PUREsuggest: Citation-based Literature Search and Visual Exploration with Keyword-controlled Rankings

TL;DR

PUREsuggest presents a citation-based literature search interface that combines explainable, glyph-based ranking with user-defined keyword control to steer discovery. It integrates seed-based exploration, a transparent scoring model, and augmented visualization of the citation network to support building and updating literature collections as well as identifying experts. Empirical evaluation through simulated sessions and a user study in information visualization demonstrates that keyword boosting improves recommendation quality and promotes author diversity, while highlighting limitations related to data coverage and sample size. The approach advances foraging-style literature discovery with a focus on transparency and user control, offering practical utility for researchers and students in rapidly evolving fields.

Abstract

Citations allow quickly identifying related research. If multiple publications are selected as seeds, specific suggestions for related literature can be made based on the number of incoming and outgoing citation links to this selection. Interactively adding recommended publications to the selection refines the next suggestion and incrementally builds a relevant collection of publications. Following this approach, the paper presents a search and foraging approach, PUREsuggest, which combines citation-based suggestions with augmented visualizations of the citation network. The focus and novelty of the approach is, first, the transparency of how the rankings are explained visually and, second, that the process can be steered through user-defined keywords, which reflect topics of interests. The system can be used to build new literature collections, to update and assess existing ones, as well as to use the collected literature for identifying relevant experts in the field. We evaluated the recommendation approach through simulated sessions and performed a user study investigating search strategies and usage patterns supported by the interface.

Figures (6)

• Figure 1: Dialog to search for publications based on a query (here: "citation visualization") or to add publications by providing DOIs.
• Figure 2: Glyph and tooltip explaining the score of a publication.
• Figure 3: The citation network visualization in full-screen, shown in cluster mode in default settings. The collection shows publications related to the visualization of clustering and classification, with two clear groups of publications forming around the respective keyword nodes; a central publication linking the two groups is highlighted, with an additional box showing the publication details.
• Figure 4: Ranked list of authors of selected publications (same as in \ref{['fig:cluster_vis']}). For each author, a glyph shows a score together with the author's initials and statistics; the summary below the author name characterizes the co-authored publications in terms of number, year range, matched keywords, and co-authors. A settings menu allows configuring the score.
• Figure 5: Alternative user interface layouts for different screen sizes: (left) ultra-wide, (right) mobile; empty selections as shown on start-up.