Digital Library

cab1

 
Title:      USING MOBILE TECHNOLOGY TO CROWDSOURCE THE AUGMENTATION OF DEEP LEARNING DATASETS
Author(s):      Chantelle Saliba and Dylan Seychell
ISBN:      978-989-8704-16-0
Editors:      Inmaculada Arnedillo Sánchez, Pedro Isaías and Boyan Bontchev
Year:      2020
Edition:      Single
Keywords:      Mobile Technology, Research, Data, Crowdsourcing, Machine Learning
Type:      Full
First Page:      3
Last Page:      10
Language:      English
Cover:      cover          
Full Contents:      click to dowload Download
Paper Abstract:      In the past decade, mobile communications have seen drastic changes and improvements with an estimate of over 3.5 billion mobile phone users worldwide. In addition, the average mobile phone has gone from being a simple communication device to a smartphone device capable of web browsing, video conferencing, gaming, photography, and videography and intelligent applications. For this reason, companies and industries have been exploring this technology to create opportunities to enhance their communications with clients and to create further business opportunities. In this research, we analyze the approach of using mobile technologies as a technique to crowdsource data that would be used to enhance research by creating digital resources. In today’s modern and technological world there are areas and fields which are still unexplored by technology due to their lack of digital resources. Modern machine learning techniques such as deep learning methods, require a large volume of data that is not always available. Such a case is the example of classifying Maltese flora. Malta is a small island in the middle of the Mediterranean with an area of 316 km2. Being such a small island with unique and indigenous flora makes it a challenging feat to find already available digital data to be able to conduct technological research. For this reason, we turn to mobile technology and how this can aid in the collection of such data to augment existing datasets that enhance academic research and render classification more effective and feasible.
   

Social Media Links

Search

Login