MultiML: A Machine Learning Model Construction Tool Based on Multidimensional Views

This paper addresses the widespread impact of machine learning on diverse industries and the current high learning threshold in designing and training models. It proposes a no-code, data-driven tool for multidimensional machine learning construction, reducing the need for domain expertise and enhancing system usability. Through modular components and optimized interaction methods, the approach significantly lowers user entry barriers, improving model usability and construction efficiency. The survey results confirm its practical value and demonstrate its positive impact on the quality of systems, information, and cross-industry interactions.