Slides of the 10 min layman's talk that preceded my PhD defence. In this talk I summarize ~4yrs of research in 10 minutes, so it's a very high-level overview.
16. • Gain new insights/discover new information
Entities of Interest
Discovery in Digital Traces
17. • Gain new insights/discover new information
• Answer questions: Who was involved? What
happened? Where, when and why did it
happen?
Entities of Interest
Discovery in Digital Traces
18. • Gain new insights/discover new information
• Answer questions: Who was involved? What
happened? Where, when and why did it
happen?
Entities of Interest
Discovery in Digital Traces
21. Entities of Interest
Discovery in Digital Traces
• “Things with distinct and independent
existence”
• Real-world entities central to answering 5
W’s.
22. Entities of Interest
Discovery in Digital Traces
• “Things with distinct and independent
existence”
• Real-world entities central to answering 5
W’s.
23. Entities of Interest
Discovery in Digital Traces
• “Things with distinct and independent
existence”
• Real-world entities central to answering 5
W’s.
32. Methods
• Information Retrieval
• Searching & finding things
• Natural Language Processing
• (automated) ’understanding’ of language
• Machine Learning
33. Methods
• Information Retrieval
• Searching & finding things
• Natural Language Processing
• (automated) ’understanding’ of language
• Machine Learning
• Using programs that ‘learn’ to do something
36. Two types of
Entities of Interest
Part 1: Entities in digital traces
37. Two types of
Entities of Interest
Part 1: Entities in digital traces
• Content/data
38. Two types of
Entities of Interest
Part 1: Entities in digital traces
• Content/data
Part 2: Entities that produce digital traces
39. Two types of
Entities of Interest
Part 1: Entities in digital traces
• Content/data
Part 2: Entities that produce digital traces
• Context/metadata
55. *****
Can we leverage collective intelligence to
construct entity representations for in-
creased retrieval effectiveness of entities
of interest?
56. *****
Can we leverage collective intelligence to
construct entity representations for in-
creased retrieval effectiveness of entities
of interest?
Yes!
69. Creation times Notification times
Creation times Notification times
Can we identify patterns in the times at
which people create reminders, and, via
notification times, when the associated
tasks are to be executed?
70. Creation times Notification times
Creation times Notification times
Can we identify patterns in the times at
which people create reminders, and, via
notification times, when the associated
tasks are to be executed?
Yes!
71. In Summary
• Part 1:
We propose methods for analyzing, predicting,
and retrieving emerging entities
• Part 2:
We propose methods for predicting future
activity by leveraging digital traces.