Exploratory Data Analysis Methods Research Articles

Angle-based joint and individual variation explained

Integrative analysis of disparate data blocks measured on a common set of experimental subjects is a major challenge in modern data analysis. This data structure naturally motivates the simultaneous exploration of the joint and individual variation within each data block resulting in new insights. For instance, there is a strong desire to integrate the multiple genomic data sets in The Cancer Genome Atlas to characterize the common and also the unique aspects of cancer genetics and cell biology for each source. In this paper we introduce Angle-Based Joint and Individual Variation Explained capturing both joint and individual variation within each data block. This is a major improvement over earlier approaches to this challenge in terms of a new conceptual understanding, much better adaption to data heterogeneity and a fast linear algebra computation. Important mathematical contributions are the use of score subspaces as the principal descriptors of variation structure and the use of perturbation theory as the guide for variation segmentation. This leads to an exploratory data analysis method which is insensitive to the heterogeneity among data blocks and does not require separate normalization. An application to cancer data reveals different behaviors of each type of signal in characterizing tumor subtypes. An application to a mortality data set reveals interesting historical lessons. Software and data are available at GitHub https://github.com/MeileiJiang/AJIVE_Project.

The Empirical Study of the Spatial Character of Chinese Provincial Human Capital Based on ESDA Method

By utilizing the human capital penal data of China’s 31 provinces that was measured and calculated by J-F life time revenue from 1985 to 2012, it introduces the spatial factor, by using exploratory spatial data analysis (ESDA) method, through space bitmap, global Moran’s I, Moran scatter diagram to describe the spatial distribution characteristic of Chinese provincial human capital and its inner mechanism. Chinese provincial human capital’s spatial distribution had the non-equilibrium characteristics and spatial aggregation tendency, and the human capital was highly centered in a limited number of provinces; there’s the high significance spatial self-correlation in Chinese provincial human capital, with the interaction between the human capitals in different provinces; there’s also the local relativity and heterogenicity in the provincial human capital distribution.

Educators’ Perceptions of Integrated STEM: A Phenomenological Study

The study utilized a semistructured interview approach to identify phenomena that are related to integrated STEM education by addressing the question: What are the critical components of an integrated STEM definition and what critical factors are necessary for an integrated STEM definition’s implementation? Thirteen expert practitioners were identified and interviewed. The interviews were transcribed and analyzed for content in three different ways: by person, by interview question, and across all interviews using exploratory data analysis methods. Ten identified phenomena were grouped into two classes: structural implementation phenomena and interpersonal implementation phenomena. The structural implementation phenomena were: subject integration, project-based learning, and design-based education; nontraditional assessment; STEM content; time; professional development; and outside support (from businesses and industry). The interpersonal implementation phenomena include: leadership; collaboration; willingness; authentic, relevant, and meaningful experiences for participants; and outside support (from people in business and industry). The analysis concluded that these phenomena could be considered both critical components and key implementation factors due to their interconnected nature. The data showed that the identified phenomena are necessary as part of an integrated STEM curriculum, which makes them critical components, and that the identified phenomena are critical to create and implement an integrated STEM setting, making them implementation factors as well. Implications for further research include: the possibility of looking at the interconnectedness of the phenomena, examining how each phenomenon contributes to integrated STEM, and measuring current STEM implementations to see if they incorporate the identified phenomena. Additionally, inclusion of an absent phenomenon could be researched to see if integrated STEM education is improved.

The use of discriminant analysis methods for diagnosis of the causes of differences in the properties of resin mortar containing various fillers

Resin mortars belong to the group of concrete-like construction composites. They are obtained by mixing a synthetic resin with a hardener and an appropriately selected aggregate. The latter component is usually as much as 90% of the composite mass and can largely shape the characteristics of the finished product. The fact that the type of filler used can significantly differentiate the values of physical and mechanical parameters of epoxy mortars is confirmed by the results of the exploratory data analysis method used in this article, which is discriminant analysis. This allows us to examine differences between groups of objects based on a set of selected independent variables (predictors). It is used to solve a wide range of classification and prediction problems. The core of discriminant analysis is a model presented in the form of a linear combination of independent variables, which allows classification of observations (e.g. test mortars) into one of the groups that are of interest to the researcher. In discriminant analysis one can distinguish the learning stage (model building), in which classification rules are created based on research results (training set) and the classification stage, i.e. the use of the model, e.g. for testing its prognostic accuracy.

The Use of Univariate and Multivariate Analyses in the Geochemical Exploration, Ravanj Lead Mine, Delijan, Iran

A geochemical exploration program was applied to recognize the anomalous geochemical haloes at the Ravanj lead mine, Delijan, Iran. Sampling of unweathered rocks were undertaken across rock exposures on a 10 × 10 meter grid (n = 302) as well as the accessible parts of underground mine A (n = 42). First, the threshold values of all elements were determined using the cut-off values used in the exploratory data analysis (EDA) method. Then, for further studies, elements with lognormal distributions (Pb, Zn, Ag, As, Cd, Co, Cu, Sb, S, Sr, Th, Ba, Bi, Fe, Ni and Mn) were selected. Robustness against outliers is achieved by application of central log ratio transformation to address the closure problems with compositional data prior to principle components analysis (PCA). Results of these analyses show that, in the Ravanj deposit, Pb mineralization is characterized by a Pb-Ba-Ag-Sb ± Zn ± Cd association. The supra-mineralization haloes are characterized by barite and tetrahedrite in a Ba- Th- Ag- Cu- Sb- As- Sr association and sub-mineralization haloes are comprised of pyrite and tetrahedrite, probably reflecting a Fe-Cu-As-Bi-Ni-Co-Mo-Mn association. Using univariate and multivariate geostatistical analyses (e.g., EDA and robust PCA), four anomalies were detected and mapped in Block A of the Ravanj deposit. Anomalies 1 and 2 are around the ancient orebodies. Anomaly 3 is located in a thin bedded limestone-shale intercalation unit that does not show significant mineralization. Drilling of the fourth anomaly suggested a low grade, non-economic Pb mineralization.

Wind Turbine Accidents: A Data Mining Study

While the global production of wind energy is increasing, there exists a significant gap in the academic and practice literature regarding the analysis of wind turbine accidents. This paper presents the results obtained from the analysis of 240 wind turbine accidents from around the world. The main focus of this paper is revealing the associations between several factors and deaths and injuries in wind turbine accidents. Specifically, the associations of death and injuries with the stage of the wind turbine's life cycle (transportation, construction, operation, and maintenance) and the main cause factor categories (human, system/equipment, and nature) were studied. To this end, we conducted a detailed investigation that integrates exploratory and statistical data analysis and data mining methods. This paper presents a multitude of insights regarding the accidents and discusses implications for wind turbine manufacturers, engineering and insurance companies, and government organizations.

Comparison of the metal contamination in water measured by diffusive gradient in thin film (DGT), biomonitoring and total metal dissolved concentration at a national scale

In this study, we propose to compare the metal contamination in water assessed by three monitoring tools: diffusive gradient in thin film (DGT), caged gammarids and grab samples of the dissolved phase. The study was conducted at a national scale on 101 sites throughout France and involved the deployment of DGT and gammarids at the same time. Cd, Cr, Co, Cu, Mn, Ni, Pb and Zn were measured in DGT, dissolved phase and gammarids. The data set was analyzed with an exploratory data analysis method: multiple factor analysis (MFA), which aims at comparing data sets structured in groups of variables in order to compare the groups and the conclusions to be drawn from the analysis of the data set by each group. No differences were observed for Co and Ni for the three methods. The gammarid group gave the richest information on the contamination of the sites because it discriminated the sites contaminated by Pb and Cd. Depending on the metals, DGT provided more information than grab samples, but less than gammarids. We conclude that the three monitoring tools gave complementary information on metal contamination, and their effectiveness in assessing metal contamination in water depends on the metal and the fraction studied.

The use of STATICO and COSTATIS, two exploratory three-ways analysis methods: an application to the ecology of aquatic heteroptera in the Medjerda watershed (Tunisia)

This article is intended as a guideline to the use of two exploratory data analysis methods, namely STATICO and COSTATIS. Both techniques have already been used in the field of ecological data analysis, and we present a rapid survey of the ecological literature on three-ways analysis methods. Here, we wish to share some advanced computation and graphical display scripts to help ecologists use these methods. We first recall the main principles of these two methods for the analysis of the relationships between the structures of two series of data tables. In the context of ecology, these two series can be for example (1) a series of species data tables and (2) a series of environmental parameters tables. A detailed, real-size example is presented to show how this strategy can be put in place using the ade4 and adegraphics packages for R. This example relates to the ecology of aquatic Heteroptera in the Medjerda watershed (Tunisia). We show how the outputs of the two methods can be used to interpret the relationships between aquatic Heteroptera species distribution and environmental parameters. Several R scripts to conduct the computations and draw suitable graphical displays are reproduced and explained in the text and in five appendices.

Analysis of the Spatial Characteristics of the Water Usage Patterns Based on ESDA-GIS: An Example of Hubei Province, China

The spatial characteristics and the high-duty water regions of the Water Usage Patterns (WUP) are very important for the allocation and management of water resources. Taken Hubei province, China as an example, we adopted the exploratory spatial data analysis (ESDA) method to investigate the spatial dependence and local patterns of the WUP from 2003 to 2012. Subsequently, the spatial variation mechanisms were analyzed through the gravity center model. The results indicated that the overall spatial dependence of the agricultural WUP was detected (more significant after 2008). Moreover, the global spatial autocorrelation analysis results on the domestic WUP showed statistical significance (Moran’s I > 0.1, P < 0.05). These indicated that the local patterns were presented. The high values clustering areas of the agricultural and domestic WUP were mainly distributed in the central province and in the western province respectively. However, the approximate random distribution was identified for the industrial WUP because the industrial development had been conducted widely in the whole province during these years. Furthermore, the governmental policies and natural environment contributed to the spatial evolution tendency of the WUP. An increasing trend of the spatial association of the agricultural WUP and a significant decreasing trend of that of the domestic WUP, which suggested that the natural circumstance superiority and the industrial structure adjustment related to water utilization has been utilized and implemented effectively. This study can provide a useful reference and guidance for scientific planning of water resource systems.

A Study of Fuzzy and Non-fuzzy clustering algorithms on Wine Data

Data clustering has been considered as the most important exploratory data analysis method used to extract the unknown valuable information from the large volume of data for many real time applications in Data Mining technology. Most of the clustering techniques proved their efficiency in many fields such as decision making systems, medical sciences, earth sciences, etc. partition based clustering is one of the main approaches used in clustering. This work reports the results of classification performance of four such widely used algorithms namely K-means (KM) or Hard c-means, Fuzzy C-means, Fuzzy Possibilistic c-Means (FPCM) and Possibilistic Fuzzy c-Means (PFCM) clustering algorithms. Well known data set from UCI machine learning repository are considered to test the algorithms. The efficiency of clustering output is compared with the results observed from the repository. The experimental results demonstrate that FCM, FPCM and PFCM gives the similar percentage of correctness and classification performance.FCM, FPCM and PFCM results are better than K-means. The experimental results prove that fuzzy clustering algorithms are better than non-fuzzy clustering algorithm.

Spatial Characteristics and Driving Factors of Provincial Wastewater Discharge in China

Based on the increasing pressure on the water environment, this study aims to clarify the overall status of wastewater discharge in China, including the spatio-temporal distribution characteristics of wastewater discharge and its driving factors, so as to provide reference for developing “emission reduction” strategies in China and discuss regional sustainable development and resources environment policies. We utilized the Exploratory Spatial Data Analysis (ESDA) method to analyze the characteristics of the spatio-temporal distribution of the total wastewater discharge among 31 provinces in China from 2002 to 2013. Then, we discussed about the driving factors, affected the wastewater discharge through the Logarithmic Mean Divisia Index (LMDI) method and classified those driving factors. Results indicate that: (1) the total wastewater discharge steadily increased, based on the social economic development, with an average growth rate of 5.3% per year; the domestic wastewater discharge is the main source of total wastewater discharge, and the amount of domestic wastewater discharge is larger than the industrial wastewater discharge. There are many spatial differences of wastewater discharge among provinces via the ESDA method. For example, provinces with high wastewater discharge are mainly the developed coastal provinces such as Jiangsu Province and Guangdong Province. Provinces and their surrounding areas with low wastewater discharge are mainly the undeveloped ones in Northwest China; (2) The dominant factors affecting wastewater discharge are the economy and technological advance; The secondary one is the efficiency of resource utilization, which brings about the unstable effect; population plays a less important role in wastewater discharge. The dominant driving factors affecting wastewater discharge among 31 provinces are divided into three types, including two-factor dominant type, three-factor leading type and four-factor antagonistic type. In addition, the proposals aimed at reducing the wastewater discharge are provided on the basis of these three types.

Characterising the nature of primary care patient safety incident reports in the England and Wales National Reporting and Learning System: a mixed-methods agenda-setting study for general practice

BackgroundThere is an emerging interest in the inadvertent harm caused to patients by the provision of primary health-care services. To date (up to 2015), there has been limited research interest and few policy directives focused on patient safety in primary care. In 2003, a major investment was made in the National Reporting and Learning System to better understand patient safety incidents occurring in England and Wales. This is now the largest repository of patient safety incidents in the world. Over 40,000 safety incident reports have arisen from general practice. These have never been systematically analysed, and a key challenge to exploiting these data has been the largely unstructured, free-text data.AimsTo characterise the nature and range of incidents reported from general practice in England and Wales (2005–13) in order to identify the most frequent and most harmful patient safety incidents, and relevant contributory issues, to inform recommendations for improving the safety of primary care provision in key strategic areas.MethodsWe undertook a cross-sectional mixed-methods evaluation of general practice patient safety incident reports. We developed our own classification (coding) system using an iterative approach to describe the incident, contributory factors and incident outcomes. Exploratory data analysis methods with subsequent thematic analysis was undertaken to identify the most harmful and most frequent incident types, and the underlying contributory themes. The study team discussed quantitative and qualitative analyses, and vignette examples, to propose recommendations for practice.Main findingsWe have identified considerable variation in reporting culture across England and Wales between organisations. Two-thirds of all reports did not describe explicit reasons about why an incident occurred. Diagnosis- and assessment-related incidents described the highest proportion of harm to patients; over three-quarters of these reports (79%) described a harmful outcome, and half of the total reports described serious harm or death (n = 366, 50%). Nine hundred and ninety-six reports described serious harm or death of a patient. Four main contributory themes underpinned serious harm- and death-related incidents: (1) communication errors in the referral and discharge of patients; (2) physician decision-making; (3) unfamiliar symptom presentation and inadequate administration delaying cancer diagnoses; and (4) delayed management or mismanagement following failures to recognise signs of clinical (medical, surgical and mental health) deterioration.ConclusionsAlthough there are recognised limitations of safety-reporting system data, this study has generated hypotheses, through an inductive process, that now require development and testing through future research and improvement efforts in clinical practice. Cross-cutting priority recommendations include maximising opportunities to learn from patient safety incidents; building information technology infrastructure to enable details of all health-care encounters to be recorded in one system; developing and testing methods to identify and manage vulnerable patients at risk of deterioration, unscheduled hospital admission or readmission following discharge from hospital; and identifying ways patients, parents and carers can help prevent safety incidents. Further work must now involve a wider characterisation of reports contributed by the rest of the primary care disciplines (pharmacy, midwifery, health visiting, nursing and dentistry), include scoping reviews to identify interventions and improvement initiatives that address priority recommendations, and continue to advance the methods used to generate learning from safety reports.FundingThe National Institute for Health Research Health Services and Delivery Research programme.

Empirical Analysis of the Effects of Parasitic Magnetic Stray Field on Force Output of Electromagnetic Actuators

Electromagnetic actuators are used in a variety of technical applications especially in the automotive industry. In-line process control methods are an essential component of the Lean and Six Sigma methodology to ensure process quality. However, the current state of the art in process and quality control is largely limited to end-of-line measurements of the force output. Analysing the magnetic stray field is a promising method that can be used to draw conclusions on the properties and defects of the flux-conducting magnetic materials. This phenomenon can potentially be used to identify defects in magnetic actuators thus allowing inline quality-monitoring. In order to realize this feature, patterns in the magnetic stray field of an actuator have to be identified and linked to a specific defect. The resulting challenge is the analysis of large datasets in order to characterize the stray field anomalies. This paper summarizes the results of a study on linear magnetic actuators trying to prove a relationship between parasitic magnetic stray field and the overall force output of an actuator by analysing the data with statistical methods. The findings of this study suggest that certain statistical methods, like regression, are not well suited to build a prediction model for defects in actuators using a similar approach of measuring stray field outside the actuator. This is mainly due to the fact that prerequisites for model building are difficult to full fill within the context of stray field analysis. Nevertheless, the findings also suggest that methods of exploratory data analysis can be used to derive quality relevant information from data of stray field measurements. The paper elaborates on the problem of defining a population, choosing variables for model building, as well as model error.

Exploratory data analysis as a foundation of inductive research

Across academic disciplines, scientific progress is maximized when there is a balance between deductive and inductive approaches. To promote this balance in organizational science, rigorous inductive research aimed at phenomenon detection must be further encouraged. To this end, the present article discusses the logic and methods of exploratory data analysis (EDA), the mode of analysis concerned with discovery, exploration, and empirically detecting phenomena in data. We begin by first describing the historical and conceptual background of EDA. We then discuss two issues related to EDA and its relationship to scientific credibility. First, we argue that EDA fosters a replication-based science by requiring cross-validation and by emphasizing the natural uncertainty of data patterns. Second, we clarify that EDA is distinguishable from other exploratory practices that are considered scientifically questionable (e.g., “p-hacking”, “data fishing” and “data-dredging”). In the following section of the paper, we present a final argument for EDA: that it helps maximize the value of data. To illustrate this point, we present several graphical methods for detecting data patterns and provide references to further techniques for the interested reader.

Detection of spatial hot spots and variation for the neon flying squid Ommastrephes bartramii resources in the northwest Pacific Ocean

With the increasing effects of global climate change and fishing activities, the spatial distribution of the neon flying squid (Ommastrephes bartramii) is changing in the traditional fishing ground of 150°–160°E and 38°–45°N in the northwest Pacific Ocean. This research aims to identify the spatial hot and cold spots (i.e. spatial clusters) of O. bartramii to reveal its spatial structure using commercial fishery data from 2007 to 2010 collected by Chinese mainland squid-jigging fleets. A relatively strongly-clustered distribution for O. bartramii was observed using an exploratory spatial data analysis (ESDA) method. The results show two hot spots and one cold spot in 2007 while only one hot and one cold spots were identified each year from 2008 to 2010. The hot and cold spots in 2007 occupied 8.2% and 5.6% of the study area, respectively; these percentages for hot and cold spot areas were 5.8% and 3.1% in 2008, 10.2% and 2.9% in 2009, and 16.4% and 11.9% in 2010, respectively. Nearly half (>45%) of the squid from 2007 to 2009 reported by Chinese fleets were caught in hot spot areas while this percentage reached its peak at 68.8% in 2010, indicating that the hot spot areas are central fishing grounds. A further change analysis shows the area centered at 156°E/43.5°N was persistent as a hot spot over the whole period from 2007 to 2010. Furthermore, the hot spots were mainly identified in areas with sea surface temperature (SST) in the range of 15–20°C around warm Kuroshio Currents as well as with the chlorophyll-a (chl-a) concentration above 0.3 mg/m3. The outcome of this research improves our understanding of spatiotemporal hotspots and its variation for O. bartramii and is useful for sustainable exploitation, assessment, and management of this squid.

Application of clustering analysis in traditonal Chinese medicine research

As one of exploratory data analysis method, the clustering analysis can provide technical support for objectivity and standardization ofTCM. This paper dealt with the application of the clustering analysis in the research of Chinese medcine, TCM pattern differentiation and the regularity of TCM herbal formulae. The clustering analysis showed great value on classifying imformation and finding out regularity from data, only if the characteristics of TCM data were considered. Key words: Clustering analysis; Traditional Chinese medicine; TCM pattern differentiation; Regularity of TCM herbal formulae

Hierarchical clustering and partitioning to characterize shrimp grow-out farms in northeast Brazil

Shrimp farming in northeast Brazil is an important sector of the regional economies, contributing as a source of revenue and poverty alleviation. In recent years, shrimp aquaculture has faced many challenges, ranging from the occurrence of viral diseases to changes in market access. Considering the past and present challenges faced by the shrimp farmers in northeast Brazil there is a current need to identify the present trends in shrimp farming when striving for the sustainable development of this important sector of the Brazilian economy. To this aim, the exploratory data analysis methods of multiple correspondence analysis and hierarchical and partitional clustering were applied to the data collected through a national census: by identifying shrimp farm clusters, we can characterize groups and identify trends that can compromise sustainable production. The present study identified not only trends in shrimp farming but also considered variables excluded from previous publications for the national census data. The general and initially well-known division between the medium and large-scale, highly tecnified shrimp farms and the micro-scale and less tecnified farms, was verified. However, two clusters of micro-scale and less tecnified farms were observed, highlighting differences in farm management and productivity and also how farmer's view shrimp farming, their concerns and their awareness of what compose biosecurity practices and which biosecurity practices they considered most important. The different stakeholders can make use of the results from the analysis and contribute to the improvement of management practices and policy-making for sustainable shrimp farming in northeast Brazil. Statement of relevanceShrimp farming is an important source of income in Brazil's northeast states. Many challenges were faced by this important economic sector and the now extinct Ministry of Fisheries and the Brazilian Association for Shrimp Farming performed a national census of shrimp farms, visiting each farm individually to meet with farmers and record information on many variables. A great deal of the information generated through the census remained unaccounted for in the publication released by these organizations and so multiple correspondence analysis as a data reduction method was used to overcome the challenge of making further use of the data set and that output analyzed through clustering methodology. Trends in shrimp farming can be identified and the information can make for better understanding of shrimp farming to help secure sustainable development through better management and policy making.

우리나라와 유럽의 근로환경조사 비교연구

근래에 들어 기업 간 경쟁이 심화되고, 경제상황이 불안해지면서 근로시간의 단축은 물론 비정규직이나 시간제 근로자의 양산, 새로운 탄력근무제나 교대제의 도입 등이 일어났고, 따라서 근로시장은 급속한 변화를 겪고 있다. 이에 이 연구에서는 우선 탐색적 자료분석 방법을 이용하여 KWCS(Korean Working Conditions Survey)와 EWCS (European Working Conditions Survey) 데이터로부터 근로환경의 측면에서 근로자의 사회 인구학적 특성이 어떤지 살펴보고자 한다. 이어서 KWCS와 EWCS의 데이터를 가지고 탐색적 자료분석 방법으로 주당 근로시간이 성별, 연령별, 종사상 지위별, 직종별, 산업별로 얼마나 차이가 나는지 알아보고자 한다. 마지막으로 근로자들을 건강이 적어도 좋은 편에 속하는 사람들과 건강이 적어도 나쁜 편에 속하는 사람들로 나누어 두 그룹 간 주당 평균근로시간이 성별, 연령별, 종사상 지위별, 직종별, 산업별로 얼마나 차이가 나는지 표를 통해 먼저 살펴보고, 주당 평균근로시간을 반응변수로 보고 이에 영향을 미치는 요인들 (건강, 성, 연령, 종사상 지위 등)의 효과를 점검하는 단순회귀 및 다중회귀 분석을 실시하고자 한다. Global competition and unstable economic situation in recent years have brought in adverse working conditions such as proliferation of temporary or part-time workers and new shift system. In this study we use exploratory data analysis method to find the relationship between some of the important variables which are related to working hours in KWCS (Korean Working Conditions Survey) and EWCS (European Working Conditions Survey). Next, working hours are explored in terms of sex, employment type, occupation and industry using 2006, 2010, 2011, 2014 KWCS data as well as 2010 EWCS data. Lastly, respondents are divided into two groups of healthy and nonhealthy workers and their working hours are explored in terms of the same categories as above. Finally, simple regression analysis is used to find the effect of health on weekly working hours and adjustment of the effect of health on weekly working hours is made using multiple regression analysis with some other independent variable such as sex, age and employment type included in the model.

Delineation of Geochemical Anomalies Based on Cu by the Boxplot as an Exploratory Data Analysis (EDA) Method and Concentration-Volume (C-V) Fractal Modeling in Mesgaran Mining Area, Eastern Iran

The target in this investigation is separation and delineation of geochemical anomalies for the single element Cu in Mesgaran mining area, eastern Iran. Mesgaran mining area is located in south part of Sarbishe county with about 29 Km distance to the county center. This region is part of an Ophiolite sequence and the copper anomalies seem to be related to a volcanic massive sulfide (VMS) deposit whose main part (massive sulfide Lens) has been eroded. In order to delineate Cu anomalies, the boxplot as an Exploratory Data Analysis (EDA) method and concentration-volume (C-V) Fractal modeling are employed. Both of the methods reveal low-deep anomalies which are highly correlated with geological and geophysical studies. As the main result of this study we show that Fractal modeling in spite of the Boxplot, is not recommended for complex geological settings. The proved shallow anomalies recorded by geophysical studies and defined by the used methods are in accordance to the stringer zone of a volcanic massive sulfide (VMS) deposit in Mesgaran mining area which means this region is the bottom of a VMS deposit and geochemical anomalies are related to the remained parts of the deposit.

Exploratory Method for Spatio-Temporal Feature Extraction and Clustering: An Integrated Multi-Scale Framework

This paper presents an integrated framework for exploratory multi-scale spatio-temporal feature extraction and clustering of spatio-temporal data. The framework combines the multi-scale spatio-temporal decomposition, feature identification, feature enhancing and clustering in a unified process. The original data are firstly reorganized as multi-signal time series, and then decomposed by the multi-signal wavelet. Exploratory data analysis methods, such as histograms, are used for feature identification and enhancing. The spatio-temporal evolution process of the multi-scale features can then be tracked by the feature clusters based on the data adaptive Fuzzy C-Means Cluster. The approach was tested with the global 0.25° satellite altimeter data over a period of 21 years from 1993 to 2013. The tracking of the multi-scale spatio-temporal evolution characteristics of the 1997–98 strong El Niño were used as validation. The results show that our method can clearly reveal and track the spatio-temporal distribution and evolution of complex geographical phenomena. Our approach is efficient for global scale data analysis, and can be used to explore the multi-scale pattern of spatio-temporal processes.