Matches in ScholarlyData for { ?s ?p ?o. }
- 491 keyword "unsupervised".
- 491 title "Targeted Disambiguation of Ad-hoc, Homogeneous Sets of Named Entities".
- authorList type List.
- authorList hasFirstItem item-1.
- authorList hasItem item-1.
- authorList hasItem item-2.
- authorList hasItem item-3.
- authorList hasItem item-4.
- authorList hasLastItem item-4.
- item-1 type ListItem.
- item-1 hasContent chi-wang.
- item-1 next item-2.
- item-2 type ListItem.
- item-2 hasContent kaushik-chakrabarti.
- item-2 next item-3.
- item-2 previous item-1.
- item-3 type ListItem.
- item-3 hasContent tao-cheng.
- item-3 next item-4.
- item-3 previous item-2.
- item-4 type ListItem.
- item-4 hasContent surajit-chaudhuri.
- item-4 previous item-3.
- 498 creator brendan-lucier.
- 498 creator eva-tardos.
- 498 creator renato-paes-leme.
- 498 type InProceedings.
- 498 label "On Revenue in the Generalized Second Price Auction".
- 498 sameAs 498.
- 498 abstract "The Generalized Second Price (GSP) auction is the primary auction used for selling sponsored search advertisements. In this paper we consider the revenue of this auction at equilibrium. Most previous work on GSP focuses on envy-free equilibria of the full information version of this game. We consider revenue in equilibria that are not necessarily envy-free, both in the full information game and in the Bayesian (partial information) game. Note that in the Bayesian setting, envy-free equilibria may not exist. We prove that if agent values are drawn from identical distributions satisfying the regularity assumption, then the GSP auction paired with an appropriate reserve price generates a constant fraction (1/6th) of the optimal revenue. In the full-information game, we show that at any Nash equilibrium of the GSP auction obtains at least half of the revenue of the VCG mechanism excluding the payment of a single participant. This bound holds also with any reserve price, and is tight. Finally, we consider the tradeoff between maximizing revenue and social welfare. We introduce a natural convexity assumption on the click-through rates and show that it implies that the revenue-maximizing equilibrium of GSP in the full information model will necessarily be envy-free. In particular, it is always possible to maximize revenue and social welfare simultaneously when click-through rates are convex. On the other hand, without this convexity assumption, we demonstrate that revenue may be maximized at a non-envy-free equilibrium that generates a socially inefficient allocation.".
- 498 hasAuthorList authorList.
- 498 isPartOf proceedings.
- 498 keyword "Ad Auction,".
- 498 keyword "Bayesian".
- 498 keyword "Equilibrium Hierarchy".
- 498 keyword "Generalized Second Price".
- 498 keyword "Revenue".
- 498 title "On Revenue in the Generalized Second Price Auction".
- authorList type List.
- authorList hasFirstItem item-1.
- authorList hasItem item-1.
- authorList hasItem item-2.
- authorList hasItem item-3.
- authorList hasLastItem item-3.
- item-1 type ListItem.
- item-1 hasContent brendan-lucier.
- item-1 next item-2.
- item-2 type ListItem.
- item-2 hasContent renato-paes-leme.
- item-2 next item-3.
- item-2 previous item-1.
- item-3 type ListItem.
- item-3 hasContent eva-tardos.
- item-3 previous item-2.
- 505 creator arjun-mukherjee.
- 505 creator bing-liu.
- 505 type InProceedings.
- 505 label "Spotting Fake Reviewer Groups in Consumer Reviews".
- 505 sameAs 505.
- 505 abstract "Opinionated social media like product reviews are now widely used by individuals and organizations for their decision making. However, due to the reason of profit or fame, people try to game the system by opinion spamming (e.g., writing fake reviews) to promote or demote some target products. For reviews to reflect genuine user experiences and opinions, such spam reviews should be detected. Prior works on opinion spam focused on detecting fake reviews and individual spammers. However, a spammer group (a group of reviewers who work collaboratively to write fake reviews) is even more damaging and can take total control of the sentiment on the target product due to its size. This paper studies spam detection in the collaborative setting, i.e., to discover review spammer groups. The proposed method first uses a frequent itemset mining method to find a set of candidate groups. It then uses several behavioral models derived from the collusion phenomenon among spammers and relation models based on inter-relationships among products, groups, and individual reviewers to detect spammer groups. Additionally, we also built a labeled group spam dataset. To the best of our knowledge, this is the first labeled dataset on group opinion spam. Experimental results show that the proposed method is highly effective and it outperforms multiple strong baselines including both the state-of-the-art learning to rank and supervised classification algorithms.".
- 505 hasAuthorList authorList.
- 505 isPartOf proceedings.
- 505 keyword "Fake reviewers".
- 505 keyword "Group deceptive behavior".
- 505 keyword "Group opinion spam".
- 505 title "Spotting Fake Reviewer Groups in Consumer Reviews".
- authorList type List.
- authorList hasFirstItem item-1.
- authorList hasItem item-1.
- authorList hasItem item-2.
- authorList hasLastItem item-2.
- item-1 type ListItem.
- item-1 hasContent arjun-mukherjee.
- item-1 next item-2.
- item-2 type ListItem.
- item-2 hasContent bing-liu.
- item-2 previous item-1.
- 518 creator alex-bocharov.
- 518 creator eric-horvitz.
- 518 creator jaime-teevan.
- 518 creator kira-radinsky.
- 518 creator krysta-svore.
- 518 creator susan-dumais.
- 518 type InProceedings.
- 518 label "Learning and Predicting Behavioral Dynamics on the Web".
- 518 sameAs 518.
- 518 abstract "Users’ behaviors on theWeb shift over time in terms of pages of interest to users, queries issued to a search engine, and the underlying informational goals behind the queries. We examine how to model and predict user behavior on theWeb over time. We explore several modeling approaches and de- velop a temporal modeling framework adapted from physics and signal processing that can be used for predicting user behavior. We also explore other dynamics of the Web, such as the detection of periodicities and surprises. We develop a learning procedure that can be used to construct models of users’ behaviors based on features of present and histor- ical behaviors. The results of experiments indicate that, by learning how to predict user behavior using our framework, we can achieve significant improvements in prediction and explicitly learn when to apply each of a set of behavior pre- diction models.".
- 518 hasAuthorList authorList.
- 518 isPartOf proceedings.
- 518 keyword "Query log mining".
- 518 keyword "Temporal User Behavior prediction".
- 518 keyword "Temporal data analysis".
- 518 keyword "Web dynamics and search".
- 518 title "Learning and Predicting Behavioral Dynamics on the Web".
- authorList type List.
- authorList hasFirstItem item-1.
- authorList hasItem item-1.
- authorList hasItem item-2.
- authorList hasItem item-3.
- authorList hasItem item-4.