraction for word cloud generation. ACM Trans. Intell. Syst. Technol. 3, 4, Article
60 (Sept. 2012), 20 pages.Google ScholarDigital Library
- [8] . 1998. Multitask learning. In Learning to Learn. Springer, 95–133.Google Scholar
Cross Ref
- [9] . 2011. Quality evaluation of product reviews using an information quality framework. Decision Support Systems 50, 4 (2011), 755–768.Google Scholar
Digital Library
- [10] . 2021. Attentive multi-task learning for group itinerary recommendation. Knowledge and Information Systems 63, 7 (2021), 1687–1716.Google Scholar
Digital Library
- [11] . 2022. Multi-view graph attention network for travel recommendation. Expert Systems with Applications 191 (2022), 116234.Google Scholar
Digital Library
- [12] . 2021. A multi-task learning approach for improving travel recommendation with keywords generation. Knowledge-Based Systems 233 (2021), 107521.Google Scholar
Digital Library
- [13] . 2006. The effect of word of mouth on sales: Online book reviews. Journal of Marketing Research 43, 3 (2006), 345–354.Google Scholar
Cross Ref
- [14] . 2014. Polarized review summarization as decision making tool. In Proc. of AVI. 355–356.Google Scholar
- [15] . 2021. An image is worth a thousand terms? Analysis of visual e-commerce search. In Proc. of SIGIR. 102–112.Google Scholar
- [16] . 2018. Bert: Pre-training of deep bidirectional transformers for language understanding. arXiv preprint arXiv:1810.04805 (2018).Google Scholar
- [17] . 2018. The hitchhiker’s guide to testing statistical significance in natural language processing. In Proc. of ACL. 1383–1392.Google Scholar
- [18] . 2008. Do online reviews matter? An empirical investigation of panel data. Decision Support Systems 45, 4 (2008), 1007–1016.Google Scholar
Digital Library
- [19] . 2019. Learning to generate personalized product descriptions. In Proc. of CIKM. 389–398.Google Scholar
- [20] . 1971. Measuring nominal scale agreement among many raters.Psychological Bulletin 76, 5 (1971), 378–382.Google Scholar
Cross Ref
- [21] . 2019. Abstractive text summarization by incorporating reader comments. In Proc. of the AAAI Conference, Vol. 33. 6399–6406.Google Scholar
Digital Library
- [22] . 2014. Abstractive summarization of product reviews using discourse structure. In Proc. of EMNLP. 1602–1613.Google Scholar
- [23] . 2017. Extracting and ranking travel tips from user-generated reviews. In Proc. of WWW. 987–996.Google Scholar
- [24] . 2018. From royals to vegans: Characterizing question trolling on a community question answering website. In Proc. of SIGIR. 835–844.Google Scholar
- [25] . 2016. A joint many-task model: Growing a neural network for multiple NLP tasks. arXiv preprint abs/1611.01587 (2016).Google Scholar
- [26] . 2016. Ups and downs: Modeling the visual evolution of fashion trends with one-class collaborative filtering. In Proc. of WWW. 507–517.Google Scholar
- [27] . 2020. Query reformulation in e-commerce search. In Proc. of SIGIR. 1319–1328.Google Scholar
- [28] . 2021. Generating tips from product reviews. In Proc. of WSDM. 310–318.Google Scholar
- [29] . 1997. Long short-term memory. Neural Computation 9, 8 (1997), 1735–1780.Google Scholar
Digital Library
- [30] . 2004. Mining and summarizing customer reviews. In Proc. of KDD. 168–177.Google Scholar
- [31] . 2012. Manipulation of online reviews: An analysis of ratings, readability, and sentiments. Decision Support Systems 52, 3 (2012), 674–684.Google Scholar
Digital Library
- [32] . 2020. Personalized review recommendation based on users’ aspect sentiment. ACM Transactions on Internet Technology (TOIT) 20, 4 (2020), 1–26.Google Scholar
Digital Library
- [33] . 2013. Cross-language knowledge transfer using multilingual deep neural network with shared hidden layers. In 2013 IEEE International Conference on Acoustics, Speech and Signal Processing. IEEE, 7304–7308.Google Scholar
Cross Ref
- [34] . 2006. Automatically assessing review helpfulness. In Proc. of EMNLP. 423–430.Google Scholar
- [35] . 2014. Adam: A method for stochastic optimization. arXiv preprint abs/1412.6980 (2014).Google Scholar
- [36] . 2012. Selecting a characteristic set of reviews. In Proc. of KDD. 832–840.Google Scholar
- [37] . 2010. Efficient confident search in large review corpora. In ECML PKDD. Springer, 195–210.Google Scholar
Digital Library
- [38] . 2022. Lot or not: Identifying multi-quantity offerings in e-commerce. In Proc. of ECNLP 5. 250–262.Google Scholar
- [39] . 2019. Persona-aware tips generation. In The World Wide Web Conference. 1006–1016.Google Scholar
Digital Library
- [40] . 2017. Neural rating regression with abstractive tips generation for recommendation. In Proc. of SIGIR. 345–354.Google Scholar
- [41] . 2008. Electronic word-of-mouth in hospitality and tourism management. Tourism Management 29, 3 (2008), 458–468.Google Scholar
Cross Ref
- [42] . 2018. Generative adversarial network for abstractive text summarization. In 32nd AAAI Conference on Artificial Intelligence.Google Scholar
Cross Ref
- [43] . 1993. Building a large annotated corpus of English: The Penn Treebank. Computational Linguistics 19, 2 (1993), 313–330.Google Scholar
- [44] . 2013. Efficient estimation of word representations in vector space. arXiv preprint abs/1301.37810 (2013).Google Scholar
- [45] . 2013. Using micro-reviews to select an efficient set of reviews. In Proc. of CIKM. 1067–1076.Google Scholar
- [46] . 1989. Computer-intensive Methods for Testing Hypotheses. Wiley New York.Google Scholar
- [47] . 2019. Generating product descriptions from user reviews. In Proc. of WWW. 1354–1364.Google Scholar
- [48] . 2020. Descriptions from the customers: Comparative analysis of review-based product description generation methods. ACM Transactions on Internet Technology (TOIT) 20, 4 (2020), 1–31.Google Scholar
Digital Library
- [49] . 2014. Glove: Global vectors for word representation. In Proc. of EMNLP, Vol. 14. 1532–1543.Google Scholar
- [50] . 2001. An empirical study of the naive Bayes classifier. In IJCAI 2001 Workshop on Empirical Methods in Artificial Intelligence, Vol. 3. 41–46.Google Scholar
- [51] . 2012. Actionable clause detection from non-imperative sentences in howto instructions: A step for actionable information extraction. In TSD. Springer, 272–281.Google Scholar
- [52] . 2017. Uncovering consumers’ returning behaviour: A study of fashion e-commerce. International Review of Retail, Distribution and Consumer Research 27, 3 (2017), 284–299.Google Scholar
- [53] . 2013. Generating comparative summaries from reviews. In Proc. of CIKM. 1853–1856.Google Scholar
- [54] . 1980. Some effects of accounting expertise and information load: An empirical study. Accounting, Organizations and Society 5, 3 (1980), 323–338.Google Scholar
Cross Ref
- [55] . 2016. Deep multi-task learning with low level tasks supervised at lower layers. In Proc. of ACL, Vol. 2. 231–235.Google Scholar
- [56] . 2018. A taxonomy of queries for e-commerce search. In Proc. of SIGIR. 1245–1248.Google Scholar
- [57] . 1999. The influence of task interruption on individual decision making: An information overload perspective. Decision Sciences 30, 2 (1999), 337–360.Google Scholar
Cross Ref
- [58] . 2009. Revrank: A fully unsupervised algorithm for selecting the most helpful book reviews. In Proc. of ICWSM.Google Scholar
- [59] . 2020. E-commerce dispute resolution prediction. In Proc. of CIKM. 1465–1474.Google Scholar
- [60] . 2020. Product bundle identification using semi-supervised learning. In Proc. of SIGIR. 791–800.Google Scholar
- [61] . 2019. Extracting API tips from developer question and answer websites. In 2019 IEEE/ACM 16th International Conference on Mining Software Repositories (MSR ’19). IEEE, 321–332.Google Scholar
Digital Library
- [62] . 2012. Answers, not links: Extracting tips from yahoo! answers to address how-to web queries. In Proc. of WSDM. 613–622.Google Scholar
- [63] . 2012. Mining advices from weblogs. In Proc. of CIKM. 2347–2350.Google Scholar
- [64] . 2013. Toward advice mining: Conditional random fields for extracting advice-revealing text units. In Proc. of CIKM. 2039–2048.Google Scholar
- [65] . 2001. Monte Carlo cross validation. Chemometrics and Intelligent Laboratory Systems 56, 1 (2001), 1–11.Google Scholar
Cross Ref
- [66] . 2021. Online user review analysis for product evaluation and improvement. Journal of Theoretical and Applied Electronic Commerce Research 16, 5 (2021), 1598–1611.Google Scholar
Cross Ref
- [67] . 2016. Hierarchical attention networks for document classification. In Proc. of NAACL. 1480–1489.Google Scholar
- [68] . 2009. The impact of online user reviews on hotel room sales. International Journal of Hospitality Management 28, 1 (2009), 180–182.Google Scholar
Cross Ref
- [69] . 2018. Unsupervised tip-mining from customer reviews. Decision Support Systems 107 (2018), 116–124.Google Scholar
Cross Ref
- [70] . 2018. A local algorithm for product return prediction in e-commerce. In Proc. of IJCAI. 3718–3724.Google Scholar
Index Terms
The Tip of the Buyer: Extracting Product Tips from Reviews
Recommendations
Generating Tips from Product Reviews
WSDM '21: Proceedings of the 14th ACM International Conference on Web Search and Data MiningProduct reviews play a key role in e-commerce platforms. Studies show that many users read product reviews before purchase and trust them as much as personal recommendations. However, in many cases, the number of reviews per product is large and finding ...






Comments