Selecting Negative Training Documents for Better Learning

ABDULMOHSEN, ALGARNI (2015) Selecting Negative Training Documents for Better Learning. In: Third International Conference on Advances in Computing, Electronics and Communication - ACEC 2015, 10-11 October, 2015, Zurich, Switzerland.

[img]
Preview
Text
20151023_070807.pdf - Published Version

Download (782kB) | Preview
Official URL: https://www.seekdl.org/conferences/paper/details/6...

Abstract

In general, there are two types of feedback documents: positive feedback documents and negative feedback documents. These types share some knowledge because they retrieved using the same query. It is clear that all feedback documents contain some noise knowledge that affects the quality of the extracted features. The amount of noise is different from document to another. Therefore, the number of feedback documents affects the amount of extracting noise features. Then, using all feedback documents can increase the number of extracted noise features. However, we believe that negative feedback documents contain more noise than positive feedback documents. In this paper, we introduce a methodology to select some negative feedback documents to extract high-quality features and to reduce the amount of noises features.

Item Type: Conference or Workshop Item (Paper)
Uncontrolled Keywords: knowledge extraction, feature selection , relevance feedback.
Depositing User: Mr. John Steve
Date Deposited: 19 Apr 2019 12:07
Last Modified: 19 Apr 2019 12:07
URI: http://publications.theired.org/id/eprint/1392

Actions (login required)

View Item View Item