Advances In Natural Language Processing:Capsule Networks for Part of Speech Tagging
PublisherThe University of Arizona.
RightsCopyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.
AbstractPart of Speech (POS) tagging tasks have constantly evolved throughout the years within the corpus of linguistics. Many shortcomings and complications have arrived with implementations of POS tagging through various computer algorithms.This task fosters such complications due to the many nuances of understanding language such as words having several different meanings, or maybe a phrase expressing totally different thoughts due to a one word difference. Its factors like these that computational linguist try to incorporate when teaching language to a machine and one common way of conducting this task revolves around utilizations of neural networks. A neural network is essentially a single algorithm made up of many artificial neurons that attempts to mimic the biological neural networks that make up human brain such as certain neurons firing when someone distinguishes a verb from a noun. In this work,we explore the applications of a new type of neural network called a Capsule Network and how its implementation serves as an efficient and promising method of conducting several Natural Language Processing (NLP) tasks. In particular we will explore this neural network architecture as well as its performance on a simple Part of Speech tagging task to show its potential within future works.
Degree ProgramComputer Science