Pengenalan Bahasa Isyarat Indonesia menggunakan Mediapipe dengan Model Random Forest dan Multinomial Logistic Regression

Abstract

Purpose: This research aims to create a random forest machine learning model and logistic regression that can perform the sign language recognition of the Indonesian Sign Language System (SIBI) using a regular RGB camera with the MediaPipe framework. Research methodology: Both variables in this study are measured using Innovative Work Behavior (IWB) Scale from Janssen (2000) and Connor-Davidson Resilience Scale (CD-RISC) from Connor & Davidson (2003) that was distributed through Google Form link. The data analysis is done with the support of the 25th version of SPSS (Statistical Package for Social Science). Results: Resilience has a significant correlation with innovative work behavior among college students. Limitations: No strict controls of questionnaire administration, the questionnaire consists of 6 different measurements from the research team, and can't be fully generalized to the college students population. Contribution: New findings of correlation between two variables among new samples.