An accurate and efficient intelligent fault diagnosis of mobile robotic roller bearings can significantly enhance the reliability and safety of mechanical systems. To improve the efficiency of intelligent fault classification of mobile robotic roller bearings, this paper proposes a parallel machine learning algorithm using fine-grained-mode Spark on a Mesos big data cloud computing software framework. Through the segmentation of datasets and the support of a parallel framework, the parallel processing technology Spark is combined with a support vector machine (SVM), and a parallel single-SVM algorithm is realized using Scala language. In this approach, empirical mode decomposition (EMD) is applied to extract the energy of the acceleration vibration signal in different frequency bands as features. The parallel EMD-SVM approach is applied to detect faults in mobile robotic roller bearings from fault vibration signals. The experimental results show that it can accurately and effectively identify the faults, and it outperforms existing methods based on parallel deep belief network (DBN) and parallel radial basis function neural network under different training set sizes. Fault classification tests are conducted on outer-race and inner-race faults: in both cases, the proposed parallel EMD-SVM outperforms the serial EMD-SVM in terms of the classification accuracy and classification time under different test sizes. For a small number of nodes, the processing time of the proposed Spark model is less than that of Hadoop but close to that of Storm. For 17 slave nodes, the average precision, average recall, and average F1 score of Spark on Mesos in the fine-grained mode reach 97.3, 97.8, and 97.9%, respectively. The parallel EMD-SVM algorithm in the big data Spark cloud computing framework can improve the accuracy of intelligent fault classification, albeit by a small margin, with higher processing speed and learning convergence rate.

/pdf/parallel-machine-learning-algorithm-using-fine-grained-mode-4wr2ra7asm.pdf

Parallel Machine Learning Algorithm Using Fine-Grained-Mode Spark on a Mesos Big Data Cloud Computing Software Framework for Mobile Robotic Intelligent Fault Recognition

Uber Data using R programming, we can predict the fare, reduce timing and choose locations based on the heat map. We are creating a website for UBER DATA. Effective taxi dispatching will allow each driver and passenger to spend less time looking for each other. The model is used to forecast demand at various locations throughout the city. In an urban area of any metropolitan city or normal city, people are unable to drive or are most not likely to take their vehicles due to rides. All the people are most likely to take buses, subwaysor taxis. The three modes of transportation are important but have different features. The main difference among all these is fare (way to pay). So, Uber is very popular nowadays. The main aim of this project is that we are creating a website of Uber data so, all the people who are willing to take uber taxis can easily find the data which is on their mobile itself. 

An Empirical Data Analytics and Visualization for UBER Services: A Data Analysis Based Web Search Engine

MSGC: Multi-scale grid clustering by fusing analytical granularity and visual cognition for detecting hierarchical spatial patterns

Cryo-electron tomography (cryo-ET) is uniquely suited to precisely localize macromolecular complexes in situ, that is in a close-to-native state within their cellular compartments, in three-dimensions at high resolution. Point pattern analysis (PPA) allows quantitative characterization of the spatial organization of particles. However, current implementations of PPA functions are not suitable for applications to cryo-ET data because they do not consider the real, typically irregular 3D shape of cellular compartments and molecular complexes. Here, we designed and implemented first and the second-order, uni- and bivariate PPA functions in a Python package for statistical spatial analysis of particles located in three dimensional regions of arbitrary shape, such as those encountered in cellular cryo-ET imaging (PyOrg). 
To validate the implemented functions, we applied them to specially designed synthetic datasets. This allowed us to find the algorithmic solutions that provide the best accuracy and computational performance, and to evaluate the precision of the implemented functions. Applications to experimental data showed that despite the higher computational demand, the use of the second-order functions is advantageous to the first-order ones, because they allow characterization of the particle organization and statistical inference over a range of distance scales, as well as the comparative analysis between experimental groups comprising multiple tomograms. 
Altogether, PyOrg is a versatile, precise, and efficient open-source software for reliable quantitative characterization of macromolecular organization within cellular compartments imaged in situ by cryo-ET, as well as to other 3D imaging systems where real-size particles are located within regions possessing complex geometry.

Statistical spatial analysis for cryo-electron tomography

Clustering is a powerful machine learning method for discovering similar patterns according to the proximity of elements in feature space. It is widely used in computer science, bioscience, geoscience, and economics. Although the state-of-the-art partition-based and connectivity-based clustering methods have been developed, weak connectivity and heterogeneous density in data impede their effectiveness. In this work, we propose a boundary-seeking Clustering algorithm using the local Direction Centrality (CDC). It adopts a density-independent metric based on the distribution of K-nearest neighbors (KNNs) to distinguish between internal and boundary points. The boundary points generate enclosed cages to bind the connections of internal points, thereby preventing cross-cluster connections and separating weakly-connected clusters. We demonstrate the validity of CDC by detecting complex structured clusters in challenging synthetic datasets, identifying cell types from single-cell RNA sequencing (scRNA-seq) and mass cytometry (CyTOF) data, recognizing speakers on voice corpuses, and testifying on various types of real-world benchmarks. 

/pdf/clustering-by-measuring-local-direction-centrality-for-data-1aa6ep48.pdf

Clustering by measuring local direction centrality for data with heterogeneous density and weak connectivity

Optimizing and accelerating space–time Ripley ’s K function based on Apache Spark for distributed spatiotemporal point pattern analysis

. Enterprises are the basis of urban economic development and an essential factor that affects urban structure. It is of great significance to study the spatiotemporal distribution of enterprises to benefit cities planning and development. Based on the enterprises registration data of Administrator for Industry and Commerce of Wuhan from 1996 to 2007, in this paper, we analysed the spatiotemporal distribution of enterprises in Wuhan city. We divided Wuhan into 2,356 square grids with a length of 2 km and counted the number of enterprises of different industry categories in each grid. Then we calculated the aggregation intensity index of enterprises in each grid. With the method of global spatial autocorrelation analysis, we discovered the change of the degree of enterprise agglomeration for different industry categories in different periods. We also used local autocorrelation analysis to find the hotspots of enterprise distribution areas furtherly and discover the changes of the location of the hotspots during each period. Based on the two analyses and context information of Wuhan city, we can know the features of spatiotemporal distribution of enterprises. It can be concluded that the spatial distribution of enterprises could be reflected directly by spatial autocorrelation analysis. Moreover, the enterprises registration data with fine-grained enterprises spatiotemporal information can support the research of spatial distribution and evolution of enterprises. This study can show how different industry categories developed and provide reference to urban geography to help cities plan the distribution of enterprises and help enterprises decide its location.

/pdf/analyzing-the-spatiotemporal-distribution-of-different-1dkppyu15m.pdf

Analyzing the Spatiotemporal Distribution of Different Industries in Wuhan City Using Enterprise Registration Data

The boom in publicly accessible big data sources, such as social networks and the internet of things, is creating new opportunities as well as new technologies and applications to study and understand socioeconomic activities. Visual analytics is promising social sensing approach that helps users effectively identify interesting events and discover hidden patterns, anomalies, and relations from datasets. Big data however also represents a challenge to current data processing and analytical technologies, given the heterogeneity of multi-source data, extremely large data volume and intensive computation impede the efficiency and feasibility of visual analysis. In this chapter, we introduce cutting-edge data storage, computing, and visualization technologies that can tackle these challenges. Data storage and index mechanism optimize query and reduce response time for multi-dimensional big data access; high Performance Computing (HPC) technology accelerates big data preprocessing and mining; web-based interactive visualization technologies allow users to synthesize and manipulate data in a dynamic and user-friendly manner. Taking big enterprise registration data as an example, we demonstrate how to apply these new technologies to build a web-based visual analytics framework that supports studies in economic geography. The necessity and value of these technologies for big socioeconomic data processing and visualization is analyzed and discussed. We conclude, by analyzing and discussing the versatility and portability of these technologies in other application scenarios such as in the humanities and social sciences.

High Performance Spatiotemporal Visual Analytics Technologies and Its Applications in Big Socioeconomic Data Analysis

. Ripley’s K functions are powerful tools for studying the spatial arrangement or spatiotemporal distribution characteristics of geographic phenomena and events in spatial analysis and has been used in many fields. However, the K functions are compute-intensive for point-wise distance comparisons, edge correction and simulations for significance test. Although parallel computing technologies have been adopted to accelerate K functions, previous works haven’t extended the optimization from space to space-time dimension. This study presents an acceleration method for K functions upon state-of-the-art distributed computing framework Apache Spark, and four optimization strategies are leveraged to simplify calculation procedures and accelerate distributed computing respectively, including 1) spatiotemporal indexing based on R-tree with Sort-Tile-Recursive (STR) algorithm for reducing distance comparison when retrieving potential spatiotemporally neighbouring points; 2) Hash-Table-based caching for spatiotemporal edge correction weights reuse and reducing repetitive computation; 3) Spatiotemporal partitioning using KDB-tree as well as cylinder intersection redundancy strategy for decreasing ghost buffer redundancy in partitions and supporting near-balanced distributed processing; 4) Customized serialization of spatiotemporal objects and indexes for lowering the overhead of data transmission. Experiments verify the effectiveness and time efficiency of the proposed optimization strategies, and also evaluate the overall performance and scalability. Based on the proposed methods, a web-based visual analytics framework has been developed and publicly shared through GitHub, and four types of the distributed K functions are implemented, including space, space-time, local and cross K functions, which demonstrates its value on promoting geographical and socioeconomic studies.

/pdf/developing-apache-spark-based-ripley-s-k-functions-for-53tfqur4wu.pdf

Developing apache spark based ripley’s k functions for accelerating spatiotemporal point pattern analysis

Background Adolescents' suicidality and peer bullying are rising as a serious public health concern globally. However, the neural correlates of perpetrators and the impact of different types of bullying perpetration/peer victimisation on preadolescents needs elucidation. Besides, there has been a lack of research studying the cumulative risk pattern of bullying perpetration/peer victimisation on preadolescents with suicidality. Methods We adopted a retrospective and longitudinal methodology by utilising the data from the Adolescent Brain Cognitive Development (ABCD) cohort. Firstly, participants were assigned into two groups (i.e., perpetrators vs non-perpetrators) based on their bullying perpetration experiences. Next, different types of bullying perpetration/peer victimisation (i.e., overt, relational, and reputational) were extracted to evaluate their associations with suicidal ideation (SI), non-suicidal self-injury (NSSI), and suicide attempt (SA) separately. Then, participants were further assigned into four groups (i.e., bully-uninvolved preadolescents without suicidality/NSSI history (HC group), bully-uninvolved preadolescents with SA (SA group), preadolescents with both SA and bullying perpetration (SA+P group), and preadolescents with both SA and peer victimisation (SA+V group)). We used multinomial logistic regression models, analysis of variance (ANOVA), and brain network analysis for assessing potential associations in study's objectives. Findings In total, 9992 individuals aged 11-12 years were included in our analysis. Of these individuals, 1111 (11.1%) were classified as perpetrators, and had significantly lower bilateral cortical volume in the superior frontal sulcus (SFS), lower functional connectivity within default mode network (DMN), and lower anti-correlation between DMN and dorsal attention network (DAN) than non-perpetrators. Additionally, preadolescents' overt, relational, and reputational perpetration/ victimisation experiences were all shown to be associated with suicidality/NSSI, despite varying strengths, while overt perpetration showed the strongest association with SA (OR 3.6 [95%CI 2.4-5.4]), followed by overt victimisation (2.4 [1.6-3.5]). Besides, the SA+V group was characterised by the highest psychopathology among four groups, while the SA+P group was characterised by the highest aggression among four groups. Lastly, lower bilateral cortical volume in the precentral gyrus (PCG) while higher cortical volume in both the lateral occipitotemporal cortex (LOTC) and posterior cingulate cortex (PCC) were found in the SA+P group when compared to SA and HC group. Interpretation Findings from the present study offered empirical evidence on the impact of bullying experience, suicidality and their cumulative risk on preadolescents' behavior and brain development, contributing to the growing literature on discerning the impact of different types of bullying perpetration/peer victimisation on preadolescents.

Zousen Cui

Papers

Optimizing and accelerating space–time Ripley ’s K function based on Apache Spark for distributed spatiotemporal point pattern analysis

Analyzing the Spatiotemporal Distribution of Different Industries in Wuhan City Using Enterprise Registration Data

High Performance Spatiotemporal Visual Analytics Technologies and Its Applications in Big Socioeconomic Data Analysis

Developing apache spark based ripley’s k functions for accelerating spatiotemporal point pattern analysis

Bullying perpetration, peer victimisation, suicidality, and their cumulative effect on preadolescents' behaviour and brain development