Nontraditional Datasets Research Articles

Performance analysis and comparison of cat and dog image classification based on different models

Image classification has widespread applications in computer vision, with significant advancements in performance due to deep learning models. Cat and dog image classification, as a classic problem, has attracted considerable research interest. This study aims to conduct a comprehensive analysis and comparison of deep learning models, including LeNet, ResNet, and VGG, in the context of cat and dog image classification. This paper employed two datasets: traditional cat and dog images and non-traditional, diverse images. Data preprocessing and augmentation were applied, and various model architectures were constructed. Through training and testing, this paper assessed the performance of these models under different conditions. The research findings indicate that ResNet excels in handling various datasets and different dataset sizes, demonstrating outstanding image classification performance. LeNet performs well on traditional datasets but experiences performance degradation when dealing with non-traditional datasets and smaller dataset sizes. VGG performs reasonably well on the original dataset but needs help processing non-traditional datasets. These results provide valuable insights for guiding model selection and optimization in image classification tasks.

Towards near-real-time air pollutant and greenhouse gas emissions: lessons learned from multiple estimates during the COVID-19 pandemic

Abstract. The 2020 COVID-19 crisis caused an unprecedented drop in anthropogenic emissions of air pollutants and greenhouse gases. Given that emissions estimates from official national inventories for the year 2020 were not reported until 2 years later, new and non-traditional datasets to estimate near-real-time emissions became particularly relevant and widely used in international monitoring and modelling activities during the pandemic. This study investigates the impact of the COVID-19 pandemic on 2020 European (the 27 EU member states and the UK) emissions by comparing a selection of such near-real-time emission estimates, with the official inventories that were subsequently reported in 2022 under the Convention on Long-Range Transboundary Air Pollution (CLRTAP) and the United Nations Framework Convention on Climate Change (UNFCCC). Results indicate that annual changes in total 2020 emissions reported by official and near-real-time estimates are fairly in line for most of the chemical species, with NOx and fossil fuel CO2 being reported as the ones that experienced the largest reduction in Europe in all cases. However, large discrepancies arise between the official and non-official datasets when comparing annual results at the sector and country level, indicating that caution should be exercised when estimating changes in emissions using specific near-real-time activity datasets, such as time mobility data derived from smartphones. The main examples of these differences are observed for the manufacturing industry NOx (relative changes ranging between −21.4 % and −5.4 %) and road transport CO2 (relative changes ranging between −29.3 % and −5.6 %) total European emissions. Additionally, significant discrepancies are observed between the quarterly and monthly distribution of emissions drops reported by the various near-real-time inventories, with differences of up to a factor of 1.5 for total NOx during April 2020, when restrictions were at their maximum. For residential combustion, shipping and the public energy industry, results indicate that changes in emissions that occurred between 2019 and 2020 were mainly dominated by non-COVID-19 factors, including meteorology, the implementation of the Global Sulphur Cap and the shutdown of coal-fired power plants as part of national decarbonization efforts, respectively. The potential increase in NMVOC emissions from the intensive use of personal protective equipment such as hand sanitizer gels is considered in a heterogeneous way across countries in officially reported inventories, indicating the need for some countries to base their calculations on more advanced methods. The findings of this study can be used to better understand the uncertainties in near-real-time emissions and how such emissions could be used in the future to provide timely updates to emission datasets that are critical for modelling and monitoring applications.

The possibilities of exploring nontraditional datasets with young children

AbstractToday's world is characterized by the extensive production of data in different scenarios that everyday citizens need to understand for their informed participation in society. With the increase in the availability of data in a society defined by the industrious production of data, the educational system needs to think of possibilities to bring young children closer to the world of data science. This paper presents a nontraditional data exploration experience with an 8‐year‐old participant helped by a data visualization tool. A task‐based interview was conducted while the participant explored a carbon dioxide emission dataset. This paper studied how the participant interrogates the data, draws inferences and exhibits dispositions. At the end, some reflections are presented when introducing the exploration of nontraditional data in teaching.

Large-scale agricultural burning and cardiorespiratory emergency department visits in the U.S. state of Kansas.

Prescribed agricultural burning is a common land management practice, but little is known about the health effects from the resulting smoke exposure. To examine the association between smoke from prescribed burning and cardiorespiratory outcomes in the U.S. state of Kansas. We analyzed a zip code-level, daily time series of primary cardiorespiratory emergency department (ED) visits for February-May (months when prescribed burning is common in Kansas) in the years 2009-2011 (n = 109,220). Given limited monitoring data, we formulated a measure of smoke exposure using non-traditional datasets, including fire radiative power and locational attributes from remote sensing data sources. We then assigned a population-weighted potential smoke impact factor (PSIF) to each zip code, based on fire intensity, smoke transport, and fire proximity. We used Poisson generalized linear models to estimate the association between PSIF on the same day and in the past 3 days and asthma, respiratory including asthma, and cardiovascular ED visits. During the study period, prescribed burning took place on approximately 8 million acres in Kansas. Same-day PSIF was associated with a 7% increase in the rate of asthma ED visits when adjusting for month, year, zip code, meteorology, day of week, holidays, and correlation within zip codes (rate ratio [RR]: 1.07; 95% confidence interval [CI]: 1.01, 1.13). Same-day PSIF was not associated with a combined outcome of respiratory ED visits (RR [95% CI]: 0.99 [0.97, 1.02]), or cardiovascular ED visits (RR [95% CI]: 1.01 [0.98, 1.04]). There was no consistent association between PSIF during the past 3 days and any of the outcomes. These results suggest an association between smoke exposure and asthma ED visits on the same day. Elucidating these associations will help guide public health programs that address population-level exposure to smoke from prescribed burning.

Open-Source Datasets for Recommender Systems Analysis

There are different traditional and nontraditional datasets available to investigate the performance of recommender systems. This article focuses on the different datasets required for the investigation of recommender systems.

Managing Editor’s Letter

1. Francesco A. Fabozzi 1. Managing Editor <!-- --> 1. To order reprints of this article, please contact David Rowe at d.rowe{at}pageantmedia.com or 646-891-2157. In asset management, alternative data are diverse nontraditional datasets utilized by quantitative and fundamental

Alternative Data in Investment Management: Usage, Challenges, and Valuation

<i>Alternative data</i> in finance is an umbrella term for diverse nontraditional datasets used by quantitative and fundamental institutional investors to enhance portfolio returns. Although the use of alternative data is a recent phenomenon, it was not until the last five years that it gained widespread acceptance and the sector started to evolve into a complex ecosystem of data originators, intermediaries, and investors. The alternative data industry faces several obstacles, including difficulty estimating a dataset’s value to investors and technical challenges in leveraging these datasets efficiently at a large scale. In this article, the authors provide an up-to-date description of the alternative data space as it relates to the institutional investment industry. The authors elaborate on what alternative data are and how they are used in investment management. The authors identify and discuss some of the key challenges that arise when working with alternative data. In particular, they address issues such as entity mapping, ticker-tagging, panel stabilization, and debiasing with modern statistical and machine learning approaches. The authors advance several methodologies for the valuation of alternative datasets, including an event study methodology they refer to as the <i>golden triangle</i>, the application of report cards, and the relationship between a dataset’s structure of information content and its potential to enhance investment returns. To illustrate the effectiveness of these methods, they apply them to a case study analysis of real-world healthcare data, delivering an improvement in revenue prediction accuracy from an 88% mean absolute error to a 2.6% mean absolute error. <b>Key Findings</b> ▪ The authors describe the alternative data industry in an investment context. They discuss the commercial incentives for different categories of organizations within the industry and methodologies to valuate alternative datasets. ▪ They identify key technical challenges in working with alternative data, including entity mapping, ticker-tagging, panel stabilization, and debiasing. ▪ Through a case study, they demonstrate how to use alternative data combined with imputation and optimization techniques to predict revenues of publicly traded companies within the healthcare sector.

Alternative Data in Investment Management: Usage, Challenges and Valuation

Alternative data in finance is an umbrella term for diverse non-traditional datasets used by quantitative and fundamental institutional investors to enhance portfolio returns. While the use of alternative data is a recent phenomenon, it was not until the last five years that it gained widespread acceptance and the sector started to evolve into a complex ecosystem of data originators, intermediaries and investors. The alternative data industry faces several obstacles, including difficulty estimating a dataset’s value to investors and technical challenges in leveraging these datasets efficiently at large scale. In this article, the authors provide an up-to-date description of the alternative data space as it relates to the institutional investment industry. The authors elaborate on what alternative data is and how it is used in investment management. The authors identify and discuss some of the key challenges that arise when working with alternative data. In particular, they address issues such as entity mapping, ticker-tagging, panel stabilization and debiasing with modern statistical and machine learning approaches. The authors advance several methodologies for the valuation of alternative datasets, including an event study methodology they refer to as the Golden Triangle, the application of report cards, and the relationship between a datasets’ structure of information content its potential to enhance investment returns. To illustrate the effectiveness of the methods, they apply them to a case study analysis of real-world healthcare data, delivering an improvement of revenue prediction accuracy from an 88% mean absolute error to a 2.6% mean absolute error.

Large-scale agricultural burning and asthma emergency department visits in the U.S. state of Kansas

Background: Prescribed agricultural burning is a common land management practice, but little is known about the respiratory health effects from the resulting smoke exposure. We examined the association between smoke from prescribed burning and asthma in the U.S. state of Kansas. Methods: We analyzed a zip code-level daily time series of primary asthma emergency department (ED) visits and environmental variables for years 2009 - 2011 (n=62,105); analyses were restricted to the months when prescribed burning is common in Kansas (February - May). Given limited monitoring data, we formulated a measure of smoke exposure using non-traditional datasets, including fire radiative power and locational attributes from remote sensing data sources and surface wind characteristics from model-based earth observations. We then assigned a population-weighted potential smoke impact factor (PSIF) to each zip code, based on fire intensity, smoke transport and proximity (to fire) considerations. We used Poisson generalized linear models to estimate the association between PSIF (both as binary and 4-level categorical variables) and asthma ED visits. Results: During the study period, prescribed burning was used to clear approximately 8 million acres in Kansas. Same-day PSIF was associated with a 3% increase in the rate of asthma ED visits when adjusting for month, year, zip code, meteorology, day of week, holidays, and correlation within zip codes (rate ratio [RR]:1.03; 95% confidence interval [CI]:0.97, 1.09). When comparing tertiles of same-day exposure to a reference group of no exposure, adjusted rates of asthma ED visits were elevated for the two lowest tertiles, but did not increase monotonically (RR[CI] from lowest to highest exposure: 1.06[0.95, 1.18], 1.05[0.97, 1.14], 1.00[0.93, 1.08]). Conclusions: Our preliminary results are suggestive of an association between smoke from prescribed burning and asthma ED visits. Elucidating these associations will help guide public health programs that address population-level exposure to smoke from prescribed burning.

Patient-Reported Outcomes in Online Communications on Statins, Memory, and Cognition: Qualitative Analysis Using Online Communities.

BackgroundIn drug development clinical trials, there is a need for balance between restricting variables by setting eligibility criteria and representing the broader patient population that may use a product once it is approved. Similarly, although recent policy initiatives focusing on the inclusion of historically underrepresented groups are being implemented, barriers still remain. These limitations of clinical trials may mask potential product benefits and side effects. To bridge these gaps, online communication in health communities may serve as an additional population signal for drug side effects.ObjectiveThe aim of this study was to employ a nontraditional dataset to identify drug side-effect signals. The study was designed to apply both natural language processing (NLP) technology and hands-on linguistic analysis to a set of online posts from known statin users to (1) identify any underlying crossover between the use of statins and impairment of memory or cognition and (2) obtain patient lexicon in their descriptions of experiences with statin medications and memory changes.MethodsResearchers utilized user-generated content on Inspire, looking at over 11 million posts across Inspire. Posts were written by patients and caregivers belonging to a variety of communities on Inspire. After identifying these posts, researchers used NLP and hands-on linguistic analysis to draw and expand upon correlations among statin use, memory, and cognition.ResultsNLP analysis of posts identified statistical correlations between statin users and the discussion of memory impairment, which were not observed in control groups. NLP found that, out of all members on Inspire, 3.1% had posted about memory or cognition. In a control group of those who had posted about TNF inhibitors, 6.2% had also posted about memory and cognition. In comparison, of all those who had posted about a statin medication, 22.6% (P<.001) also posted about memory and cognition. Furthermore, linguistic analysis of a sample of posts provided themes and context to these statistical findings. By looking at posts from statin users about memory, four key themes were found and described in detail in the data: memory loss, aphasia, cognitive impairment, and emotional change.ConclusionsCorrelations from this study point to a need for further research on the impact of statins on memory and cognition. Furthermore, when using nontraditional datasets, such as online communities, NLP and linguistic methodologies broaden the population for identifying side-effect signals. For side effects such as those on memory and cognition, where self-reporting may be unreliable, these methods can provide another avenue to inform patients, providers, and the Food and Drug Administration.

An Evolutionary Schema for Using “it-is-what-it-is” Data in Official Statistics

AbstractThe linking of disparate data sets across time, space and sources is probably the foremost current issue facing Central Statistical Agencies (CSA). If one reviews the current literature looking for the prevalent challenges facing CSAs, three issues stand out: 1) using administrative data effectively; 2) big data and what it means for CSAs; and 3) integrating disparate data set (such as health, education and wealth) to provide measurable facts that can guide policy makers. CSAs are being challenged to explore the same kind of challenges faced by Google, Facebook, and Yahoo, which are using graphical/semantic web models for organizing, searching and analysing data. Additionally, time and space (geography) are becoming more important dimensions (domains) for CSAs as they start to explore new data sources and ways to integrate those to study relationships. Central agency methodologists are being pushed to include these new perspectives into their standard theories, practises and policies. Like most methodologists, the authors see surveys and the publications of their results as a process where estimation is the key tool to achieve the final goal of an accurate statistical output. Randomness and sampling exists to support this goal, and early on it was clear to us that the incoming “it-is-what-it-is” data sources were not randomly selected. These sources were obviously biased and thus would produce biased estimates. So, we set out to design a strategy to deal with this issue.This article presents a schema for integrating and linking traditional and non-traditional datasets. Like all survey methodologies, this schema addresses the fundamental issues of representativeness, estimation and total survey error measurement.

ORBiT - The Oak Ridge Biosurveillance Toolkit

We describe ORBiT, the Oak Ridge Biosurveillance Toolkit, the objective of which is to provide 1) forecasting and early warning, and 2) an extensible data analytics platform for biosurveillance by enabling the use of traditional and non-traditional datasets, consisting of heterogeneous data types and modalities. ORBiT includes a data collection interface, analytic components, higher-order statistical tools, and a visual front-end with visual analytics. ORBiT is implemented as a distributed analytic platform consisting of a software stack atop of Hadoop and uses Titan, a distributed graph database as a backend for data storage.

Taming the control of chronology in ancient agricultural structures in the Calchaqui Valley, Argentina. Non-traditional data sets

The absolute chronology of the agricultural soil from organic matter has always been difficult to measure due to several reasons. In our work on the Prehispanic agriculture in the Calchaqui Valley – Argentina, however, an absolute chronology is necessary to distinguish the Inca prevalence of the previous agricultural structure. Instead of looking for a new dating methodology, we have used the 14C method trying to eliminate the error sources typical of soils. Thus, we have dated new types of agricultural structures, with the assumption that they would have worked as archaeological “seals”. In the present paper we show that the dating of the beginning of the use of the Prehispanic agricultural systems has been successful.

Using alternative data sources to study rural migration: examples from Illinois

AbstractOne of the problems frequently faced by migration researchers is the paucity of national census migration data for small rural areas. Drawing upon a larger, multi‐level research project that focused upon rural poverty migration within Illinois, this article illustrates the opportunities and challenges associated with using community‐based data sources to measure mobility in small rural communities. Use of alternative migration data sources can present opportunities to circumvent common problems with census‐based data sources, including addressing temporal shortfalls, providing custom geographies, and capturing the movement of difficult‐to‐enumerate subpopulations. Alternative data‐sets may also allow researchers to formulate new questions that are more oriented to society's pressing problems. Yet, users of non‐traditional data‐sets also face issues of indirect measurement, data quality, comparability, replication, costs, and moral and ethical concerns. As use of non‐traditional migration data sources increases, researchers will need to become familiar with both the benefits and pitfalls of these data‐sets. Copyright © 2008 John Wiley &amp; Sons, Ltd.

Climate Change in Nontraditional Data Sets

Scientific examinations of nontraditional datasets collected by naturalists, hobbyists, or indig-enous peoples have been instrumental in linkingchanges in biological communities to recentclimatic changes. Many phenological records inparticular show close correlations between tim-ing of biological events and climatic conditions(