Unifying Local and Global Knowledge: Empowering Large Language Models as Political Experts with Knowledge Graphs

Large Language Models (LLMs) have revolutionized solutions for general natural language processing (NLP) tasks. However, deploying these models in specific domains still confronts challenges like hallucination. While existing knowledge graph retrieval-based approaches offer partial solutions, they can not be well adapted to the political domain. On the one hand, existing generic knowledge graphs lack vital political context, hindering deductions for practical tasks. On the other hand, the nature of political questions often renders the direct facts elusive, necessitating deeper aggregation and comprehension of retrieved evidence. To address these challenges, we present a Political Experts through Knowledge Graph Integration (PEG) framework. PEG entails the creation and utilization of a multi-view political knowledge graph (MVPKG), which integrates U.S. legislative, election, and diplomatic data, as well as conceptual knowledge from Wikidata. With MVPKG as its foundation, PEG enhances existing methods through knowledge acquisition, aggregation, and injection. This process begins with refining evidence through semantic filtering, followed by its aggregation into global knowledge via implicit or explicit methods. The integrated knowledge is then employed in LLMs through prompts. Experiments on three real-world datasets across diverse LLMs reaffirm PEG's superiority in tackling political modeling tasks. CCS CONCEPTS • Computing methodologies → Natural language processing; • Human-centered computing → Collaborative and social computing.