• Publications
  • Influence
Beyond DCG: user behavior as a predictor of a successful search
TLDR
This work shows empirically that user behavior alone can give an accurate picture of the success of the user's web search goals, without considering the relevance of the documents displayed. Expand
Modeling dwell time to predict click-level satisfaction
TLDR
It is shown that the topic of the page, its length and its readability level are critical in determining the amount of dwell time needed to predict whether any click is associated with satisfaction, and a method to model and provide a better understanding of click dwell time is proposed. Expand
Understanding and Predicting Graded Search Satisfaction
TLDR
This work is the first to study the problem of understanding and predicting graded (multi-level) search satisfaction and shows that its approach can predict subtle changes in search satisfaction more accurately than state-of-the-art methods, affording greater insight into search satisfaction. Expand
Identifying Text Polarity Using Random Walks
TLDR
A Markov random walk model is applied to a large word related-ness graph, producing a polarity estimate for any given word, and outperforms the state of the art methods in the semi-supervised setting. Expand
Automatic Online Evaluation of Intelligent Assistants
TLDR
This paper uses implicit feedback from users to predict whether users are satisfied with the intelligent assistant as well as its components, i.e., speech recognition and intent classification, and develops consistent and automatic approaches that can evaluate different tasks in voice-activated intelligent assistants. Expand
Detecting Subgroups in Online Discussions by Modeling Positive and Negative Relations among Participants
TLDR
It is found that the polarity of the links between users can be predicted with high accuracy given the text they exchange, and it is shown that the automatically predicted networks are consistent with social psychology theories of balance. Expand
Multi-Source Cross-Lingual Model Transfer: Learning What to Share
TLDR
This model leverages adversarial networks to learn language-invariant features, and mixture-of-experts models to dynamically exploit the similarity between the target language and each individual source language to further boost target language performance. Expand
Beyond clicks: query reformulation as a predictor of search satisfaction
TLDR
It is shown that a query-based model (with no click information) can indicate satisfaction more accurately than click-based models, and that search success is an incremental process for successful tasks with multiple queries. Expand
Enhancing personalized search by mining and modeling task behavior
TLDR
A method whereby other users performing similar tasks to the current user and leverage their on-task behavior to identify Web pages to promote in the current ranking yields promising gains in retrieval performance, and has direct implications for improving personalization in search systems. Expand
Struggling or exploring?: disambiguating long search sessions
TLDR
This paper analyzes struggling and exploring behavior in Web search using log data from a commercial search engine, and builds classifiers that can accurately distinguish between exploring and struggling sessions using behavioral and topical features. Expand
...
1
2
3
4
5
...