Incomplete Data Sources Research Articles

Estimation of dynamic Origin–Destination matrices in a railway transportation network integrating ticket sales and passenger count data

Accurately estimating Origin–Destination matrices is a pressing challenge in transportation management and urban planning. However, traditional methods like travel surveys have limitations in availability and comprehensiveness, which have been further exacerbated by the recent changes in mobility patterns induced by the COVID-19 pandemic. To address this issue, we focused on the Trenord railway network in Lombardy, Italy, and developed an innovative pipeline to integrate ticket and subscription sales and Automated Passenger Counting data using the Iterative Proportional Fitting algorithm. By effectively navigating the complexities of diverse and incomplete data sources, our approach showcases adaptability across various transportation contexts. Our research offers a valuable tool for operators, policymakers, and researchers, bridging the gap between data availability and the need for precise OD matrices. Additionally, we emphasise the potential of dynamic OD matrices and showcase methods for detecting anomalies in mobility trends, interpreting them in the context of events from the last months of 2022.

Inferring school district learning modalities during the COVID-19 pandemic with a hidden Markov model.

During the COVID-19 pandemic, many public schools across the United States shifted from fully in-person learning to alternative learning modalities such as hybrid and fully remote learning. In this study, data from 14,688 unique school districts from August 2020 to June 2021 were collected to track changes in the proportion of schools offering fully in-person, hybrid and fully remote learning over time. These data were provided by Burbio, MCH Strategic Data, the American Enterprise Institute's Return to Learn Tracker and individual state dashboards. Because the modalities reported by these sources were incomplete and occasionally misaligned, a model was needed to combine and deconflict these data to provide a more comprehensive description of modalities nationwide. A hidden Markov model (HMM) was used to infer the most likely learning modality for each district on a weekly basis. This method yielded higher spatiotemporal coverage than any individual data source and higher agreement with three of the four data sources than any other single source. The model output revealed that the percentage of districts offering fully in-person learning rose from 40.3% in September 2020 to 54.7% in June of 2021 with increases across 45 states and in both urban and rural districts. This type of probabilistic model can serve as a tool for fusion of incomplete and contradictory data sources in order to obtain more reliable data in support of public health surveillance and research efforts.

The discrepancy between expenditure- and income-side estimates of US output

The United States has two measures of economic output: gross domestic product (GDP) and gross domestic income (GDI). While these are conceptually equivalent, their initial estimates differ because these initial estimates are computed from different and incomplete data sources. I study the difference, or “statistical discrepancy,” between GDP and GDI in percent and document three features. First, its size does not materially shrink on average as more data become available. Second, the size of the initial discrepancy in absolute value does not predict the size of the discrepancy in absolute value after revisions. Third, the initial discrepancy has some predictive information about revisions to lagged GDP growth but no predictive information about revisions to lagged GDI growth.

Use of the Capture-Recapture Method to Estimate the Frequency of Community- and Hospital-Acquired Drug-Induced Acute Kidney Injuries in French Databases.

Drug-induced acute kidney injury (AKI) can occur both in primary care (i.e., community-acquired AKI (CA-AKI)) and in hospital settings (i.e., hospital-acquired AKI (HA-AKI)). The reported prevalence of these events varies markedly from one study to another, mainly due to differences in the study design. To estimate the frequency of drug-induced AKIs (both CA-AKIs and HA-AKIs) observed in a French university hospital, we applied the capture-recapture method to 1) the French national pharmacovigilance database (FPVD) and 2) a cohort of hospitalized patients with drug-induced AKIs (documented by analyzing the French national hospital discharge database and the patients’ electronic medical records). Drug-induced AKIs were determined according to the Naranjo algorithm and then categorized as CA-AKIs or HA-AKIs. A total number of 1,557 episodes of AKI were record during the study period, of them, the estimated total number of drug-induced AKIs was 593 [95% confidence interval (CI): 485–702], and the estimated prevalence was 38.1% [95%CI: 35.67–40.50]. The prevalences of HA-AKIs and CA-AKIs were similar (39.4% [36.24–42.54] and 37.4% [33.67–41.21], respectively). Only 6.1% of the drug-induced AKIs were recorded in the FPVD, and the proportions of recorded HA-AKIs and CA-AKI differed markedly (3.0% vs. 10.5%, respectively). One of the most frequently involved drug classes were antibiotics in the HA-AKI subgroup (13.0%) and antineoplastics in the CA-AKI subgroup (8.3%). Application of the capture-recapture method to two incomplete data sources can improve the ability to identify and quantify adverse drug reactions like AKIs. The frequency of drug-induced AKI is relatively high and is probably underestimated. The clinical management of an AKI might depend on where it originated.

Distrust of Artificial Intelligence: Sources & Responses from Computer Science & Law

Abstract Social distrust of AI stems in part from incomplete and faulty data sources, inappropriate redeployment of data, and frequently exposed errors that reflect and amplify existing social cleavages and failures, such as racial and gender biases. Other sources of distrust include the lack of “ground truth” against which to measure the results of learned algorithms, divergence of interests between those affected and those designing the tools, invasion of individual privacy, and the inapplicability of measures such as transparency and participation that build trust in other institutions. Needed steps to increase trust in AI systems include involvement of broader and diverse stakeholders in decisions around selection of uses, data, and predictors; investment in methods of recourse for errors and bias commensurate with the risks of errors and bias; and regulation prompting competition for trust.

Heterogeneity and Interoperability in Local Public Health Information Systems.

Heterogeneity and Interoperability in Local Public Health Information Systems.

Utilization of semantic web technologies to improve BIM-LCA applications

Currently, designers require to use numerous incomplete data sources and make multiple assumptions to conduct an LCA. We argue that traditional methods are incapable of addressing LCA data requirements, and for such a data and meta-data intensive task, the semantic web is the only logical choice of technology. This paper presents a review of the semantic web and its applications in the construction industry, along with an overview of the current LCA tools. LCA challenges are thoroughly discussed, and a framework using the semantic web is developed to address the data management issues of BIM-LCA applications. For comparison, the implementation is also done using an LCA tool (Tally), and a method using Relational Database Management Systems (RDBMS). Results show that the proposed method is superior to RDBMS methods in terms of capturing semantics and can improve LCA tools by providing reliable data in the early design stages.

The incidence of construction site injuries to women in Delhi: capture-recapture study

BackgroundIn India, the construction sector provides the main alternative to agricultural work - seasonal migration to and from construction work is widespread and construction work remains the second-largest employer of women in the country behind agriculture. Occupational injuries, which kill over 300,000 people annually, are a serious public health concern. However, data on construction site injuries to women are lacking, as India does not publish statistics on occupational injuries and illnesses. Our objectives were to: Estimate the number of women injured in construction site accidents in Delhi; and to estimate and compare the annual construction site injury rates per 100,000 workers of males and females in Delhi.MethodsWe conducted a two-sample capture-recapture study using data for accidents reported to the Delhi Police, Employee State Insurance Corporation (ESIC), and Commissioners of Workmen Compensation (CWC) of Delhi Government. The capture-recapture method has been used in epidemiology, to estimate morbidity and mortality using multiple, overlapping, but incomplete data sources. This study is based on the injuries reported from construction site accidents in Delhi in 2017. We linked the data from each of the data sources using the name, gender, and age of each injured person, the date and place of the accident, and the name of the employer. We used the Chapman estimator to estimate the total incidence of construction injuries in Delhi.ResultsWe estimated that there was a total of 37 female construction site workers injured (17 fatal and 20 non-fatal) in Delhi in 2017. There was a total of 1043 male construction site workers injured (236 fatal and 807 non-fatal). FIRs ascertained two-thirds (68%) of all injuries to females but only one third (34%) of those to males. The annual construction site injury rate per 100,000 workers of females was 82.26 (95%CI: 57.92 to 113.39). The annual construction site injury rate per 100,000 workers of males was 146.5 (95%CI: 137.7 to 155.6). There was strong evidence (p = 0.001) that the overall construction site injury rate per 100,000 workers of females was about one half the rate of males [rate ratio 0.56 (95%CI: 0.40 to 0.78)]. There was no evidence (p = 0.601) that the rates of fatal injuries differed in males and females (rate ratio 1.14 (95%CI: 0.70 to 1.87).ConclusionsThis study is the first to estimate the incidence of injuries to female construction site workers in India. The overall injury rate of female construction workers was over half as great as the rate of males. This implies that female construction workers face a not insignificant risk. Hence, safety measures (e.g., personal protective equipment) that are appropriate and culturally acceptable to Indian women are needed.

Improving tree biomass models through crown ratio patterns and incomplete data sources

Aboveground biomass quantification is essential for determining carbon stocks in forests. Multiple tree biomass models are available, but estimations can be biased outside the fitting range. This is due to the lack of data for larger trees, mainly because of the cost and time required. This study proposed a methodology based on tree crown biomass ratio (crown biomass: total aboveground biomass) modelling. The original data used in the existing biomass models in Spain have been notably extended by the inclusion of stem data from First Spanish National Forest Inventory and other databases, covering better tree size variability. The analysis of the crown biomass ratio against tree size (d2h), allowed us to distinguish three different patterns: an increasing pattern, a constant one, and a decreasing pattern. A new system of biomass models was fitted simultaneously by species, including a model for crown biomass ratio according to the identified pattern, a stem biomass model, and a total aboveground biomass model. Using this methodology, models were fitted for the 29 most important species in Spain. The fitted models result in more accurate and unbiased predictions for stem biomass, and realistic estimations for the crown biomass. This methodology means more robust and flexible biomass estimations with the possibility of using different data sources. The absence of crown information is not an obstacle because this component is a percentage of total aboveground biomass. Moreover, determining the crown biomass ratio pattern allows improving the accuracy of tree biomass estimation beyond the range of tree sizes (2–70 cm) for which these models were fitted.

Causal Effect Identification from Multiple Incomplete Data Sources: A General Search-Based Approach

Causal effect identification considers whether an interventional probability distribution can be uniquely determined without parametric assumptions from measured source distributions and structural knowledge on the generating system. While complete graphical criteria and procedures exist for many identification problems, there are still challenging but important extensions that have not been considered in the literature. To tackle these new settings, we present a search algorithm directly over the rules of do-calculus. Due to generality of do-calculus, the search is capable of taking more advanced data-generating mechanisms into account along with an arbitrary type of both observational and experimental source distributions. The search is enhanced via a heuristic and search space reduction techniques. The approach, called do-search, is provably sound, and it is complete with respect to identifiability problems that have been shown to be completely characterized by do-calculus. When extended with additional rules, the search is capable of handling missing data problems as well. With the versatile search, we are able to approach new problems such as combined transportability and selection bias, or multiple sources of selection bias. We perform a systematic analysis of bivariate missing data problems and study causal inference under case-control design. We also present the R package dosearch that provides an interface for a C++ implementation of the search.

1384. Conceptual Economic Model Methodology for Infant Pneumococcal Conjugate Vaccine Program and its Impact on Antimicrobial Resistance

BackgroundAntimicrobial resistance (AMR) is a global threat to effective prevention and treatment of an ever-increasing range of infections. Pneumococcal conjugate vaccines (PCV) used in infant national immunization programs have been shown to decrease AMR pneumococci. Cost-effectiveness models evaluating the value for money of PCV programs have not considered the economic impact of reducing antimicrobial prescribing or prolonged infections due to treatment failures. Standardized frameworks are needed for models to address outcomes and impact on health resource utilization related to AMR.MethodsWe developed a conceptual modeling methodology suitable for a health economic evaluation of an infant PCV program. We considered impact of PCVs on pneumococcal disease (PD) specifically related to clinical management of AMR-PD, including AMR epidemiology, antibiotic prescribing patterns, and healthcare resource utilization. Model inputs were evaluated regarding optimal and available data sources considering the complex nature of AMR at the national, regional, and global level.ResultsThe proposed framework considers impact of PCVs on antimicrobial prescribing due to invasive pneumococcal disease (IPD), community acquired pneumonia (CAP), and acute otitis media (AOM) across 3 pathways (Figure 1). The population and pathogen-level pathway describe epidemiology and vaccine impact. The care level pathway describes clinical disease management. The health outcomes pathway characterizes resistant or successfully treated PD costs and quality of life.Conceptual Economic Model Methodology ConclusionWe present a generalizable methodology to quantify impact of PCVs on cases and outcomes of PD related to AMR. Modelling vaccine-preventable burden of AMR-PD requires data extrapolations and assumptions due to the myriad of interconnected pathways (i.e. microbiology, epidemiology, environment, health systems). Further work is needed to validate assumptions and linkages across incomplete data sources.Disclosures Raymond Farkouh, PhD, Pfizer (Employee) Arianna Nevo, MPH, Pfizer, Inc. (Other Financial or Material Support, I am an employee of IQVIA. IQVIA received funding from Pfizer to carry out the project.) Jennifer Uyei, PhD, MPH, Pfizer, Inc. (Other Financial or Material Support, I am an employee of IQVIA. IQVIA received funding from Pfizer to carry out the project.) Cassandra Hall-Murray, PharmD, Pfizer, Inc. (Employee) Joseph Lewnard, PhD, Pfizer, Inc. (Consultant, Grant/Research Support, Advisor or Review Panel member) Matthew Wasserman, MSc., Pfizer Inc. (Employee)

Characterizing Tractability of Simple Well-Designed Pattern Trees with Projection

We study the complexity of evaluating well-designed pattern trees, a query language extending conjunctive queries with the possibility to define parts of the query to be optional. This possibility of optional parts is important for obtaining meaningful results over incomplete data sources as it is common in semantic web settings. Recently, a structural characterization of the classes of well-designed pattern trees that can be evaluated in polynomial time was shown. However, projection—a central feature of many query languages—was not considered in this study. We work towards closing this gap by giving a characterization of all tractable classes of simple well-designed pattern trees with projection (under some common complexity theoretic assumptions). Since well-designed pattern trees correspond to the fragment of well-designed {AND, OPTIONAL}-SPARQL queries this gives a complete description of the tractable classes of queries with projections in this fragment that can be characterized by the underlying graph structures of the queries. For non-simple pattern trees the tractability criteria for simple pattern trees do not capture all tractable classes. We thus extend the characterization for the non-simple case in order to capture some additional tractable cases.

Querying Rich Ontologies by Exploiting the Structure of Data

Ontology-based data access (OBDA) has emerged as a paradigm for accessing heterogeneous and incomplete data sources. A fundamental reasoning service in OBDA, the ontology mediated query (OMQ) answering has received much attention from the research community. However, there exists a disparity in research carried for OMQ algorithms for lightweight DLs which have found their way into practical implementations, and algorithms for expressive DLs for which the work has had mainly theoretical oriented goals. In the dissertation, a technique that leverages the structural properties of data to help alleviate the problems that typically arise when answering the queries in expressive settings is developed. In this paper, a brief summary of the technique along with the different algorithms developed for OMQ for expressive DLs is given.

A climate services perspective on Norwegian stormwater-related databases

Floods and stormwater events are the costliest natural catastrophes. Costs are expected to increase due to urbanization and climate change. Mitigation is needed. Different stakeholders with different motivations unfortunately often evaluate vulnerability by using fragmented and incomplete data sources. This paper intends to review the different approaches for collecting and analyzing data, and to evaluate their usefulness within the proposed framework for a “smart” use of data. The objectives of this work have been to review qualitatively and quantitatively a selection of Norwegian stormwater-related databases and to propose measures for improvement. The findings are seen according to the climate services literature and show that that data is spread around a heterogeneous community of stakeholders concerned with different motivations, different needs, and different levels of data processing. In general, the needs of the different stakeholders have not been surveyed and defined systematically enough and there is a substantial potential in upgrading from the delivery of passive raw data to the delivery of knowledge-driven decision-support tools.

State Trauma Registries as a Resource for Occupational Injury Surveillance and Research: Lessons From Washington State, 1998-2009.

Work-related traumatic injury is a leading cause of death and disability among US workers. Occupational injury surveillance is necessary for effective prevention planning and assessing progress toward Healthy People 2020 objectives. Our objectives were to (1) describe the Washington State Trauma Registry (WTR) as a resource for occupational injury surveillance and research, (2) compare the WTR with 2 population-based data sources more widely used for these purposes, and (3) compare the number of injuries ascertained by the WTR with other data sources. We linked WTR records to hospital discharge records in the Comprehensive Hospital Abstract Reporting System for 2009 and to workers' compensation claims from the Washington State Department of Labor and Industries for 1998 to 2008. We assessed the 3 data sources for overlap, concordance, and case ascertainment. Of 9185 work-related injuries in the WTR, 3380 (37%) did not link to workers' compensation claims. Use of payer information in hospital discharge records along with the WTR work-relatedness field identified 20% more linked injuries as work related (n = 720) than did use of payer information alone (n = 602). The WTR identified substantial numbers of work-related injuries that were not identified through workers' compensation or hospital discharge records. Workers' compensation and hospital discharge databases are important but incomplete data sources for work-related injuries; many work-related injuries are not billed to, reported to, or covered by workers' compensation. Trauma registries are well positioned to capture severe work-related injuries and should be included in comprehensive injury surveillance efforts.

Towards periodic and time-referenced flood risk assessment using airborne remote sensing

Abstract Flood risk management is founded on the regular assessment of damage potential. A significant parameter for assessing damage potential is the number of at-risk objects. However, data sets on exposure are often incomplete and/or lack time-references. Airborne remote sensing data, such as orthophotos, offers a regularly-updated, time-referenced depiction of land use. As such, remote sensing data compensates for incomplete data sources (such as digital cadastral maps). Orthophotos can even be used to analyze the temporal dynamics of flood damage potential, providing that time-referenced information is available for multiple time points. This paper describes a method for integrating orthophotos into flood risk analyses. In Austria, orthophotos are updated every three years, allowing their integration into cyclical flood risk assessments. The results of a case study presented in this paper demonstrate that orthophotos are most useful where other data sources, such as digital cadastral maps, are incomplete. In such situations, orthophotos lead to a significant increase in estimated damage potential. Orthophoto analysis allows damage potentials to be re-assessed at regular intervals, another major advantage over digital cadastral maps. Orthophoto analysis thus supports the evaluation of flood risk management options.

An Application of a Zero‐inflated Lifetime Distribution with Multiple and Incomplete Data Sources

We analyze data sampled from a population of parts in which an associated anomaly can occur at assembly or after assembly. Using a zero‐inflated lifetime distribution to fit left‐censored and right‐censored data as well data from a supplementary sample, we make predictions about the proportion of the population with anomalies today and in the future. Goodness‐of‐fit is also addressed. Copyright © 2016 John Wiley & Sons, Ltd.

Ontology-Based Data Access

Ontology-based data access is concerned with querying incomplete data sources in the presence of domain-specific knowledge provided by an ontology. A central notion in this setting is that of an ontology-mediated query , which is a database query coupled with an ontology. In this article, we study several classes of ontology-mediated queries, where the database queries are given as some form of conjunctive query and the ontologies are formulated in description logics or other relevant fragments of first-order logic, such as the guarded fragment and the unary negation fragment. The contributions of the article are threefold. First, we show that popular ontology-mediated query languages have the same expressive power as natural fragments of disjunctive datalog, and we study the relative succinctness of ontology-mediated queries and disjunctive datalog queries. Second, we establish intimate connections between ontology-mediated queries and constraint satisfaction problems (CSPs) and their logical generalization, MMSNP formulas. Third, we exploit these connections to obtain new results regarding: (i) first-order rewritability and datalog rewritability of ontology-mediated queries; (ii) P/NP dichotomies for ontology-mediated queries; and (iii) the query containment problem for ontology-mediated queries.

REGIONAL AND SECTORAL DISAGGREGATION OF MULTI-REGIONAL INPUT–OUTPUT TABLES – A FLEXIBLE ALGORITHM

A common shortcoming of available multi-regional input–output (MRIO) data sets is their lack of regional and sectoral detail required for many research questions (e.g. in the field of disaster impact analysis). We present a simple algorithm to refine MRIO tables regionally and/or sectorally. By the use of proxy data, each MRIO flow in question is disaggregated into the corresponding sub-flows. This downscaling procedure is complemented by an adjustment rule ensuring that the sub-flows match the superordinate flow in sum. The approximation improves along several iteration steps. The algorithm unfolds its strength through the flexible combination of multiple, possibly incomplete proxy data sources. It is also flexible in a sense that any target sector and region resolution can be chosen. As an exemplary case we apply the algorithm to a regional and sectoral refinement of the Eora MRIO database.

Real time adaptive non‐linear estimator/predictor design for traffic systems with inadequate detectors

Traffic control and management needs accurate estimation and prediction of traffic variables such as flow, speed, volume, travel time etc. Linear stochastic time series methods are powerful analytical tools. The capability of state space reconstruction makes them popular in traffic prediction and estimation. In this study, to overcome the linearity assumption of these methods, a non-linear kernel-based stochastic time series method with state space reconstruction capability is proposed. To minimise the prediction error of the method, adaptive time variant transformation from primary space to reproducing kernel Hilbert space is proposed by employing extended Kalman filter. Owing to high costs of traffic detectors, not all the metropolitan areas are equipped with these sensors; therefore in this study, an extended Kalman observer based on the new dynamic-adaptive-non-linear predictor is designed and applied for traffic flow estimation and prediction in the areas that suffer from lack of detectors. Practical data simulations and evaluations justify the high strength and accuracy of the proposed method in prediction of traffic speed with incomplete data sources.