Peningkatan Efisiensi dan Transparan Layanan Jasa Tukang melalui Aplikasi Berbasis Machine Learning

Abstract

The increasing rate of home ownership in Indonesia has created a growing demand for construction services that are transparent, efficient, and reliable, posing challenges for consumers in finding skilled workers. This study formulates the problem of how perceived usefulness and perceived ease of use, facilitated by Machine Learning-based features, influence user acceptance of the Kanggo application. The objective is to analyze the effects of Perceived Ease of Use and Perceived Usefulness on attitude, intention, and actual use within the framework of the Technology Acceptance Model (TAM). A quantitative method employing Partial Least Squares Structural Equation Modeling (PLS-SEM) was applied to survey data from Kanggo users. The findings reveal that all hypotheses are supported, with significant relationships among latent variables, confirming that ease of use and perceived benefits drive user intention and actual application usage. Recommendations include strengthening recommendation and price prediction features to enhance user adoption.