User profiles for Mark Craven
Mark CravenProfessor of Biostatistics and Medical Informatics, University of Wisconsin Verified email at biostat.wisc.edu Cited by 12749 |
[PDF][PDF] An analysis of active learning strategies for sequence labeling tasks
Active learning is well-suited to many problems in natural language processing, where
unlabeled data may be abundant but annotation is slow and expensive. This paper aims to shed …
unlabeled data may be abundant but annotation is slow and expensive. This paper aims to shed …
Extracting tree-structured representations of trained networks
A significant limitation of neural networks is that the represen (cid: 173) tations they learn
are usually incomprehensible to humans. We present a novel algorithm, TREPAN, for …
are usually incomprehensible to humans. We present a novel algorithm, TREPAN, for …
[PDF][PDF] Learning to extract symbolic knowledge from the World Wide Web
The World Wide Web is a vast source of information accessible to computers, but understandable
only to humans. The goal of the research described here is to automatically create a …
only to humans. The goal of the research described here is to automatically create a …
[PDF][PDF] Constructing biological knowledge bases by extracting information from text sources.
M Craven, J Kumlien - ISMB, 1999 - cdn.aaai.org
Recently, there has been much effort in making databases for Inolecular biology more
accessible osld interoperable. However, information in text. form, such as MEDLINE records, …
accessible osld interoperable. However, information in text. form, such as MEDLINE records, …
Multiple-instance active learning
In a multiple instance (MI) learning problem, instances are naturally organized into bags
and it is the bags, instead of individual instances, that are labeled for training. MI learners …
and it is the bags, instead of individual instances, that are labeled for training. MI learners …
Learning to construct knowledge bases from the World Wide Web
The World Wide Web is a vast source of information accessible to computers, but understandable
only to humans. The goal of the research described here is to automatically create a …
only to humans. The goal of the research described here is to automatically create a …
Incorporating domain knowledge into topic modeling via Dirichlet forest priors
Users of topic modeling methods often have knowledge about the composition of words that
should have high or low probability in various topics. We incorporate such domain …
should have high or low probability in various topics. We incorporate such domain …
Using sampling and queries to extract rules from trained neural networks
MW Craven, JW Shavlik - Machine learning proceedings 1994, 1994 - Elsevier
Abstract Concepts learned by neural networks are difficult to understand because they are
represented using large assemblages of real-valued parameters. One approach to …
represented using large assemblages of real-valued parameters. One approach to …
Using neural networks for data mining
MW Craven, JW Shavlik - Future generation computer systems, 1997 - Elsevier
Neural networks have been successfully applied in a wide range of supervised and
unsupervised learning applications. Neural-network methods are not commonly used for data-…
unsupervised learning applications. Neural-network methods are not commonly used for data-…
[BOOK][B] Extracting comprehensible models from trained neural networks
MW Craven - 1996 - search.proquest.com
Although neural networks have been used to develop highly accurate classifiers in numerous
real-world problem domains, the models they learn are notoriously difficult to understand. …
real-world problem domains, the models they learn are notoriously difficult to understand. …