Preliminary Proceedings of the 70th International Academic Conference, Madrid

NETWORK ANALYSIS AND CENTRALITY IN DATA SCIENCE

ELISABED ASABASHVILI, TAMAR TEVZADZE

Abstract:

Data science is one of the fastest-growing interdisciplinary fields, traditionally taught and practiced using dominant programming languages such as Python and R. However, the growing emphasis on pedagogy, conceptual clarity, and rapid prototyping raises the question of whether alternative languages can effectively support data science education. This paper presents a novel, education-focused exploration of Ruby as a viable language for illustrating foundational data science concepts without reliance on specialized external libraries. Using social network analysis as a unifying case study, we demonstrate how degree centrality, friends-of-friends exploration, and interest-based similarity can be implemented transparently using Ruby’s native data structures. The novelty of this work lies not in proposing new analytical algorithms, but in reframing data science instruction through a language that prioritizes readability and conceptual explicitness. By exposing the underlying mechanics of graph construction and similarity computation, the proposed approach helps learners focus on core analytical reasoning rather than library-specific abstractions. The results indicate that Ruby offers a clear and accessible environment for teaching network-based data science fundamentals and for rapid exploratory analysis, complementing established Python-based workflows.

Keywords: Data science education, Ruby programming, social network analysis, degree centrality, graph modeling.

PDF: Download