Aurora: Neuro-Symbolic AI Driven Advising Agent
Authors:
- Lorena Amanda Quincoso LugonesLLLorena Amanda Quincoso Lugones
- Christopher Lukas KverneCKChristopher Lukas Kverne
- Nityam Sharadkumar BhimaniNBNityam Sharadkumar Bhimani
- Ana Carolina OliveiraAOAna Carolina Oliveira
- Agoritsa PolyzouAPAgoritsa Polyzou
- Christine LisettiCLChristine Lisetti
- Janki BhimaniJBJanki Bhimani
Open Access
- 23 March 2026
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 22-29
Abstract
Academic advising in higher education is under severe strain, with advisor-to-student ratios commonly exceeding 300:1. These structural bottlenecks limit timely access to guidance, increase the risk of delayed graduation, and contribute to inequities in student support. We introduce Aurora, a modular neuro-symbolic advising agent that unifies retrieval-augmented generation (RAG), symbolic reasoning, and normalized curricular databases to deliver policy-compliant, verifiable recommendations at scale. Aurora integrates three components: (i) a Boyce-Codd Normal Form (BCNF) catalog schema for consistent program rules, (ii) a Prolog engine for prerequisite and credit enforcement, and (iii) an instruction-tuned large language model for natural-language explanations of its recommendations. To assess performance, we design a structured evaluation suite spanning common and edge-case advising scenarios, including short-term scheduling, long-term roadmapping, skill-aligned pathways, and out-of-scope requests. Across this diverse set, Aurora improves semantic alignment with expert-crafted answers from 0.68 (Raw LLM baseline) to 0.93 (+36%), achieves perfect precision and recall in nearly half of in-scope cases, and consistently produces correct fallbacks for unanswerable prompts. On commodity hardware, Aurora averages 32.99 s per query (vs. 20.13 s for the Raw-LLM baseline), reflecting the runtime overhead of neuro-symbolic grounding. By combining symbolic rigor with neural fluency, Aurora advances a paradigm for accurate, explainable, and scalable AI-driven advising.Keywords
All Related Versions
Funding Information
- National Science Foundation (CSR-2402328)
- National Science Foundation (CAREER-2338457)
- National Science Foundation (CSR- 2406069)
- National Science Foundation (CSR-2323100)
- National Science Foundation (HRD-2225201)
This publication has 13 references indexed in Scilit:
- Large Language Model in Higher Education: Leveraging Llama2 for Effective Academic Advising2024
- Neurosymbolic AI: the 3rd waveArtificial Intelligence Review, 2023
- Engaging Students With a Chatbot-Based Academic Advising SystemInternational Journal of Human–Computer Interaction, 2022
- Development of Grammatical Aspect in Specific Language Impairment: Evidence from an Experimental Design with Video StimuliProcedia Computer Science, 2015
- Social Networks Mining for Analysis and Modeling Drugs UsageProcedia Computer Science, 2014
- Implementation of an Intelligent Course Advisory Expert SystemInternational Journal of Advanced Research in Artificial Intelligence, 2014
- SWI-PrologTheory and Practice of Logic Programming, 2011
- Design and development of a Web-based academic advising system2002
- A decision support system for academic advising1995
- Using an Expert System to Support Academic AdvisingJournal of Research on Computing in Education, 1992