*3.1. Machine Learning Algorithms*

There are many platforms and learning libraries that can be used to implement different ML algorithms, many of which circumvent the need to write codes. However, in this section, we present only a summary of each ML algorithm as a basis for understanding how they work. According to the meta-analysis of the recent literature provided in Table 1, the underlying ML algorithms include the regression and artificial neural network-based approaches. As a result, we considered these foundational techniques in our research because, in addition to being simple, they use fewer computational and memory resources than the more recent deep learning approaches. Furthermore, in the aftermath of new smart grid applications, which are based on the Internet of Things (IoT), it is critical to consider these simpler methods due to the limited processing and memory capacities of many IoT-based devices, which justifies the inclusion of the methods discussed below in our study.
