OmniScholar: An AI-Powered System for Researcher Profiling and Publication Analysis

Abstract

The exponential growth of scientific publications has made it increasingly challenging to monitor researcher contributions, assess impact, and align scholarly work with global priorities such as the United Nations Sustainable Development Goals (SDGs). Existing platforms like Google Scholar, ResearchGate, and Semantic Scholar host vast amounts of academic content but lack automation for extracting, filtering, and mapping research output to sustainability goals. Manual approaches are timeconsuming, fragmented, and insufficient for comprehensive profiling. This paper presents OmniScholar, an AI-assisted research intelligence system that automates the end-to-end process of data extraction, SDG classification, and profile visualization. Given an author’s name, the system retrieves publication metadata via the Semantic Scholar API, maps each paper to relevant SDGs through NLP-based keyword matching on abstracts, and produces a structured Excel report along with an interactive web visualization. OmniScholar delivers tangible results, retrieving almost all the manuscripts for a given author. Among the retrieved papers, around 50\% papers have successful abstract content. The results also include achieving $92 \%$ accuracy in keyword-based SDG mapping and reducing data profiling time by over 95\% compared to manual tracking. Its adaptability, lightweight design, and visualization capabilities make it a powerful tool for researchers, institutions, and policymakers seeking evidence-based insights and alignment with sustainability frameworks.

Ähnliche Arbeiten