When a phone fails and its owner keeps the messaging account running on a second device, the account's history can end up in two archives, each intact and each missing what the other holds. We call this Temporal Communication Fragmentation (TCF), a loss by non-union: no message is deleted, but the platform offers no operation that joins the two halves. WhatsApp's own documentation states that old and new chats cannot be merged. We define TCF by three jointly necessary and sufficient conditions and specify CARTA 1.1, a framework for reconstructing one archive from several. CARTA is an output contract owed to a reader who cannot check the result. Records that cannot be read unambiguously are quarantined with a typed reason, and a report gives each record's provenance, the conflicts and deduplication ambiguities found, and every setting used. We evaluated a reference implementation, CHAMP-Pipeline, on synthetic fragment sets with known ground truth. With its declared date inference it reproduced the ground truth in all 4,500 configurations of a parameter sweep; the invariants and the predicted outcome held in all 5,000 random configurations of two to four fragments; and reconstruction stayed exact at one million messages. A zone-offset advisory, frozen before testing, flagged all 2,800 misdeclared zones in a designed test of overlapping fragment pairs and 1,470 of 1,542 in random configurations, and raised false alarms in 2 of 600 correctly declared runs. A systematic search found 60 chat-merging tools; none fully provides more than 2 of the five output properties CARTA specifies. With a fragment's time zone left undeclared, every benchmarked tool and CHAMP-Pipeline produce the same duplicated and misdated records, and only CHAMP-Pipeline's report says so. All data are synthetic; no real archive has been reconstructed.
This cross-sectional study examined how MBA students appraise ethical concerns, practical benefits, social expectations, control, and institutional governance related to educational artificial intelligence (AI). The analysis used 275 eligible responses submitted by the prespecified study cut-off. Descriptive statistics, omega and Spearman-Brown reliability, exploratory factor analysis (EFA), rank-based and bootstrap robustness checks, and regression models of ordered AI experience were used. Three exploratory dimensions were retained in 90.4% of 500 bootstrap samples. A proportional-odds reference model improved on the null model, chi-square(6) = 17.81, p = .007, but the proportional-odds assumption was rejected, chi-square(6) = 17.81, p = .007; threshold-specific models were therefore primary. AI self-efficacy/control was associated with reporting at least moderate rather than none/minimal AI experience (OR = 1.51, 95% CI [1.13, 2.03], p = .005), and this result persisted after background adjustment, bootstrap validation, and response-quality screens. In the smaller extensive-experience comparison (n = 26), efficiency benefit and perceived ethical-issue frequency were positive exploratory correlates. Concern was highest for data privacy, while governance responses indicated comparatively limited knowledge of personal-data practices. Predictive performance was modest. The findings support clearer task-specific guidance, critical AI literacy, and transparent data practices, but the cross-sectional self-report design does not establish causal adoption effects or validated moral profiles
This study examines the association between the reproductive policy environment and female labor force participation (FLFP) in Canada and Sri Lanka using panel data from 2013-2023. The analysis introduces a Reproductive Policy Restrictiveness Index (RPRI), a ten-component composite integrating three externally sourced legal variables (abortion legality, contraception access, and the World Bank Women, Business and the Law composite) with seven institutional indicators. The analytical approach in this work uses correlation analysis, Principal Component Analysis, Random Forest feature ranking, and ARIMA forecasting. Findings document a 55.7-point RPRI gap between Canada (mean = 90.6) and Sri Lanka (mean = 34.9), co-varying with a 13.4-percentage-point FLFP gap (pooled r = 0.996). Within-country temporal correlation is essentially zero, indicating the RPRI-FLFP association is between-country. The study contributes the first integrated legal-institutional restrictiveness index for these countries, framed associational rather than causally.
Management analytics is commonly evaluated with model-centric statistics such as the area under the receiver operating characteristic curve (ROC-AUC), yet managerial deployment converts scores into costly actions under finite capacity, changing base rates, and asymmetric economic consequences. This study introduces Decision-Adjusted Analytical Value (DAAV), a normalized framework that evaluates a predictive policy against random-capacity and attainable-oracle benchmarks after incorporating an economic action threshold and an operational capacity constraint. The framework is tested on the UCI Bank Marketing dataset using 41,188 chronologically ordered campaign records. To prevent post-contact leakage, call duration is excluded. Logistic regression, random forest, and Extreme Gradient Boosting (XGBoost) models are trained on the first 75% of observations, calibrated on the next 10%, and evaluated on the final 15% temporal holdout. Performance is assessed across 20 benefit-cost and capacity scenarios, with paired nonparametric bootstrapping for the focal policy comparison. The temporal holdout exhibits a pronounced prevalence shift, from 6.8% in validation to 38.5% in testing. Random forest attains the highest ROC-AUC (0.653), but is not the value-maximizing model in 16 of 20 operating scenarios. At a benefit-cost ratio of 3 and 50% capacity, logistic regression produces 226.9 normalized net-value units per 1,000 cases versus 95.6 for random forest; the paired difference is 131.0 units (95% bootstrap interval: 110.4 to 151.3). Static Platt calibration fitted before the shift further reduces downstream value despite leaving discrimination unchanged. The findings show that ranking models by predictive accuracy can misallocate managerial attention when economic thresholds and resource constraints bind. DAAV provides managers, analytics teams, and governance functions with a transparent, decision-centered complement to conventional predictive metrics for model selection and deployment under temporal shift.
Rapidly expanding metropolitan areas across Sub-Saharan Africa face systemic risk from hazard-induced disruptions, exacerbated by infrastructure deficits and socioeconomic vulnerabilities. Traditional risk assessment frameworks frequently fail decision-makers by offering static, qualitative narratives that cannot prioritize capital allocation or track resilience over dynamic budget cycles. This paper presents the Risk Mechanism Theory Index (RMTI) framework, a single, decision-ready mathematical model, grounded in Risk Mechanism Design and Theory, and built to assess, quantify, and treat systemic risk across organizational, health, infrastructure, and urban contexts. Structurally, the RMTI reduces every risk environment to three correctly signed dimensions drawn from Table 9.4: RMTre (risk exposure, L · E), RMTseen (socio-economic and environmental vulnerability, V), and RMTrl (resilience capacity, ρ), combined into a single bounded 0–1 composite residual score R = L · Esel · V · (1−ρ). The RMTI is operationalized through a purpose-built calculator that automates the full computational pipeline and is complemented by a Python reference implementation (RMTI-Py) for registry-scale and pipeline use. Both modalities are validated end-to-end against three World Bank-profiled urban infrastructure cases in Sub-Saharan Africa (Bamako, Mali; Lagos, Nigeria; Lilongwe, Malawi) and benchmarked against three comparator cities (Nairobi, Kenya; Dar es Salaam, Tanzania; Lusaka, Zambia). Empirical results demonstrate that targeted $50 million capital investments reduce composite risk scores by 30.7% to 47.5%, systematically transitioning all three primary environments from Tier 4 (Severe) to Tier 3 (High). The framework provides a transparent, algorithmic pipeline for embedding risk-informed governance into capital-budgeting workflows across sectors, and is offered as a management decision-support tool for institutional investors, development banks, and municipal finance officers seeking a reproducible, data-analytic alternative to narrative-based risk reporting.
