{"schema_version":1,"experiment_id":"eoa_inverse_innovation_exp04_retrieval_first_paired20_20260802","cell_id":"invariant_mode_decomposition_design__library_information_science","hypothesis_id":"H5","search_queries":["library authority control automated change detection network analysis controlled vocabularies","knowledge organization system vocabulary evolution change detection thesaurus maintenance research","dynamic graph spectral change detection eigenvector subspace rotation spectral gap monitoring","semantic drift detection temporal word graphs spectral methods","patent spectral graph change detection eigenvector drift monitoring network","authority control software automated authority record maintenance change reports product","ISO 25964 thesaurus maintenance review terms changes official standard","Library of Congress authority file weekly changes distribution authority records"],"sources":[{"source_id":"H5-C1","title":"ISO 25964-1:2011 — Information and documentation — Thesauri and interoperability with other vocabularies — Part 1","publisher":"International Organization for Standardization","url":"https://www.iso.org/standard/53657.html","source_class":"STANDARD","claims_supported":["Development and maintenance of information-retrieval thesauri is standardized professional practice.","The standard supplies a thesaurus data model and anticipates thesaurus-management requirements, but its public abstract does not prescribe spectral monitoring."]},{"source_id":"H5-C2","title":"The NACO FTP process","publisher":"Library of Congress, Program for Cooperative Cataloging","url":"https://www.loc.gov/aba/pcc/naco/nodes.html","source_class":"OFFICIAL_GUIDANCE","claims_supported":["The LC name-authority workflow processes daily files of created, updated, and deleted records.","Authority-record changes have transaction timestamps, version checks, error responses, and distribution histories that could support longitudinal evaluation labels.","The documented workflow manages submitted changes and concurrency rather than predicting future correction clusters from graph modes."]},{"source_id":"H5-C3","title":"Represent Changes of Knowledge Organization Systems on the Semantic Web","publisher":"International Journal of Librarianship","url":"https://journal.calaijol.org/index.php/ijol/article/view/64","source_class":"PRIMARY_RESEARCH","claims_supported":["Change, version control, and evolution of thesauri, subject-heading systems, and name authorities are recognized KOS-management problems.","The work focuses on representing changes with SKOS/SKOS-XL, not detecting latent structural drift or issuing predictive review alarms."]},{"source_id":"H5-C4","title":"Observing trajectories of KOSs Across Space and Time: The DANS KOS Observatory (KOSo)","publisher":"North American Symposium on Knowledge Organization","url":"https://journals.lib.washington.edu/index.php/nasko/article/view/15619","source_class":"PRIMARY_RESEARCH","claims_supported":["The KOS Observatory was created to study how knowledge-organization systems change over time and how their versions can be identified, controlled, and archived.","Longitudinal observation of KOS trajectories is therefore prior practice, but the opened source does not describe eigenvector rotation, spectral-gap alarms, or correction prediction."]},{"source_id":"H5-C5","title":"Change Detection in Noisy Dynamic Networks: A Spectral Embedding Approach","publisher":"arXiv","url":"https://arxiv.org/abs/1910.02301","source_class":"PRIMARY_RESEARCH","claims_supported":["Dynamic networks are monitored as graph sequences using spectral embeddings and alignment across time.","The CDP method detects vertex-level changes in degree, community membership, and edge weight and compares performance against spectral baselines.","This substantially overlaps term-level structural-change localization, although its evaluated domains are not governed authority files."]},{"source_id":"H5-C6","title":"Profile Monitoring via Eigenvector Perturbation","publisher":"arXiv","url":"https://arxiv.org/abs/2205.15422","source_class":"PRIMARY_RESEARCH","claims_supported":["Eigenvector perturbation has been used to construct sequential control-chart alarms with explicitly calibrated false-alarm and detection-delay behavior.","The paper establishes thresholded eigenvector-based monitoring as a neighboring-domain mechanism, but not for semantic graphs or authority maintenance."]},{"source_id":"H5-C7","title":"Eigenspace-based Anomaly Detection in Computer Systems","publisher":"IBM Research / ACM SIGKDD","url":"https://ide-research.net/papers/2004_KDD_Ide_p140.pdf","source_class":"PRIMARY_RESEARCH","claims_supported":["The method models an evolving system as a time sequence of weighted graphs, extracts principal eigenvector directions, compares their time-series behavior, and adaptively updates an anomaly threshold.","It automatically raises fault indications and localizes implicated nodes where individual surface observations are inconclusive.","This is a close mechanism-level predecessor to dominant-mode-rotation alarms, differing chiefly in application domain and evaluation outcome."]},{"source_id":"H5-C8","title":"Graph-based Clustering for Detecting Semantic Change Across Time and Languages","publisher":"Association for Computational Linguistics","url":"https://aclanthology.org/2024.eacl-long.93/","source_class":"PRIMARY_RESEARCH","claims_supported":["Graph-based analysis already detects nuanced semantic changes, including acquisition and loss of word senses across time.","The method was benchmarked against a semantic-change classification task, bringing temporal graph detection into the semantic-vocabulary neighborhood.","It does not evaluate governed authority-record corrections or spectral-gap and eigenspace-rotation alarms."]}],"proximity":"SUBSTANTIAL_COLLISION","closest_analogues":[{"name":"Eigenspace-based anomaly detection from graph sequences","similarity":"It already converts successive weighted graphs into principal-eigenvector directions, monitors directional departures with an adaptive threshold, raises automated alarms, and localizes affected nodes despite noisy surface activity.","remaining_difference":"Its nodes are computer services and its target is contemporaneous faults, not authority terms and prospective clusters of governed corrections; it also does not center spectral-gap collapse.","source_ids":["H5-C7"]},{"name":"Spectral embedding change detection using Procrustes alignment","similarity":"It compares spectral representations of successive graphs and detects entity-level structural changes, including community and edge-weight changes, while benchmarking against alternative spectral methods.","remaining_difference":"It detects changes in generic dynamic networks rather than using authority-reference and co-assignment graphs to allocate human vocabulary-review capacity or predict later corrections.","source_ids":["H5-C5"]},{"name":"Graph-based lexical semantic-change detection","similarity":"It applies temporal graph structure to identify changing meanings, including low-frequency changes that simpler measurements can miss.","remaining_difference":"It targets corpus word senses and semantic-change benchmarks, not controlled authority concepts, spectral separation, governed alerts, or maintenance outcomes.","source_ids":["H5-C8"]},{"name":"Longitudinal KOS change and version governance","similarity":"KOS evolution, version control, observation across time, and formal thesaurus maintenance are established concerns and practices in library and information science.","remaining_difference":"The opened KOS and authority-management sources record or govern explicit changes after they occur; they do not disclose modal-rotation or spectral-gap alarms that predict correction clusters.","source_ids":["H5-C1","H5-C2","H5-C3","H5-C4"]}],"overlapping_components":["Successive graph or network snapshots","Weighted adjacency or relational representation","Spectral embedding and dominant eigenvectors","Comparison of eigendirections or latent graph coordinates over time","Thresholded anomaly or control alarms","Entity-level localization of structural change","Longitudinal KOS version and change management","Human-governed authority-file maintenance"],"remaining_contrastive_claim":"In a governed authority-file workflow, dominant-subspace rotation or spectral-gap collapse in successive term-reference and co-assignment graphs can predict later clusters of approved authority corrections better than edit-volume and term-frequency baselines at equal review capacity.","claim_falsifier":"The contrastive claim would be falsified as a prior-art distinction by a pre-existing paper, patent, product, or documented deployment that builds longitudinal authority-reference and co-assignment graphs, alarms on eigenspace rotation or spectral-gap thresholds, and demonstrates capacity-matched prediction of subsequent authority-correction clusters against edit-volume or term-frequency baselines.","problem_support":"MODERATE","recommendation":"RESEARCH","world_novelty_boundary":"This bounded search found substantial prior art for every generic mechanism—longitudinal KOS maintenance, spectral comparison of evolving graphs, principal-eigenvector direction alarms, calibrated eigenvector-based control charts, and graph-based semantic-change detection—but no opened source established their specific use for prospective, capacity-matched authority-correction review. The remaining distinction is therefore an application-and-validation claim, not a new spectral-monitoring mechanism; absence from these eight principal queries and eight opened sources is not proof of worldwide novelty."}