PENERAPAN MACHINE LEARNING DENGAN TENSORFLOW UNTUK MENDETEKSI BAHASA ISYARAT BISINDO BERBASIS APLIKASI ANDROID

DEWANGGA, TIARA and Firdaus, Masagus Afriyan (2022) PENERAPAN MACHINE LEARNING DENGAN TENSORFLOW UNTUK MENDETEKSI BAHASA ISYARAT BISINDO BERBASIS APLIKASI ANDROID. Undergraduate thesis, Sriwijaya University.

	Text RAMA_57201_09031381823095.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (2MB) | Request a copy
	Text RAMA_57201_09031381823095_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (4MB) | Request a copy
Preview	Text RAMA_57201_09031381823095_0012028203_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (717kB) | Preview
	Text RAMA_57201_09031381823095_0012028203_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (523kB) | Request a copy
	Text RAMA_57201_09031381823095_0012028203_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (505kB) | Request a copy
	Text RAMA_57201_09031381823095_0012028203_04.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (418kB) | Request a copy
	Text RAMA_57201_09031381823095_0012028203_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (5kB) | Request a copy
	Text RAMA_57201_09031381823095_0012028203_06_ref.pdf - Bibliography Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (239kB) | Request a copy
	Text RAMA_57201_09031381823095_0012028203_07_lamp.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy

Abstract

Deaf or hearing loss is a condition in which the person with hearing impairment has difficulty hearing sound perfectly, and even cannot hear at all. Due to the limitations of verbal communication, the hearing loss people communicate using sign language. Problems arise when the hearing loss people need to communicate with hearing people, while that person does not understand sign language, so that communication becomes hampered. Machine learning technology can be used to solve this problem. With Tensorflow, machine learning can be trained to recognize sign languages, especially alphabet with BISINDO (Indonesian Sign Language) standard. The output of the machine learning will then be inserted in the Android application so that it is practical to use. The method is Rapid Application Development with two development cycles. Machine learning training with transfer learning pre-trained model SSD MobileNet V2 produces a model which has 11MB in size and tflite extension that can specifically recognize the BISINDO alphabet with an accuracy of 89% in the dataset test and 93% in the black box test using an application. The Android application is called Auris with version 1.1 and has 51.8 MB in size with several features, such as sign language detection, dictionaries, and about. The score of usability testing is 75.75 which means the application is good and is feasible for users to use.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	Android, Bahasa Isyarat, Machine Learning, Tensorflow
Subjects:	Q Science > Q Science (General) > Q300-390 Cybernetics > Q325.5 Machine learning
Divisions:	09-Faculty of Computer Science > 57201-Information Systems (S1)
Depositing User:	Tiara Dewangga
Date Deposited:	24 Jun 2022 02:56
Last Modified:	24 Jun 2022 02:56
URI:	http://repository.unsri.ac.id/id/eprint/72784

Actions (login required)

View Item