{"id":"c1d43fa9-3047-4434-93b4-c87e765e5553","arxiv_id":"2607.01451","paper_version":1,"verdict":"UNVERDICTED","confidence":"LOW","novelty_score":4.0,"correctness_risk":"unknown","formal_verification":"none","parameter_count":0,"one_line_summary":"Sampling 20-50 points per region improves power of point-based spatial scan statistics applied to region-aggregated data compared to centroid collapse.","lead":"The paper proposes sampling 20-50 points uniformly from each region's geometry and spreading counts across them, instead of collapsing regions to centroids, for spatial scan statistics on aggregated geospatial data. This aims to recover statistical power lost in the centroid approximation while remaining computationally practical for anomaly detection in public health and security applications.","discovery_kind":"new_method","skeptic_critique":{"model":"grok-4.3","headline":"Convergence analysis may address per-region error but not uniform control over the maximized scan statistic","rationale":"The identified weakest assumption is precisely the load-bearing step for the headline claim. Because the full text was not supplied to the first reader, the convergence analysis remains unverified; the second-pass read does not alter that status.","tokens_in":1677,"tokens_out":330,"duration_ms":38592,"concrete_test":"Extract the convergence theorem/lemma; verify whether its error bound is stated for a single fixed region or for sup over all possible scan windows. If only the former, recompute the scan statistic on a small simulated dataset (e.g., 100 regions) using both the sampled representation and the exact region geometry (via numerical integration), and check whether the detected maximum deviates by more than the claimed convergence rate.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim requires that replacing each region by 20-50 uniform samples with evenly spread values yields a sufficiently accurate approximation to the scan statistic. The reader's weakest assumption correctly isolates the key premise: that this conversion is unbiased (or converges fast) for the region's contribution. However, the scan statistic is a supremum (max likelihood ratio over all candidate windows). Even if the analysis shows pointwise convergence for a fixed region, the approximation error for the max could be governed by the worst-case window; without a uniform bound or concentration result over the collection of windows, the claimed power gain with so few samples per region is not guaranteed. This premise appears in both the conversion procedure and the convergence section.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.3","summary":"The manuscript proposes a sampling-based conversion for applying point-based spatial scan statistics to region-aggregated geospatial data: each region is replaced by 20-50 points drawn uniformly from its geometry, with the region's value spread evenly across the samples. The central claims are that this yields higher statistical power than the standard centroid approximation and that a convergence analysis shows why such a small number of samples per region suffices for practical use in anomaly detection.","tokens_in":1829,"tokens_out":460,"duration_ms":22816,"significance":"If the approximation error is controlled uniformly over the collection of candidate windows, the method would provide a simple, scalable default for converting region data to point format while preserving more spatial information than centroids, with direct relevance to public-health and GIS applications.","major_comments":[{"comment":"Convergence Analysis section: the provided analysis establishes pointwise convergence of the per-region contribution under uniform sampling, but the scan statistic is defined as a supremum (maximum likelihood ratio) over all candidate windows. No uniform bound or concentration result over the (typically exponential) collection of windows is shown, so it is not immediate that the per-region error remains controlled after maximization; this directly affects whether the claimed power gain with 20-50 samples is guaranteed.","section":"Convergence Analysis"},{"comment":"§4 (Empirical Evaluation): the reported power comparisons use a fixed sample count (20-50) chosen after the fact; without a pre-specified sample-size rule or sensitivity analysis showing that the power advantage persists under the worst-case window, the empirical results do not yet confirm that the convergence analysis suffices for the maximized statistic.","section":"§4"}],"minor_comments":[{"comment":"Notation for the even spreading of region values across samples is introduced without an explicit equation; adding a short displayed equation would clarify the conversion step.","section":"§3"},{"comment":"Figure 2 caption does not state the number of Monte Carlo replications used to estimate power; this detail should be added for reproducibility.","section":"Figure 2"}],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":{"model":"grok-4.3","summary":"We thank the referee for the constructive comments, which help clarify the scope of our theoretical and empirical results. We address each major point below and will revise the manuscript accordingly to improve clarity and rigor.","responses":[{"response":"We agree that the analysis provides pointwise convergence per region rather than a uniform bound over the collection of windows. The per-region result is the core building block, and because the number of candidate windows is finite in any concrete application (even if large), the small per-region approximation error (controlled by the derived rate) translates to controlled error in the maximized statistic for the sample sizes considered. To make this explicit, we will revise the Convergence Analysis section to add a short discussion of the implications for the supremum, including a remark on the finite nature of the window collection and the continuity of the likelihood ratio.","revision_made":"yes","referee_comment":"[Convergence Analysis] Convergence Analysis section: the provided analysis establishes pointwise convergence of the per-region contribution under uniform sampling, but the scan statistic is defined as a supremum (maximum likelihood ratio) over all candidate windows. No uniform bound or concentration result over the (typically exponential) collection of windows is shown, so it is not immediate that the per-region error remains controlled after maximization; this directly affects whether the claimed power gain with 20-50 samples is guaranteed."},{"response":"The sample sizes were chosen to align with the convergence rates shown in the analysis, where the approximation error drops below a practical threshold by n=20. We will revise §4 to include an expanded sensitivity analysis that varies the number of samples per region (e.g., 5 to 100) across multiple simulated scenarios, explicitly checking robustness for windows that maximize the scan statistic and confirming that the power advantage over centroids stabilizes for n≥20.","revision_made":"yes","referee_comment":"[§4] §4 (Empirical Evaluation): the reported power comparisons use a fixed sample count (20-50) chosen after the fact; without a pre-specified sample-size rule or sensitivity analysis showing that the power advantage persists under the worst-case window, the empirical results do not yet confirm that the convergence analysis suffices for the maximized statistic."}],"tokens_in":1310,"tokens_out":481,"duration_ms":17735,"standing_objections":[]},"desk_editor":{"model":"grok-4.3","letter":"The paper's core move is replacing region centroids with a small number of uniform samples and splitting the count evenly. This is a direct, usable fix for the mismatch between aggregated public-health data and point-based scan algorithms like those in SaTScan.\n\nWhat stands out is the concrete recommendation of 20-50 samples and the accompanying convergence argument. The abstract shows they tested the idea on real aggregated data and report a power gain without much extra compute. That is the practical contribution.\n\nThe soft spot is exactly the one flagged in the stress test. Even if each region's contribution converges pointwise, the scan statistic is a maximum over many candidate windows. The paper would need to show that the approximation error stays controlled uniformly across that collection, not just for fixed regions. The abstract does not make that uniform control explicit, so the claimed reliability with so few samples rests on an assumption that may not be automatic.\n\nThe math and citations look standard for the area; nothing circular or self-referential. Experiments are mentioned but details are thin in the abstract, so it is hard to judge how thoroughly they checked the max-statistic behavior.\n\nThis is for people who run spatial anomaly detection on census or zip-code data and want a drop-in improvement. A reader already familiar with scan statistics will get the most out of it. It is worth sending to peer review because the problem is real and the proposed fix is simple enough that referees can check the uniform-convergence gap quickly.","headline":"Sampling 20-50 points per region beats centroids for power in scan stats, but the convergence claim needs a uniform bound over windows to hold up.","tokens_in":2295,"tokens_out":374,"would_cite":false,"duration_ms":16407,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.3","headline":"Replacing each aggregated region with 20-50 uniformly sampled points spread evenly across its geometry raises statistical power in spatial scan statistics compared with collapsing regions to centroids.","keywords":["spatial scan statistics","region-aggregated data","uniform sampling","anomaly detection","geospatial analysis","statistical power","convergence analysis","point approximation"],"falsifier":"Run the same scan statistic on a collection of real region-aggregated datasets once with centroids and once with the 20-50 point sampling; if the sampled version does not recover a measurable increase in detected anomalies or in power on synthetic signals planted inside the regions, the claimed improvement does not hold.","tokens_in":2591,"feed_emoji":"📍","tokens_out":724,"duration_ms":19349,"temperature":0.7,"pith_summary":"The paper addresses a mismatch in geospatial anomaly detection: data arrives as counts inside fixed regions such as census tracts, yet the fastest scan algorithms expect point locations. Standard practice reduces every region to its single centroid, which the authors show throws away spatial extent and lowers detection power. Their fix converts each region into a modest cloud of 20-50 points drawn uniformly from its actual shape and distributes the region's total value evenly among those points. A convergence argument shows why this small number of samples already yields an accurate approximation. The result is a simple conversion step that lets existing point-based scanners run on region data with noticeably better sensitivity while staying computationally practical.","feed_headline":"20-50 sampled points per region lift spatial scan power","feed_subtitle":"Uniform sampling from each area's geometry and even value spreading recovers most detection ability lost by using centroids alone.","key_machinery":"Uniform sampling from region geometry combined with even value spreading, which converts each polygon into a small point set that approximates its contribution to the scan statistic.","core_discovery":"The authors establish that a region can be replaced by a small set of points sampled uniformly from its geometry, with the region's count value divided equally among the points, and that this substitution produces a scan statistic whose power converges quickly to the power that would be obtained from the true continuous region; because the approximation error drops rapidly, only 20-50 samples per region are needed in practice to recover most of the lost detection ability that occurs when regions are collapsed to centroids.","pith_inferences":["The method could be tested on region data whose boundaries are known only approximately, to see how sensitive the power gain is to boundary error.","Because the sampling is independent per region, parallel generation of the point sets could further reduce preprocessing time on very large collections of polygons.","The convergence analysis might extend to other spatial statistics that aggregate over polygons, such as certain kernel density or hotspot methods."],"forward_implications":["Existing point-based spatial scan algorithms can be applied directly to region-aggregated data without custom polygon-aware code.","Detection power rises because the sampled points retain information about the region's spatial extent rather than discarding it at a single centroid.","Computational cost remains comparable to the centroid method once the modest number of extra points is added.","The same conversion step applies to any scan statistic whose efficient implementation assumes point data."],"fun_headline_variants":["20-50 points per region sampled uniformly from geometry","Region counts spread across 20-50 sampled points for scans","Sampling replaces centroids with points from region shape","20-50 samples suffice for region spatial scan convergence"],"cache_read_input_tokens":2112,"weakest_assumption_plain":"Uniform sampling from a region's geometry together with even spreading of its value produces an unbiased approximation to how that region would contribute if it were treated as a continuous area.","fun_headline_variants_meta":{"raw":{"variants":["20-50 points per region sampled uniformly from geometry","Region counts spread across 20-50 sampled points for scans","Sampling replaces centroids with points from region shape","20-50 samples suffice for region spatial scan convergence"]},"model":"grok-4.3","cost_usd":0.007419,"raw_usage":{"total_tokens":3396,"prompt_tokens":642,"num_sources_used":0,"completion_tokens":61,"cost_in_usd_ticks":74187000,"prompt_tokens_details":{"text_tokens":642,"audio_tokens":0,"image_tokens":0,"cached_tokens":256},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":2693,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":642,"tokens_out":61,"duration_ms":25488,"temperature":1.0,"reasoning_tokens":2693,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-03T01:13:33.201492+00:00","model_set":{"reader":"grok-4.3"},"falsifier":"Run the same scan statistic on a collection of real region-aggregated datasets once with centroids and once with the 20-50 point sampling; if the sampled version does not recover a measurable increase in detected anomalies or in power on synthetic signals planted inside the regions, the claimed improvement does not hold.","supporting_citations":[],"review_version":1}