National Library of Australia

We’re delighted to be able to re-open the Library for pre-booked ticketed access to our collections. Read more...

Record 1 of 1
You must be logged in to Tag Records
Learning probabilistic relational concept descriptions [microform]
Bib ID 3283589
Format MicroformMicroform, BookBook
Author
Ali, Kamal Mahmood
 
Description 224 p. 
Summary

This dissertation presents methods for increasing the accuracy of probabilistic classification rules learned from noisy, relational data. It addresses the problem of learning probabilistic rules in noisy, "real-world" data sets, the problem of "small disjuncts" in which rules that apply to rare subclasses have high error rates, and the problems that arise in domains in which the learning algorithm is forced to pick from many rules that appear to be equally good. It is shown that learning a class description for each class in the data--the one-per-class approach--and attaching probabilistic estimates to the learned rules allows accurate classifications to be made on real-world data sets. The thesis presents the system HYDRA which implements this approach. It is shown that the resulting classifications are often more accurate than those made by three major methods for learning from noisy, relational data.

Furthermore, the learned rules are relational and so are more expressive than the attribute-value rules learned by most induction systems. Several results are also presented in the arena of multiple models. The multiple models approach is relevant to the problem of making accurate classifications in "real-world" domains since it facilitates evidence combination which is needed to accurately learn on such domains. The most important result of the multiple models research is that the amount of error reduction afforded by the multiple models approach is linearly correlated with the degree to which the individual models make errors in an uncorrelated manner. It is shown that it is possible to learn models that make less correlated errors in domains in which there are many gain ties.

The third major result of the research on multiple models is the realization that models should be learned that make errors in a negatively-correlated manner rather than those that make errors in an uncorrelated manner. Finally, results are presented on the small-disjuncts problem in which rules that apply to rare subclasses have high error rates. It is shown that the one-per-class approach reduces error rates for such rare rules while not sacrificing the error rates of the other rules.

Notes

(UnM)AAI9612058

Source: Dissertation Abstracts International, Volume: 56-12, Section: B, page: 6852.

Chair: Michael J. Pazzani.

Thesis (Ph.D.)--University of California, Irvine, 1996.

Reproduction Microfiche. Ann Arbor, Mich.: University Microfilms International. 
Subjects Computer Science.  |  Artificial Intelligence.
Other authors/contributors University of California, Irvine

Online

    In the Library

    Request this item to view in the Library's reading rooms using your library card. To learn more about how to request items watch this short online video Help Video.

    Details Collect From
    mc 252 96-12058 Temporarily unavailable for request - contact staff

    Order a copy

    - Copyright or permission restrictions may apply. We will contact you if necessary.

    Help Video To learn more about Copies Direct watch this short online video Help Video.
    close Can I borrow items from the Library?

    You need Flash player 8+ and JavaScript enabled to view this video embedded.

    You can view this on the NLA website.

    close What can I get online?

    You need Flash player 8+ and JavaScript enabled to view this video embedded.

    You can view this on the NLA website.

    close Can I get copies of items from the Library?

    You need Flash player 8+ and JavaScript enabled to view this video embedded.

    You can view this on the NLA website.

    Aboriginal and Torres Strait Islander Flags
    Aboriginal, Torres Strait Islander and other First Nations people are advised that this catalogue contains names, recordings and images of deceased people and other content that may be culturally sensitive. Please also be aware that you may see certain words or descriptions in this catalogue which reflect the author’s attitude or that of the period in which the item was created and may now be considered offensive.