In e-commerce sites, customer questions on the product details-page express the customers' information needs about the product. The answers to these questions often provide the necessary information. In this work, we present and address the novel task of generating product insights from community questions and answers (Q&A). These insights can be presented to customers to assist them in their shopping journey. Our method first generates concise, self-contained sentences based on the information in the Q&A. Then insights are selected based on the prominence of their associated questions. Empirical evaluation attests to the effectiveness of our approach in generating well-formed, objective, and helpful insights that are often not available in the product description or in summaries of customer reviews.

Useful tips extracted from product reviews assist customers to take a more informed purchase decision, as well as making a better, easier, and safer usage of the product. In this work we argue that extracted tips should be examined based on the amount of support and opposition they receive from all product reviews. A classifier, developed for this purpose, determines the degree to which a tip is supported or contradicted by a single review sentence. These support-levels are then aggregated over all review sentences, providing a global support score, and a global contradiction score, reflecting the support-level of all reviews to the given tip, thus improving the customer confidence in the tip validity. By analyzing a large set of tips extracted from product reviews, we propose a novel taxonomy for categorizing tips as highly-supported, highly-contradicted, controversial (supported and contradicted), and anecdotal (neither supported nor contradicted).

In this study we address the problem of identifying the purchase-state of users, based on product-related questions they ask on an eCommerce website. We differentiate between questions asked before buying a product (pre-purchase) and after (post-purchase). At first, we study the ambiguity that exists in purchase-states' definition, and then investigate the linguistic characteristics of the questions in each state. We analyze the discrepancy between the language models of pre- and post-purchase questions, and offer two classification schemes for this task, both outperform human judgments. We additionally show the effectiveness of our classification models in improving real world applications for both consumers and sellers.

The connection between messaging and action is fundamental both to web applications, such as web search and sentiment analysis, and to economics. However, while prominent online applications exploit messaging in natural (human) language in order to predict non-strategic action selection, the economics literature focuses on the connection between structured stylized messaging to strategic decisions in games and multi-agent encounters. This paper aims to connect these two strands of research, which we consider highly timely and important due to the vast online textual communication on the web. Particularly, we introduce the following question: can free text expressed in natural language serve for the prediction of action selection in an economic context, modeled as a game? We initiate research on this question by providing preliminary positive results.

The connection between messaging and action is fundamental both to web applications, such as web search and sentiment analysis, and to economics. However, while prominent online applications exploit messaging in natural (human) language in order to predict non-strategic action selection, the economics literature focuses on the connection between structured stylized messaging to strategic decisions in games and multi-agent encounters. This paper aims to connect these two strands of research, which we consider highly timely and important due to the vast online textual communication on the web. Particularly, we introduce the following question: Can free text expressed in natural language serve for the prediction of action selection in an economic context, modeled as a game In order to initiate the research on this question, we introduce the study of an individual’s action prediction in a one-shot game based on free text he/she provides, while being unaware of the game to be played. We approach the problem by attributing commonsensical personality attributes via crowd-sourcing to free texts written by individuals, and employing transductive learning to predict actions taken by these individuals in one-shot games based on these attributes. Our approach allows us to train a single classifier that can make predictions with respect to actions taken in multiple games. In experiments with three well-studied games, our algorithm compares favorably with strong alternative approaches. In ablation analysis, we demonstrate the importance of our modeling choices—the representation of the text with the commonsensical personality attributes and our classifier—to the predictive power of our model.