Paper year
2024
Detect emerging, bridge-candidate, and undercited papers inside a curated audio-ML corpus, then expose the signals behind every recommendation.
Paper dossier
Review source metadata, abstract, authors, topics, and local similarity context before moving into explanation and ranking views.
Paper year
2024
Citations
2
Authors
5
Topic labels
3
Source readout
Journal of the Audio Engineering Society
jaes
Core corpus
6
Ranking readout
This block uses the same resolved ranking run as Recommended. Ranks here are materialized paper_scores ranks; live Emerging may be reordered by the bounded ML scorer. Family rank is global within each family, but rank is only shown when this paper lands inside the surfaced top 50.
Families present
3
Top 50
0
Run label
shadow-generalization-product-candidate-ranking-v1
Snapshot
source-snapshot-shadow-generalization-v1-20260521
Scope: family global | run rank-83787b91ef
Emerging
Present in run, outside top 50
Emerging: embedding slice fit vs included-corpus centroid (title+abstract), plus citation velocity and topic growth; not universal relevance. Bridge signal not used here.
Signals: semantic=0.8377, citation_velocity=0.0800, topic_growth=0.2000, diversity_penalty=0.0000
Embedding slice fit (corpus centroid): high; used in final ranking (contribution to score: 0.1675)
Recent attention: low; used in final ranking (contribution to score: 0.0400)
Topic momentum: low; used in final ranking (contribution to score: 0.0600)
Cross-cluster signal: not computed for this run
Similarity penalty: reduces score when non-zero (contribution to score: 0.0000)
Bridge
Present in run, outside top 50
Multi-topic paper in active topics; no cluster_version on this run so bridge_score was not computed.
Signals: citation_velocity=0.0800, topic_growth=0.2000, diversity_penalty=0.0000
Semantic match: not computed for this run
Recent attention: low; used in final ranking (contribution to score: 0.0280)
Topic momentum: low; used in final ranking (contribution to score: 0.1300)
Cross-cluster signal: not computed for this run
Topic breadth penalty: reduces score when non-zero (contribution to score: 0.0000)
Under-cited
Present in run, outside top 50
Low-cite candidate pool (see docs/candidate-pool-low-cite.md v0): core corpus, recency floor, citation ceiling, title+abstract gate; popularity penalty among pool members only. Semantic and bridge not yet modeled.
Signals: citation_velocity=0.0800, topic_growth=0.2000, diversity_penalty=0.4421
Semantic match: not computed for this run
Recent attention: low; used in final ranking (contribution to score: 0.0240)
Topic momentum: low; used in final ranking (contribution to score: 0.1400)
Cross-cluster signal: not computed for this run
Pool popularity penalty: reduces score when non-zero (contribution to score: -0.1105)
This study focuses on a corpus of 445 sonification projects currently available in the Data Sonification Archive (DSA). The DSA develops in a collaborative process that involves researchers and creative communities, and has been online since early 2021. Projects are heuristically classified according to several aspects, in particular their intended purpose, targeted users, subject matter, sonification method, and combination of media. In the present study, we analyse six curatorial classification strategies, labelled <i>Goal</i>, <i>Method</i>, <i>User,</i> <i>Macro Topic</i>, <i>Micro Topic</i>, and <i>MediaMix</i>, and discuss their definitions and usefulness for the archive. We then introduce two computational classification strategies, respectively based on clustering of music information retrieval of sonification audio, and topic modelling of the descriptive texts that accompany DSA projects. Correlation analysis between curatorial and computational classifications, correspondingly sized, showed that the text-based method was more powerful than the audio-based methods. We then explored predictive modelling, tentatively achieving results for <i>Goal, Method, and Macro Topic</i>. This points towards the potential for automatic classification to assist in the curatorial management of the archive, as well as for similar repositories. The discussion focuses on how analysis of classification strategies supports a broadening of the definition of sonification, both as theoretical construct and as practice, where the communicative intention of the author, the aesthetic quality of the listening experience, a more explicit focus on narrative patterns, and other emerging aspects within sonification design, are all contributing factors to transitioning the field towards a mass medium for data representation, communication, and meaning-making.
Neighborhood labels
Topic labels are imported metadata and can be noisy; use them as coarse navigation hints, not authoritative classifications.
BIM and Construction IntegrationDesign Education and PracticeTactile and Sensory Interactions
Neighbor surface
Similar papers use a separately configured neighbor embedding; it may differ from the embedding version used by the current ranked run.
Next handoff
01
Use Recommended to see whether this paper behaves like an emerging or undercited signal in the current ranked feed, or how it appears on the bridge preview / diagnostics view.
02
Use Trends to understand whether its attached labels are heating up or cooling down inside the curated corpus.
03
Use Evaluation to compare the dossier readout against citation and recency baselines for the same resolved family run.