KnowFIRES: A Knowledge-Graph Framework for Interpreting Retrieved Entities from Search

Negar Arabzadeh, Kiarash Golzadeh, Christopher Risi, Charles L. A. Clarke, Jian Zhao

Abstract: Entity retrieval is essential in information access domains where people search for specific entities, such as individuals, organizations, and places. While entity retrieval is an active research topic in Information Retrieval, it is necessary to explore the explainability and interpretability of them more extensively. KnowFIRES addresses this by offering a knowledge graph-based visual representation of entity retrieval results, focusing on contrasting different retrieval methods. KnowFIRES allows users to better understand these differences through the juxtaposition and superposition of retrieved sub-graphs. As part of our demo, we make KnowFIRES (Demo: http://knowfires.live, Source: https://github.com/kiarashgl/KnowFIRES) web interface and its source code publicly available (A demonstration of the tool: https://www.youtube.com/watch?v=9u-877ArNYE).