Dafna Shahaf

Publications51
h-index20
Citations1,345
Highly Influential Citations91
Claim Author Page
Author pages are created from data sourced from our academic publisher partnerships and public sources.

Recommended Authors

  • Carlos Guestrin
  • 207 Publications • 43,451 Citations
  • A. Kittur
  • 112 Publications • 9,757 Citations
  • Joel Chan
  • 56 Publications • 1,190 Citations
  • E. Horvitz
  • 482 Publications • 33,838 Citations
  • Publications
  • Influence
Connecting the dots between news articles
TLDR
This paper investigates methods for automatically connecting the dots -- providing a structured, easy way to navigate within a new topic and discover hidden connections in the news domain. Expand
Trains of thought: generating information maps
TLDR
This work proposes a methodology for creating structured summaries of information, which is able to produce maps which help users acquire knowledge efficiently and integrates user interaction into the framework, allowing users to alter the maps to better reflect their interests. Expand
Inside Jokes: Identifying Humorous Cartoon Captions
TLDR
This work describes how judgments about the humorousness of different captions are acquired and builds a classifier to identify funnier captions automatically, and uses it to find the best captions and study how its predictions could be used to significantly reduce the load on the cartoon contest's judges. Expand
Connecting Two (or Less) Dots: Discovering Structure in News Articles
TLDR
This article focuses on the news domain: given two news articles, a system automatically finds a coherent chain linking them together, and provides a fast search-driven algorithm to connect two fixed endpoints. Expand
Turning down the noise in the blogosphere
TLDR
This work defines the problem of learning a personalized coverage function by providing an appropriate user-interaction model and formalizing an online learning framework for this task, and provides a no-regret algorithm which can quickly learn a user's preferences from limited feedback. Expand
Learning to Route
TLDR
The preliminary results regarding the power of data-driven routing suggest that applying ML (specifically, deep reinforcement learning) to this context yields high performance and is a promising direction for further research. Expand
Towards a Theory of AI Completeness
TLDR
This work serves as a formal basis for investigation of problems that researchers treat as hard AI problems and allows progress in AI as a field to be more measurable, instead of measurable with respect to problem-specific quantities. Expand
Information cartography: creating zoomable, large-scale maps of information
TLDR
Pilot user studies over real-world datasets demonstrate that the proposed methodology for creating structured summaries of information, called zoomable metro maps, helps users comprehend complex stories better than prior work. Expand
Connecting the dots between news articles
TLDR
This paper investigates methods for automatically connecting the dots - providing a structured, easy way to navigate within a new topic and discover hidden connections in the news. Expand
Learning To Route with Deep RL
TLDR
The results suggest that applying deep reinforcement learning to this context yields high performance and is thus a promising direction for further research, and outlines a research agenda for data-driven routing. Expand
...
1
2
3
4
5
...