Open Access. Powered by Scholars. Published by Universities.®

Physical Sciences and Mathematics Commons

Open Access. Powered by Scholars. Published by Universities.®

Articles 1 - 2 of 2

Full-Text Articles in Physical Sciences and Mathematics

Using Agents For Unification Of Information Extraction And Data Mining, Sharjeel Imtiaz, Azmat Hussain, Dr. Sikandar Hiyat Aug 2005

Using Agents For Unification Of Information Extraction And Data Mining, Sharjeel Imtiaz, Azmat Hussain, Dr. Sikandar Hiyat

International Conference on Information and Communication Technologies

Early work for unification of information extraction and data mining is motivational and problem stated work. This paper proposes a solution framework for unification using intelligent agents. A Relation manager agent extracted feature with cross feedback approach and also provide a Unified Undirected graphical handle. An RPM agent an approach to minimize loop back proposes pooling and model utilization with common parameter for both text and entity level abstractions.


Collective Multi-Label Classification, Nadia Ghamrawi, Andrew Mccallum Jan 2005

Collective Multi-Label Classification, Nadia Ghamrawi, Andrew Mccallum

Computer Science Department Faculty Publication Series

Common approaches to multi-label classification learn independent classifiers for each category, and employ ranking or thresholding schemes for classification. Because they do not exploit dependencies between labels, such techniques are only well-suited to problems in which categories are independent. However, in many domains labels are highly interdependent. This paper explores multilabel conditional random field (CRF) classification models that directly parameterize label co-occurrences in multi-label classification. Experiments show that the models outperform their singlelabel counterparts on standard text corpora. Even when multilabels are sparse, the models improve subset classification error by as much as 40%.