Reinforced Product Metadata Selection for Helpfulness Assessment of Customer Reviews
9337Authors
nan
Abstract
AbstractTo automatically assess the helpfulness of a customer review online, conventional approaches generally acquire various linguistic and neural embedding features solely from the textual content of the review itself as the evidence. We, however, find out that a helpful review is largely concerned with the metadata (such as the name, the brand, the category, etc.) of its target product. It leaves us with a challenge of how to choose the correct key-value product metadata to help appraise the helpfulness of free-text reviews more precisely. To address this problem, we propose a novel framework composed of two mutual-benefit modules. Given a product, a selector (agent) learns from both the keys in the product metadata and one of its reviews to take an action that selects the correct value, and a successive predictor (network) makes the free-text review attend to this value to obtain better neural representations for helpfulness assessment. The predictor is directly optimized by SGD with the loss of helpfulness prediction, and the selector could be updated via policy gradient rewarded with the performance of the predictor. We use two real-world datasets from Amazon.com and Yelp.com, respectively, to compare the performance of our framework with other mainstream methods under two application scenarios: helpfulness identification and regression of customer reviews. Extensive results demonstrate that our framework can achieve state-of-the-art performance with substantial improvements.
Keywords
Full Content
Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing, pages 1675–1683, Hong Kong, China, November 3–7, 2019. c⃝2019 Association for Computational Linguistics 1675 Reinforced Product Metadata Selection for Helpfulness Assessment of Customer Reviews Miao Fan, Chao Feng, Mingming Sun, Ping Li Cognitive Computing Lab Baidu Research No.10 Xibeiwang East Road, Beijing, 10085, China 10900 NE 8th St, Bellevue, WA 98004, USA {fanmiao, v fegchao, sunmingming01, liping11}@baidu.com Abstract To automatically assess the helpfulness of a customer review online, conventional ap- proaches generally acquire various linguistic and neural embedding features solely from the textual content of the review itself as the ev- idence. We, however, find out that a help- ful review is largely concerned with the meta- data (such as the name, the brand, the cate- gory, etc.) of its target product. It leaves us with a challenge of how to choose the correct key-value product metadata to help appraise the helpfulness of free-text reviews more pre- cisely. To address this problem, we propose a novel framework composed of two mutual- benefit modules. Given a product, a selector (agent) learns from both the keys in the prod- uct metadata and one of its reviews to take an action that selects the correct value, and a successive predictor (network) makes the free- text review attend to this value to obtain better neural representations for helpfulness assess- ment. The predictor is directly optimized by SGD with the loss of helpfulness prediction, and the selector could be updated via policy gradient rewarded with the performance of the predictor. We use two real-world datasets from Amazon.com and Yelp.com, respectively, to compare the performance of our framework with other mainstream methods under two ap- plication scenarios: helpfulness identification and regression of customer reviews. Exten- sive results demonstrate that our framework can achieve state-of-the-art performance with substantial improvements. 1 Introduction The massive number of reviews left by many ex- perienced consumers on online products are our priceless treasure in e-commerce. We believe that online customer reviews can provide more sub- jective and informative opinions from various per- spectives on the products besides the objective de- Figure 1: A screenshot of the top review on a product sold in Amazon.com: SodaStream Fruit Drops Variety Pack, 1.67 Pound. Online users can vote for the help- fulness of each customer review by clicking the “Help- ful” button underneath. scriptions given by their merchants. Hence, we prefer browsing the reviews online for the sake of finding our desirable products. However, it is quite time-consuming for potential buyers to sift through millions of online reviews with uneven qualities to make purchase decisions. To discover and surface helpful reviews for cus- tomers, some well-known websites (e.g., Ama- zon.com) have launched a module illustrated by Figure 1 which asks for feedback on the help- fulness of online reviews. And a recent study1 reported that this featured module can increase the revenue of Amazon with an estimated 27 bil- lion U.S. dollars …
Content truncated to 500 words
Labeled Text
AbstractTo automatically assess the helpfulness of a customer review online, conventional approaches generally acquire various linguistic and neural embedding features solely from the textual content of the review itself as the evidence. We, however, find out that a helpful review is largely concerned with the metadata (such as the name, the brand, the category, etc.) of its target product. It leaves us with a challenge of how to choose the correct key-value product metadata to help appraise the helpfulness of free-text reviews more precisely. To address this problem, we propose a novel framework composed of two mutual-benefit modules. Given a product, a selector (agent) learns from both the keys in the product metadata and one of its reviews to take an action that selects the correct value, and a successive predictor (network) makes the free-text review attend to this value to obtain better neural representations for helpfulness assessment. The predictor is directly optimized by SGD with the loss of helpfulness prediction, and the selector could be updated via policy gradient rewarded with the performance of the predictor. We use two real-world datasets from Amazon.com and Yelp.com, respectively, to compare the performance of our framework with other mainstream methods under two application scenarios: helpfulness identification and regression of customer reviews. Extensive results demonstrate that our framework can achieve state-of-the-art performance with substantial improvements.
Publication Details
- Paper ID:
- 9337
- Year:
- 2019
- Publication:
- TIPSTER
- Publication Type:
- conf
- DOI:
- nan