Sažetak
The integration of machine learning (ML) within the social sciences has emerged as
a transformative force, enabling researchers to uncover complex patterns and insights
from vast datasets. However, the effectiveness of machine learning applications in this
domain is heavily dependent on the selection of appropriate algorithms, a process that
can be intricate due to the diversity of available algorithms and their varying performance
across different contexts. This highlights the critical importance of algorithm
selection, which can significantly influence the outcomes of social science research. In
response to this challenge, we present idea of SIMON (Intelligent System for Automatic
Selection of Machine Learning Algorithms), a novel system designed to streamline the
algorithm selection process tailored specifically for social science applications. By leveraging
advanced methodologies, SIMON aims to empower researchers to make infor med
decisions regarding algorithm choice, thereby enhancing the robustness and relevance
of their findings. This paper explores the design and implications of the SIMON system,
alongside a literature review that contextualizes existing methods and identifies the
challenges faced in current algorithm selection approaches. Through this exploration,
we aim to contribute to the ongoing research surrounding the intersection of machine
learning and social sciences.
Ključne riječi
Machine learning intelligent systems meta-learning
