Evolution of ESG-focused DLT Research: An NLP Analysis of the Literature

2023-08-23

arXiv.org

AI Summary

Existing research on distributed ledger technology (DLT) lacks systematic, multidimensional analysis of its impacts across environmental, social, and governance (ESG) dimensions. Method: We propose the first NLP-driven systematic literature review (SLR) framework for the DLT–ESG intersection, leveraging a corpus of 24,000 papers. Our approach integrates domain-adapted named entity recognition (NER; F1 = 92.3%), directed citation network evolution analysis, and Transformer fine-tuning to identify 505 high-impact studies. Contribution/Results: We introduce a scalable SLR methodology, release the first DLT–ESG–specific NER dataset (54,808 annotated entities), and deliver the field’s inaugural temporal SLR—revealing distinct evolutionary trajectories across ESG dimensions. This work establishes an empirical foundation for sustainable technology governance and evidence-based policymaking.

Abstract

As Distributed Ledger Technologies (DLTs) rapidly evolve, their impacts extend beyond technology, influencing environmental and societal aspects. This evolution has increased publications, making manual literature analysis increasingly challenging. We address this with a Natural Language Processing (NLP)-based systematic literature review method to explore the intersection of Distributed Ledger Technology (DLT) with its Environmental, Social, and Governance (ESG) aspects. Our approach involves building and refining a directed citation network from 107 seed papers to a corpus of 24,539 publications and fine-tuning a transformer-based language model for Named Entity Recognition (NER) on DLT and ESG domains. Applying this model, we distilled the corpus to 505 key publications, enabling an inaugural literature review and temporal graph analysis of DLT's evolution in ESG contexts. Our contributions include an adaptable and scalable NLP-driven systematic literature review methodology and a unique NER dataset of 54,808 entities, tailored for DLT and ESG research. Our inaugural literature review demonstrates their applicability and effectiveness in analyzing DLT's evolution and impacts, proving invaluable for stakeholders in the DLT domain.

Problem

• Distributed Ledger Technology
• Environmental Social Governance
• Proof of Work

Innovation

• Natural Language Processing (NLP)
• Time Series Analysis
• Distributed Ledger Technology (DLT) Database