Discovering Associations Among Technologies Using Modified Term Frequency Inverse Document Frequency

Abstract

‎The development of novel and efficient technologies increasingly depends on a comprehensive understanding of the relationships among various existing technologies‎. ‎In this study‎, ‎we propose a parameterized model to identify and visualize associations between technologies using an enhanced Term Frequency-Inverse Document Frequency (TF-IDF) approach applied to diverse textual data‎, ‎such as academic publications‎. ‎The discovered associations are represented in a weighted graph structure‎, ‎referred to as the \textit{association tech-graph}‎, ‎where nodes denote individual technologies and edge weights quantify their degrees of interconnection‎. ‎The model introduces parameter symmetry to regulate graph construction‎, ‎allowing for the emergence of symmetric patterns in the relationships based on parameter selection‎. ‎Our analysis reveals that exploiting the symmetry in associations such as between Aerial Robotics and Advanced Driver Assistance Systems or between Actuators and Adjustable Hoisting Machines can significantly enhance the design of autonomous systems and increase efficiency in industrial applications‎. ‎While the inclusion of parameterized symmetry elevates computational complexity‎, ‎it also leads to more accurate and interpretable representations of technological interrelations‎. ‎The study demonstrates how the interplay of technological components‎, ‎guided by structured symmetry‎, ‎supports innovation in complex industrial systems‎.