Citation Proximity-Context Analysis for Interdisciplinary Common Knowledge Discovery

The academic barrier between disciplines can impede scientific communication and collaboration, resulting in the isolated development of common concepts and methodologies within independent fields. However, identifying and elucidating common knowledge across disciplines can enhance cross-field visibility, reduce redundant research efforts, as well as promote the diffusion and application of knowledge across domains. Therefore, this study proposes a methodology that leverages citation context to identify common knowledge among disciplines. By analyzing co-cited patterns in citation contexts drawn from existing cases of interdisciplinary common knowledge and extracting these patterns from a large-scale dataset, we designed evaluation metrics that consider commonality, interdisciplinarity, and novelty to identify valid, novel, potentially useful, and understandable interdisciplinary common knowledge. Eventually, 1,044 candidate pairs of interdisciplinary knowledge were identified. Among them, seven cases with remarkable interdisciplinary common novelty were analyzed. Each of these cases demonstrated the vital intersections in problem-solving approaches and fundamental methodologies, highlighting the potential for knowledge integration across different disciplines. The study underscores the challenges posed by disciplinary barriers in recognizing interdisciplinarity knowledge and contributes a systematic approach to fostering interdisciplinary communication of knowledge.