Hierarchical Classification with Rare Categories and Inconsistencies

Simple item page
dc.contributor.advisorRangwala, Huzefa
dc.contributor.authorNaik, Azad
dc.creatorNaik, Azad
dc.date.accessioned2018-10-22T01:21:15Z
dc.date.available2018-10-22T01:21:15Z
dc.date.issued2017
dc.description.abstractAdvancement in digital technology has generated a massive amount of data. Large amount of information streaming in from various sources such as phones, tablets, computers and internet has made an immense need to provide a structured and organized view of the data. Hierarchy (taxonomy) is one of the most easy and convenient way of data organization. It has been used extensively to store large volumes of data in various application domains ranging from biological datasets (for organizing genes and protein sequences) to image and text datasets (for providing the structured view of billions of images and web pages). Hierarchical structure representation of the data can be effectively used to eliminate the expensive and tedious task of manual classification. To this end, Hierarchical Classification (HC) deals with the task of automatically classifying the instances (examples) within the topic hierarchy have been developed. Although, HC is popular among the researchers due to its wide application, it faces severe challenges due to the following reasons:
dc.format.extent149 pages
dc.identifier.urihttps://hdl.handle.net/1920/11299
dc.language.isoen
dc.rightsCopyright 2017 Azad Naik
dc.subjectComputer science
dc.subjectEngineering
dc.subjectHierarchical Classification
dc.subjectHierarchy (Taxonomy)
dc.subjectHybrid Prediction
dc.subjectInconsistent hierarchy
dc.subjectLogistic Regression
dc.subjectSupervised Learning
dc.titleHierarchical Classification with Rare Categories and Inconsistencies
dc.typeDissertation
thesis.degree.disciplineComputer Science
thesis.degree.grantorGeorge Mason University
thesis.degree.levelPh.D.

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
Naik_gmu_0883E_11363.pdf
Size:
3.99 MB
Format:
Adobe Portable Document Format
Download

Collections

College of Engineering and Computing