Data mining and linked open data – New perspectives for data analysis in environmental research

A knowledge elicitation approach to traffic accident analysis in open data: comparing periods before and after the Covid-19 outbreak

Extracting knowledge from open data of traffic accidents has been attracting increasing attention to policymakers responsible for road safety. This article presents a knowledge elicitation approach to exploring the determinants of traffic accidents from open government data of an urban area in Taiwan. The collected open dataset contains 34 decisional attributes and one predictive attribute (i.e., type of injury, including head, breast, leg), and 47,974 cases. Prediction models using a classification-oriented mechanism and generated rules that considered datasets from before (B-dataset; 30,116 cases) and after (A-dataset; 17,868 cases) beginning to combat the Covid-19 pandemic in an urban area of Taiwan were compared. The findings showed that prediction accuracy was acceptable but not high, at 70.73% for B-dataset and 74.77% for A-dataset. Determinants in the human and vehicle categories revealed higher classification ranks than those in the temporal and environment categories. Traffic accidents involving motorcycles were 5.13% higher in A-dataset, whereas those involving cars were 4.11% lower. Injury on leg or foot was 3.46% higher in A-dataset, whereas other types of injury were up to 1.00% lower. The average support for rules in the A-dataset rule base and the simplicity of the A-dataset decision tree were higher than those of B-dataset. The research demonstrates the value of open government data in prediction model development and knowledge elicitation to support policymaking in the traffic safety domain.

Remote Sensing of Geomorphodiversity Linked to Biodiversity—Part III: Traits, Processes and Remote Sensing Characteristics

Remote sensing (RS) enables a cost-effective, extensive, continuous and standardized monitoring of traits and trait variations of geomorphology and its processes, from the local to the continental scale. To implement and better understand RS techniques and the spectral indicators derived from them in the monitoring of geomorphology, this paper presents a new perspective for the definition and recording of five characteristics of geomorphodiversity with RS, namely: geomorphic genesis diversity, geomorphic trait diversity, geomorphic structural diversity, geomorphic taxonomic diversity, and geomorphic functional diversity. In this respect, geomorphic trait diversity is the cornerstone and is essential for recording the other four characteristics using RS technologies. All five characteristics are discussed in detail in this paper and reinforced with numerous examples from various RS technologies. Methods for classifying the five characteristics of geomorphodiversity using RS, as well as the constraints of monitoring the diversity of geomorphology using RS, are discussed. RS-aided techniques that can be used for monitoring geomorphodiversity in regimes with changing land-use intensity are presented. Further, new approaches of geomorphic traits that enable the monitoring of geomorphodiversity through the valorisation of RS data from multiple missions are discussed as well as the ecosystem integrity approach. Likewise, the approach of monitoring the five characteristics of geomorphodiversity recording with RS is discussed, as are existing approaches for recording spectral geomorhic traits/ trait variation approach and indicators, along with approaches for assessing geomorphodiversity. It is shown that there is no comparable approach with which to define and record the five characteristics of geomorphodiversity using only RS data in the literature. Finally, the importance of the digitization process and the use of data science for research in the field of geomorphology in the 21st century is elucidated and discussed.

From data mining to wisdom mining

The knowledge gained from data mining is highly dependent on the experience of an expert for further analysis to increase effectiveness and wise decision-making. This mined knowledge requires actionability enhancement before it can be applied to real-world problems. The literature highlights the reasons that emerged the need to incorporate human wisdom in decision-making for complex problems. To solve this problem, a domain called ‘Wisdom Mining’ is recommended, proposing a set of algorithms parallel to the algorithms proposed by the data mining. In wisdom mining, a process to extract wisdom needs to be defined with less influence from an expert. This review proposed improvements to data mining techniques and their applications in the real world and emphasised the need to seek ways to harness wisdom from data. This study covers the diverse definitions and different perspectives of wisdom within philosophy, psychology, management and computer science. This comprehensive literature review served as a foundation for constructing a wise decision framework that aided in identifying the wisdom factors like context, utility, location and time. The inclusion of these wisdom factors in existing data mining algorithms makes the transition from data mining to wisdom mining possible. This research includes the relationship between these two mining process that facilitated further elucidation of the wisdom mining process. Potential research trends in the domain are also seen as a potential endeavour to improve the analysis and use of data.

A rough set based algorithm for updating the modes in categorical clustering

The categorical clustering problem has attracted much attention especially in the last decades since many real world applications produce categorical data. The k-mode algorithm, proposed since 1998, and its multiple variants were widely used in this context. However, they suffer from a great limitation related to the update of the modes in each iteration. The mode in the last step of these algorithms is randomly selected although it is possible to identify many candidate ones. In this paper, a rough density mode selection method is proposed to identify the adequate modes among a list of candidate ones in each iteration of the k-modes. The proposed method, called Density Rough k-Modes (DRk-M) was experimented using real world datasets extracted from the UCI Machine Learning Repository, the Global Terrorism Database (GTD) and a set of collected Tweets. The DRk-M was also compared to many states of the art clustering methods and has shown great efficiency.

Monitoring Glyphosate-Based Herbicide Treatment Using Sentinel-2 Time Series—A Proof-of-Principle

In this paper we aim to show a proof-of-principle approach to detect and monitor weed management using glyphosate-based herbicides in agricultural practices. In a case study in Germany, we demonstrate the application of Sentinel-2 multispectral time-series data. Spectral broadband vegetation indices were analysed to observe vegetation traits and weed damage arising from herbicide-based management. The approach has been validated with stakeholder information about herbicide treatment using commercial products. As a result, broadband NDVI calculated from Sentinel-2 data showed explicit feedback after the glyphosate-based herbicide treatment. Vegetation damage could be detected after just two days following of glyphosate-based herbicide treatment. This trend was observed in three different application scenarios, i.e., during growing stage, before harvest and after harvest. The findings of the study demonstrate the feasibility of satellite based broadband NDVI data for the detection of glyphosate-based herbicide treatment and, e.g., the monitoring of latency to harvesting. The presented results can be used to implement monitoring concepts to provide the necessary transparency about weed treatment in agricultural practices and to support environmental monitoring.

Linking Remote Sensing and Geodiversity and Their Traits Relevant to Biodiversity—Part I: Soil Characteristics

In the face of rapid global change it is imperative to preserve geodiversity for the overall conservation of biodiversity. Geodiversity is important for understanding complex biogeochemical and physical processes and is directly and indirectly linked to biodiversity on all scales of ecosystem organization. Despite the great importance of geodiversity, there is a lack of suitable monitoring methods. Compared to conventional in-situ techniques, remote sensing (RS) techniques provide a pathway towards cost-effective, increasingly more available, comprehensive, and repeatable, as well as standardized monitoring of continuous geodiversity on the local to global scale. This paper gives an overview of the state-of-the-art approaches for monitoring soil characteristics and soil moisture with unmanned aerial vehicles (UAV) and air- and spaceborne remote sensing techniques. Initially, the definitions for geodiversity along with its five essential characteristics are provided, with an explanation for the latter. Then, the approaches of spectral traits (ST) and spectral trait variations (STV) to record geodiversity using RS are defined. LiDAR (light detection and ranging), thermal and microwave sensors, multispectral, and hyperspectral RS technologies to monitor soil characteristics and soil moisture are also presented. Furthermore, the paper discusses current and future satellite-borne sensors and missions as well as existing data products. Due to the prospects and limitations of the characteristics of different RS sensors, only specific geotraits and geodiversity characteristics can be recorded. The paper provides an overview of those geotraits.

Attribute Selection Based on Constraint Gain and Depth Optimal for a Decision Tree

Uncertainty evaluation based on statistical probabilistic information entropy is a commonly used mechanism for a heuristic method construction of decision tree learning. The entropy kernel potentially links its deviation and decision tree classification performance. This paper presents a decision tree learning algorithm based on constrained gain and depth induction optimization. Firstly, the calculation and analysis of single- and multi-value event uncertainty distributions of information entropy is followed by an enhanced property of single-value event entropy kernel and multi-value event entropy peaks as well as a reciprocal relationship between peak location and the number of possible events. Secondly, this study proposed an estimated method for information entropy whose entropy kernel is replaced with a peak-shift sine function to establish a decision tree learning (CGDT) algorithm on the basis of constraint gain. Finally, by combining branch convergence and fan-out indices under an inductive depth of a decision tree, we built a constraint gained and depth inductive improved decision tree (CGDIDT) learning algorithm. Results show the benefits of the CGDT and CGDIDT algorithms.

Understanding Forest Health with Remote Sensing, Part III: Requirements for a Scalable Multi-Source Forest Health Monitoring Network Based on Data Science Approaches

Forest ecosystems fulfill a whole host of ecosystem functions that are essential for life on our planet. However, an unprecedented level of anthropogenic influences is reducing the resilience and stability of our forest ecosystems as well as their ecosystem functions. The relationships between drivers, stress, and ecosystem functions in forest ecosystems are complex, multi-faceted, and often non-linear, and yet forest managers, decision makers, and politicians need to be able to make rapid decisions that are data-driven and based on short and long-term monitoring information, complex modeling, and analysis approaches. A huge number of long-standing and standardized forest health inventory approaches already exist, and are increasingly integrating remote-sensing based monitoring approaches. Unfortunately, these approaches in monitoring, data storage, analysis, prognosis, and assessment still do not satisfy the future requirements of information and digital knowledge processing of the 21st century. Therefore, this paper discusses and presents in detail five sets of requirements, including their relevance, necessity, and the possible solutions that would be necessary for establishing a feasible multi-source forest health monitoring network for the 21st century. Namely, these requirements are: (1) understanding the effects of multiple stressors on forest health; (2) using remote sensing (RS) approaches to monitor forest health; (3) coupling different monitoring approaches; (4) using data science as a bridge between complex and multidimensional big forest health (FH) data; and (5) a future multi-source forest health monitoring network. It became apparent that no existing monitoring approach, technique, model, or platform is sufficient on its own to monitor, model, forecast, or assess forest health and its resilience. In order to advance the development of a multi-source forest health monitoring network, we argue that in order to gain a better understanding of forest health in our complex world, it would be conducive to implement the concepts of data science with the components: (i) digitalization; (ii) standardization with metadata management after the FAIR (Findability, Accessibility, Interoperability, and Reusability) principles; (iii) Semantic Web; (iv) proof, trust, and uncertainties; (v) tools for data science analysis; and (vi) easy tools for scientists, data managers, and stakeholders for decision-making support.

Advanced data mining approaches in the assessment of urinary concentrations of bisphenols, chlorophenols, parabens and benzophenones in Brazilian children and their association to DNA damage

Human exposure to endocrine disrupting chemicals (EDCs) has received considerable attention over the last three decades. However, little is known about the influence of co-exposure to multiple EDCs on effect-biomarkers such as oxidative stress in Brazilian children. In this study, concentrations of 40 EDCs were determined in urine samples collected from 300 Brazilian children of ages 6-14 years and data were analyzed by advanced data mining techniques. Oxidative DNA damage was evaluated from the urinary concentrations of 8-hydroxy-2'-deoxyguanosine (8OHDG). Fourteen EDCs, including bisphenol A (BPA), methyl paraben (MeP), ethyl paraben (EtP), propyl paraben (PrP), 3,4-dihydroxy benzoic acid (3,4-DHB), methyl-protocatechuic acid (OH-MeP), ethyl-protocatechuic acid (OH-EtP), triclosan (TCS), triclocarban (TCC), 2-hydroxy-4-methoxybenzophenone (BP3), 2,4-dihydroxybenzophenone (BP1), bisphenol A bis(2,3-dihydroxypropyl) glycidyl ether (BADGE·2H₂O), 2,4-dichlorophenol (2,4-DCP), and 2,5-dichlorophenol (2,5-DCP) were found in >50% of the urine samples analyzed. The highest geometric mean concentrations were found for MeP (43.1 ng/mL), PrP (3.12 ng/mL), 3,4-DHB (42.2 ng/mL), TCS (8.26 ng/mL), BP3 (3.71 ng/mL), and BP1 (4.85 ng/mL), and exposures to most of which were associated with personal care product (PCP) use. Statistically significant associations were found between urinary concentrations of 8OHDG and BPA, MeP, 3,4-DHB, OH-MeP, OH-EtP, TCS, BP3, 2,4-DCP, and 2,5-DCP. After clustering the data on the basis of i) 14 EDCs (exposure levels), ii) demography (age, gender and geographic location), and iii) 8OHDG (effect), two distinct clusters of samples were identified. 8OHDG concentration was the most critical parameter that differentiated the two clusters, followed by OH-EtP. When 8OHDG was removed from the dataset, predictability of exposure variables increased in the order of: OH-EtP > OH-MeP > 3,4-DHB > BPA > 2,4-DCP > MeP > TCS > EtP > BP1 > 2,5-DCP. Our results showed that co-exposure to OH-EtP, OH-MeP, 3,4-DHB, BPA, 2,4-DCP, MeP, TCS, EtP, BP1, and 2,5-DCP was associated with DNA damage in children. This is the first study to report exposure of Brazilian children to a wide range of EDCs and the data mining approach further strengthened our findings of chemical co-exposures and biomarkers of effect.

Open data mining for Taiwan's dengue epidemic

By using a quantitative approach, this study examines the applicability of data mining technique to discover knowledge from open data related to Taiwan's dengue epidemic. We compare results when Google trend data are included or excluded. Data sources are government open data, climate data, and Google trend data. Research findings from analysis of 70,914 cases are obtained. Location and time (month) in open data show the highest classification power followed by climate variables (temperature and humidity), whereas gender and age show the lowest values. Both prediction accuracy and simplicity decrease when Google trends are considered (respectively 0.94 and 0.37, compared to 0.96 and 0.46). The article demonstrates the value of open data mining in the context of public health care.

Upland vegetation mapping using Random Forests with optical and radar satellite data

Uplands represent unique landscapes that provide a range of vital benefits to society, but are under increasing pressure from the management needs of a diverse number of stakeholders (e.g. farmers, conservationists, foresters, government agencies and recreational users). Mapping the spatial distribution of upland vegetation could benefit management and conservation programmes and allow for the impacts of environmental change (natural and anthropogenic) in these areas to be reliably estimated. The aim of this study was to evaluate the use of medium spatial resolution optical and radar satellite data, together with ancillary soil and topographic data, for identifying and mapping upland vegetation using the Random Forests (RF) algorithm. Intensive field survey data collected at three study sites in Ireland as part of the National Parks and Wildlife Service (NPWS) funded survey of upland habitats was used in the calibration and validation of different RF models. Eight different datasets were analysed for each site to compare the change in classification accuracy depending on the input variables. The overall accuracy values varied from 59.8% to 94.3% across the three study locations and the inclusion of ancillary datasets containing information on the soil and elevation further improved the classification accuracies (between 5 and 27%, depending on the input classification dataset). The classification results were consistent across the three different study areas, confirming the applicability of the approach under different environmental contexts.