TY - CHAP
T1 - A Geo-Contextualized Multi-genre Scientific Search Engine
T2 - A Novel Conceptual Design and Prototype Evaluation
AU - Honeder, Johannes
AU - El Baff, Roxanne
AU - Hecking, Tobias
AU - Nussbaumer, Alexander
AU - Guetl, Christian
N1 - Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2026.
PY - 2025/10/1
Y1 - 2025/10/1
N2 - Scientific research is nowadays challenging due to the overwhelming access to information. Current scientific search tools have two shortcomings: (1) they lack a contextualized search framework where the user can search based on geolocation, and (2) they do not interconnect different text genres (e.g., publications and blogs). To tackle these issues, this paper presents a novel, generic conceptual design for a scientific search tool that applies to any domain. Our approach focuses on two main components within the search engine: (1) steering search results towards geolocation-oriented search where location expressions are extracted from the user query and used within the search process, and (2) enabling multi-genre text retrieval to increase synergy discovery. We built a prototype for the earth observation and science domain, where geolocation is salient. Our qualitative evaluations show the benefits of visually representing scientific research on map views, but they also reveal crucial points to improve our design further. The code is available here: https://zenodo.org/records/13788713, and the post-evaluation version is available here: https://zenodo.org/records/13789303.
AB - Scientific research is nowadays challenging due to the overwhelming access to information. Current scientific search tools have two shortcomings: (1) they lack a contextualized search framework where the user can search based on geolocation, and (2) they do not interconnect different text genres (e.g., publications and blogs). To tackle these issues, this paper presents a novel, generic conceptual design for a scientific search tool that applies to any domain. Our approach focuses on two main components within the search engine: (1) steering search results towards geolocation-oriented search where location expressions are extracted from the user query and used within the search process, and (2) enabling multi-genre text retrieval to increase synergy discovery. We built a prototype for the earth observation and science domain, where geolocation is salient. Our qualitative evaluations show the benefits of visually representing scientific research on map views, but they also reveal crucial points to improve our design further. The code is available here: https://zenodo.org/records/13788713, and the post-evaluation version is available here: https://zenodo.org/records/13789303.
KW - Open Web Search
KW - science search
KW - Specialized information retrieval
UR - https://www.scopus.com/pages/publications/105017956446
U2 - 10.1007/978-3-032-01005-6_9
DO - 10.1007/978-3-032-01005-6_9
M3 - Chapter
AN - SCOPUS:105017956446
T3 - Lecture Notes on Data Engineering and Communications Technologies
SP - 104
EP - 122
BT - Lecture Notes on Data Engineering and Communications Technologies
PB - Springer Science and Business Media Deutschland GmbH
ER -