Enhancing SDG Text Classification Through Combinatorial Fusion Analysis
Abstract
The goal of this master’s thesis is to address the challenges of Sustainable Development Goals (SDGs) text classification using Combinatorial Fusion Analysis (CFA). CFA provides methods and workflow of combining multiple scoring systems. It characterizes each scoring system using a rank-score function and measures the diversity between two scoring systems. Human experts are subjective and no single machine learning algorithm can always correctly identify labels for documents. SDGs have many overlapping areas, which makes classification process challenging. CFA is thus used to combine three computational models each with different methodologies. We use CFA to develop a metric in arranging documents into three categories and SDGs into three groups. Finally, we compare the results of CFA with those of human experts to enhance SDG classification process.
Subject Area
Computer science|Artificial intelligence|Information science
Recommended Citation
Xu, Jingyan, "Enhancing SDG Text Classification Through Combinatorial Fusion Analysis" (2024). ETD Collection for Fordham University. AAI31298963.
https://research.library.fordham.edu/dissertations/AAI31298963
