Dharma-Centric Value Network: A Knowledge Graph Approach to Indian Ethical Values
摘要
This study develops a Dharma-centric Value Network (DVN) that organizes Indian ethical values from Sanskrit literature into a transparent, ontology-driven knowledge graph. Primarily, the corpus is lexically anchored in Śabdakalpadruma and Vācaspatya. Next, entries are traced across the Bhagavadgītā, Manusmṛti, Mahābhārata, and selected Purāṇas to secure scriptural grounding. Then, classical commentaries are consulted to resolve semantic nuances. Using qualitative content analysis with an ontology-driven, lexical–hermeneutic approach, the study codifies 121 values and maps 235 relations across nine relationship types: avayava (an aspect of), kāraṇa (a cause of), rūpa (a form of), samārthaka (a synonym), lābhasya upāya (a way of achieving), abhāva (absence of), mūla (a source of), phala (a result of), and viruddhabhāva (an opposite of). The results are presented as a human and machine-readable coded table and a knowledge graph model designed to be machine-actionable, directly portable to formal knowledge-representation standards in future work. The DVN links virtues together with their vices within the Indian worldview of bhāva. It aligns with India’s National Education Policy (NEP) 2020 and supports curriculum design, teacher resources, and AI-assisted ethical reasoning while retaining scriptural fidelity. In comparative perspective, it complements Value Network Analysis (VNA) and Living Values Education (LVEP) by supplying a scripturally anchored moral ontology rather than a flow map or fixed value list. Current limitations include the focus on Hindu sources; Buddhist and Jain traditions are planned for subsequent phases.