Implementation of Machine Learning Using Google's Teachable Machine Based on Android
Diki Agustian, Pande Putu Gede Putra Pertama, Padma Nyoman Crisnapati, Putu Devi Novayanti
Abstract
Many people think that machine learning is difficult to build and do not know how to implement it. This is one of the problems in writing this report. In this study, the system development method used is the waterfall which starts from the requirements analysis stage, system design, system implementation, system testing and system maintenance. The design of this system uses flowcharts, Unified Modeling Languages (UML) and interface design. In this study, it is known that Android devices can be implemented with a machine learning model. Each model in this study has been tested for models and each model has an accuracy, precision and sensitivity ranging from 97-100%. Google's Teachable Machine can create a machine learning model with a test accuracy rate of up to 100%. The level of accuracy can be reduced according to the level of lighting at the detection site.