Novelty Detection in Patient Histories: Experiments with Measures Based on Text Compression thumbnail
Pause
Mute
Subtitles
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

Novelty Detection in Patient Histories: Experiments with Measures Based on Text Compression

Published on Oct 08, 20073274 Views

Reviewing a patient history can be very time consuming, partly because of the large number of consultation notes. Often, most of the notes contain little new information. Tools facilitating this and o

Related categories

Chapter list

Novelty Detection in Patient Histories: Experiments with Measures Based on Text Compression00:00
The Problem pt 100:32
The Problem pt 201:41
Evaluation Criterion02:22
Example pt 103:57
Example pt 204:09
Methods04:12
Preliminaries: Kolmogorov Complexity05:31
Preliminaries: Information Distance06:51
NCD - The Normalized Compression Distance07:47
Our Proposal: Normalized Asymmetric Compression Distance (NACD)08:26
How to Use these Distances to Rate Novelty10:33
TF-IDF Vector Space Measure11:06
Trivial Measures11:27
Combined Measure11:36
Results: Areas under the ROC Curves11:50
Results: The ROC Curves13:21
Conclusions14:11
Future Work14:47
Thanks for Listening!15:46