Different Views Of Data Research Articles

Personalized Edge Intelligence via Federated Self-Knowledge Distillation

<italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Federated Learning</i> (FL) is an emerging approach in edge computing for collaboratively training machine learning models among multiple devices, which aims to address limited bandwidth, system heterogeneity, and privacy issues in traditional centralized training. However, the existing federated learning methods focus on learning a shared global model for all devices, which may not always be ideal for different devices. Such situations become even worse when each edge device has its own data distribution or task. In this paper, we study personalized federated learning in which our goal is to train models to perform well for individual clients. We observe that the initialization in each communication round causes the forgetting of historical personalized knowledge. Based on this observation, we propose a novel <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Personalized Federated Learning</i> (PFL) framework via self-knowledge distillation, named pFedSD. By allowing clients to distill the knowledge of previous personalized models to current local models, pFedSD accelerates the process of recalling the personalized knowledge for the latest initialized clients. Moreover, self-knowledge distillation provides different views of data in feature space to realize an implicit ensemble of local models. Extensive experiments on various datasets and settings demonstrate the effectiveness and robustness of pFedSD.

Multi-omic analysis reveals enriched pathways associated with COVID-19 and COVID-19 severity.

COVID-19 is a disease characterized by its seemingly unpredictable clinical outcomes. In order to better understand the molecular signature of the disease, a recent multi-omics study was done which looked at correlations between biomolecules and used a tree- based machine learning approach to predict clinical outcomes. This study specifically looked at patients admitted to the hospital experiencing COVID-19 or COVID-19 like symptoms. In this paper we examine the same multi-omics data, however we take a different approach, and we identify stable molecules of interest for further pathway analysis. We used stability selection, regularized regression models, enrichment analysis, and principal components analysis on proteomics, metabolomics, lipidomics, and RNA sequencing data, and we determined key molecules and biological pathways in disease severity, and disease status. In addition to the individual omics analyses, we perform the integrative method Sparse Multiple Canonical Correlation Analysis to analyse relationships of the different view of data. Our findings suggest that COVID-19 status is associated with the cell cycle and death, as well as the inflammatory response. This relationship is reflected in all four sets of molecules analyzed. We further observe that the metabolic processes, particularly processes to do with vitamin absorption and cholesterol are implicated in COVID-19 status and severity.

Identifying multimodal imaging‐driven subtypes in mild cognitive impairment using deep multiview learning

AbstractBackgroundMultimodal neuroimaging data can provide complementary information that a single modality cannot about neurodegenerative diseases such as Alzheimer's disease (AD). Deep Generalized Canonical Correlation Analysis (DGCCA) is able to learn a shared feature representation from different views of data by applying non‐linear transformation using neural network. We utilize DGCCA to extract maximally correlated components from multi‐modal neuroimaging data to identify potential imaging‐driven MCI subtypes.MethodWe study 308 Mild Cognitive Impairment (MCI) participants (195 early MCI and 113 late MCI) from the Alzheimer’s Disease Neuroimaging Initiative (ADNI), each with voxel level features from FDG PET, amyloid PET (AV45) and structural MRI processed using voxel‐based morphometry (VBM). Six experimental settings are designed to compare single modality with multiview methods ‐ GCCA and DGCCA, see Figure 1. Agglomerative clustering was used to generated 2 clusters with features from each experiment. To investigate differences between the clusters, Wilcoxon rank‐sum tests are conducted on 11 baseline AD biomarkers including 5 cognitive assessments and 6 brain volume measures, from the ADNI QT‐PAD dataset http://www.pi4cs.org/qt‐pad‐challenge.ResultAmong the two multiview methods, DGCCA is able to explain 68.57% variance with 20 features, while GCCA explains 68.66% variance with 94 features. To evaluate the potential subtypes from clustering, the Calinski‐Harabasz (CH) score, Silhouette score and adjusted mutual information (AMI) score are computed, see Table 1. AV45 generates the best defined clusters, where DGCCA generates clusters with quality comparable to single modality features. In our QT analysis, clusters from FDG and DGCCA features show differential measure in all biomarkers where DGCCA learns from multimodal data, see Figure 2.ConclusionDGCCA is able to learn maximally correlated features from multimodal neuroimaging data with reduced dimensionality, and explain more variance than its linear counterpart GCCA. Cluster analysis shows these imaging‐driven MCI subtypes are different from the currently diagnosis with differential QT measures, by incorporating complementary information from 3 imaging modalities. DGCCA shows to be an effective feature learning method, and this multiview learning framework can identify potentially novel MCI subtypes to facilitate early detection of AD.

A generalized kernel machine approach to identify higher-order composite effects in multi-view datasets, with application to adolescent brain development and osteoporosis

In recent years, a comprehensive study of complex disease with multi-view datasets (e.g., multi-omics and imaging scans) has been a focus and forefront in biomedical research. State-of-the-art biomedical technologies are enabling us to collect multi-view biomedical datasets for the study of complex diseases. While all the views of data tend to explore complementary information of disease, analysis of multi-view data with complex interactions is challenging for a deeper and holistic understanding of biological systems. In this paper, we propose a novel generalized kernel machine approach to identify higher-order composite effects in multi-view biomedical datasets (GKMAHCE). This generalized semi-parametric (a mixed-effect linear model) approach includes the marginal and joint Hadamard product of features from different views of data. The proposed kernel machine approach considers multi-view data as predictor variables to allow a more thorough and comprehensive modeling of a complex trait. We applied GKMAHCE approach to both synthesized datasets and real multi-view datasets from adolescent brain development and osteoporosis study. Our experiments demonstrate that the proposed method can effectively identify higher-order composite effects and suggest that corresponding features (genes, region of interests, and chemical taxonomies) function in a concerted effort. We show that the proposed method is more generalizable than existing ones. To promote reproducible research, the source code of the proposed method is available at.

Multi-View Representation Learning With Deep Gaussian Processes.

Multi-view representation learning is a promising and challenging research topic, which aims to integrate multiple data information from different views to improve the learning performance. The recent deep Gaussian processes (DGPs) have the advantages of good uncertainty estimates, powerful non-linear mapping ability and great generalization capability, which can be used as an excellent data representation learning method. However, DGPs only focus on single view data and are rarely applied to the multi-view scenario. In this paper, we propose a multi-view representation learning algorithm with deep Gaussian processes (named MvDGPs), which inherits the advantages of deep Gaussian processes and multi-view representation learning, and can learn more effective representation of multi-view data. The MvDGPs consist of two stages. The first stage is multi-view data representation learning, which is mainly used to learn more comprehensive representations of multi-view data. The second stage is classifier design, which aims to select an appropriate classifier to better employ the representations obtained in the first stage. In contrast with DGPs, MvDGPs support asymmetrical modeling depths for different views of data, resulting in better characterizations of the discrepancies among different views. Experimental results on real-world multi-view data sets verify the effectiveness of the proposed algorithm, which indicates that MvDGPs can integrate the complementary information in multiple views to discover a good representation of the data.

Research of algorithms of Data Mining

The article explores data mining algorithms, which based on rules and calculations, that allow us to create a model that analyzes the data provided by searching for specific patterns and trends. The purpose of this work is to analyze correlation-regression algorithms on a statistical dataset of chronic diseases. Data mining allows building many models, multiple algorithms can be used within a single solution. The article explores the algorithms of clustering, correlation analysis, Naive Bayes algorithm for obtaining different views of data. Since diabetes is one of the most dangerous chronic diseases, the pathogenesis of which is a lack of insulin in the human body, which causes metabolic disorders and pathological changes in various organs and tissues. As a result, it leads to disability of all functional systems of the body. It was decided to investigate the data related to this disease. Also, the quality of the developed methods of information retrieval from the dataset was evaluated and the most informative features were identified. The developed methods were implemented in the system of intellectual data processing. Past studies show promise of using data mining methods to improve the quality of patient care.

Robust Multiview Subspace Learning With Nonindependently and Nonidentically Distributed Complex Noise.

Multiview Subspace Learning (MSL), which aims at obtaining a low-dimensional latent subspace from multiview data, has been widely used in practical applications. Most recent MSL approaches, however, only assume a simple independent identically distributed (i.i.d.) Gaussian or Laplacian noise for all views of data, which largely underestimates the noise complexity in practical multiview data. Actually, in real cases, noises among different views generally have three specific characteristics. First, in each view, the data noise always has a complex configuration beyond a simple Gaussian or Laplacian distribution. Second, the noise distributions of different views of data are generally nonidentical and with evident distinctiveness. Third, noises among all views are nonindependent but obviously correlated. Based on such understandings, we elaborately construct a new MSL model by more faithfully and comprehensively considering all these noise characteristics. First, the noise in each view is modeled as a Dirichlet process (DP) Gaussian mixture model (DPGMM), which can fit a wider range of complex noise types than conventional Gaussian or Laplacian. Second, the DPGMM parameters in each view are different from one another, which encodes the "nonidentical" noise property. Third, the DPGMMs on all views share the same high-level priors by using the technique of hierarchical DP, which encodes the "nonindependent" noise property. All the aforementioned ideas are incorporated into an integrated graphics model which can be appropriately solved by the variational Bayes algorithm. The superiority of the proposed method is verified by experiments on 3-D reconstruction simulations, multiview face modeling, and background subtraction, as compared with the current state-of-the-art MSL methods.

The learnability of the dimensional view of data and what to do with it

PurposeThe purpose of this paper is to examine the usability of the dimensional view of data in the context of its presumed learnability.Design/methodology/approachIn total, 303 participants were asked to solve 12 analytical problems in an experiment using the dimensional view of data for half of the problems and an operational view of data for the other half. Inferential statistics and structural equation modeling were performed with participants’ objective results and affective reactions.FindingsShowing that the order of exposure to the two views of data impacts the overall usability of ad hoc querying, the study provided evidence for the learnability potential of the dimensional view of data. Furthermore, the study showed that affective reactions to the different views of data follow objective usability parameters in a way that can be explained using models from affective computing research.Practical implicationsThe paper proposes a list of guidelines for use of the dimensional view of data in business analytics.Originality/valueThis study is the first to confirm the learnability of the dimensional view of data and the first to take a deeper look at affective reactions to an ad hoc business analytics solution. Also, it is one of few studies that examined the usability of different views of data directly on these views, rather than using paper representations of data models.

Information-Theoretic Multi-view Domain Adaptation: A Theoretical and Empirical Study

Multi-view learning aims to improve classification performance by leveraging the consistency among different views of data. The incorporation of multiple views was paid little attention in the studies of domain adaptation, where the view consistency based on source data is largely violated in the target domain due to the distribution gap between different domain data. In this paper, we leverage multiple views for cross-domain document classification. The central idea is to strengthen the views' consistency on target data by identifying the associations of domain-specific features from different domains. We present an Information-theoretic Multi-view Adaptation Model (IMAM) using a multi-way clustering scheme, where word and link clusters can draw together seemingly unrelated features across domains, which boosts the consistency between document clusterings that are based on the respective word and link views. Moreover, we demonstrate that IMAM can always find the document clustering with the minimal disagreement rate to the overlap of view-based clusterings. We provide both theoretical and empirical justifications of the proposed method. Our experiments show that IMAM significantly outperforms traditional multi-view algorithm co-training, the co-training-based adaptation algorithm CODA, the single-view transfer model CoCC and the large-margin-based multi-view transfer model MVTL-LM.

Information-theoretic multi-view domain adaptation

Multi-view learning aims to improve classification performance by leveraging the consistency among different views of data. The incorporation of multiple views was paid little attention in the stud...

Electronic medical records

and bed-request functionality not needed several floors upstairs in the same hospital. Doctors and nurses may want to see different views of data for the same patients. Administrators and researchers interested in mining data to improve care will have further requirements. Highly trained physicians in wellequipped hospitals might be willing to use complex interfaces, while rural health clinics in low-resource countries may need interfaces that can easily be used with much less computer experience. Although systems developed in these environments may differ drastically, lessons learned may transfer. Just as clinical decision-support systems developed in research hospital environments might be adapted to provide guidance to clinical health workers in low-resource settings (both in developing countries and in under-served regions of industrial nations), simplified interfaces created for use in those low-resource settings might help developers fight the staggering complexity of some EMR interfaces. Overlaying these complexities on an evolving infrastructure that records. Computerized decisionsupport systems that pester users with over-abundant or irrelevant alarms are ignored, and systems that place high cognitive loads on time-stressed doctors and nurses lead to increased medical errors. Designing effective medical records requires the full complement of usability expertise, from the contextual inquiry and ethnography required for understanding user needs to the development of comprehensible information and interaction designs, and finally to the evaluations necessary for understanding successes and failures in complex medical environments. The diversity of medical care contexts complicates this alreadychallenging scenario even further. Vastly differing scales, disciplinary requirements, user perspectives, and social/cultural contexts complicate design and limit generalizability. Small, independent medical primary-care physicians may have needs that are radically different from those of physicians in large academic medical centers. Systems used in emergency departments may need to provide triage Electronic Medical Records (EMRs) have begun to transform healthcare worldwide. Some small countries like Denmark, the Netherlands, and Israel have built effective systems based on national standards, while in the U.S. dozens of diverse systems are competing for market share. In developing nations local projects can have huge effects, but the trade-offs in spending for medical facilities versus computing resources may require painfully difficult decisions. In all cases, meeting ambitious goals while maintaining high-quality care and realizing anticipated cost savings will require solving a host of problems, many of which are not yet well understood. Despite the many uncertainties, the critical importance of usability to the success of these goals is clear. Numerous studies and extensive clinical experience testify to the difficulties associated with poor usability. Systems designed by software architects who expect clinicians to adapt their workflows to meet the needs of the computers are subject to workarounds that subvert system goals and reduce the value of the electronic in te ra c ti o n s N o ve m b e r + D e c e m b e r 2 0 11

Statistics and the Modern Student

RésuméLes cours d’initiation à la statistique ont traditionellement visé les consomammateurs de la statistique avec l’intention de produire une population capable de faire une analyse critique des statistiques élémentaires publiées. Plus récemment, les professeurs de la statistique ont tenté d’orienter les cours d’initiation vers des données réelles, afin de motiver les élèves d’un part, et de créer un cours plus pertinent d’autre part. Le succès de cette approche repose sur une provision de données que les étudiants considèrent comme réels et pertinents. Cependent, les étudiants modernes ont une vision des données qui est différente de celle qu’ont eu les élèves d’il y a 10 ou même 5 ans. Les cours modernes de statistique doivent s’adapter au fait que la première rencontre des élèves aux données a lieu en dehors de l’académie.

Micro-views, or on how to protect privacy while enhancing data usability

The large availability of repositories storing various types of information about individuals has raised serious privacy concerns over the past decade. Nonetheless, database technology is far from providing adequate solutions to this problem that requires a delicate balance between an individual's privacy and convenience and data usability by enterprises and organizations - a database which is rigid and over-protective may render data of little value. Though these goals may seem odd, we argue that the development of solutions able to reconcile them will be an important challenge to be addressed in the next few years. We believe that the next wave of database technology will be represented by a DBMS that provides high-assurance privacy and security. In this paper, we elaborate on such challenges. In particular, we argue that we need to provide different views of data at a very fine level of granularity; conventional view technology is able to select only up to a single attribute value for a single tuple. We need to go even beyond this level. That is, we need a mechanism by which even a single value inside a tuple's attribute may have different views; we refer them as micro-views. We believe that such a mechanism can be an important building block, together with other mechanisms and tools, of the next wave of database technology.

Xml Data Mediation and Collaboration: A Proposed Comprehensive Architecture and Query Requirements for Using Xml to Mediate Heterogeneous Data Sources and Targets

This paper discusses the potential of using XML technologies as mediators for accessing and processing structured, relational data and semi-structured data over the Internet.Because significant gaps exist in all the reviewed XML data mediation approaches, a more comprehensive framework is presented. Some of the potential benefits of this architecture include: provides a clean abstraction of heterogeneous data for end users, applications, and Internet applications; supports all major data formats; uses one standard query language for all XML data queries; thus, optimizes and simplifying query writing; can be implemented in phases; consistent query language regardless of data source; returns all data results in XML format; provides different views of data according to different application needs; improves Internet spidering; continues to support all necessary DBMS functions; provides a platform for future growth; and encourages coherent and comprehensive data management and data modeling with organizations.

PARSIMONY: An Infrastructure for Parallel Multidimensional Analysis and Data Mining

Multidimensional analysis and online analytical processing (OLAP) operations require summary information on multidimensional data sets. Most common are aggregate operations along one or more dimensions of numerical data values. Simultaneous calculation of multidimensional aggregates are provided by the Data Cube operator, used to calculate and store summary information on a number of dimensions. This is computed only partially if the number of dimensions is large. Query processing for these applications requires different views of data to gain insight and for effective decision support. Queries may either be answered from a materialized cube in the data cube or calculated on the fly. The multidimensionality of the underlying problem can be represented both in relational and in multidimensional databases, the latter being a better fit when query performance is the criteria for judgment. Relational databases are scalable in size for OLAP and multidimensional analysis and efforts are on to make their performance acceptable. On the other hand multidimensional databases have proven to provide good performance for such queries, although they are not very scalable. In this article we address (1) scalability in multidimensional systems for OLAP and multidimensional analysis and (2) integration of data mining with the OLAP framework. We describe our system PARSIMONY, parallel and scalable infrastructure for multidimensional online analytical processing, used for both OLAP and data mining. Sparsity of data sets is handled by using chunks to store data either as a dense block using multidimensional arrays or as sparse representation using a bit encoded sparse structure. Chunks provide a multidimensional index structure for efficient dimension oriented data accesses much the same as multidimensional arrays do. Operations within chunks and between chunks are a combination of relational and multidimensional operations depending on whether the chunk is sparse or dense. Further, we develop parallel algorithms for data mining on the multidimensional cube structure for attribute-oriented association rules and decision-tree-based classification. These take advantage of the data organization provided by the multidimensional data model. Performance results for high dimensional data sets on a distributed memory parallel machine (IBM SP-2) show good speedup and scalability.

High Performance OLAP and Data Mining on Parallel Computers

On-Line Analytical Processing (OLAP) techniques are increasingly being used in decision support systems to provide analysis of data. Queries posed on such systems are quite complex and require different views of data. Analytical models need to capture the multidimensionality of the underlying data, a task for which multidimensional databases are well suited. Multidimensional OLAP systems store data in multidimensional arrays on which analytical operations are performed. Knowledge discovery and data mining requires complex operations on the underlying data which can be very expensive in terms of computation time. High performance parallel systems can reduce this analysis time. Precomputed aggregate calculations in a Data Cube can provide efficient query processing for OLAP applications. In this article, we present algorithms for construction of data cubes on distributed-memory parallel computers. Data is loaded from a relational database into a multidimensional array. We present two methods, sort-based and hash-based for loading the base cube and compare their performances. Data cubes are used to perform consolidation queries used in roll-up operations using dimension hierarchies. Finally, we show how data cubes are used for data mining using Attribute Focusing techniques. We present results for these on the IBM-SP2 parallel machine. Results show that our algorithms and techniques for OLAP and data mining on parallel systems are scalable to a large number of processors, providing a high performance platform for such applications.

The entity-relationship model—toward a unified view of data

A data model, called the entity-relationship model, is proposed. This model incorporates some of the important semantic information about the real world. A special diagrammatic technique is introduced as a tool for database design. An example of database design and description using the model and the diagrammatic technique is given. Some implications for data integrity, information retrieval, and data manipulation are discussed. The entity-relationship model can be used as a basis for unification of different views of data: the network model, the relational model, and the entity set model. Semantic ambiguities in these models are analyzed. Possible ways to derive their views of data from the entity-relationship model are presented.

The entity-relationship model

A data model, called the entity-relationship model, is proposed. This model incorporates some of the important semantic information in the real world. A special diagramatic technique is introduced as a tool for data base design. An example of data base design and description using the model and the diagramatic technique is given. Some implications on data integrity, information retrieval, and data manipulation are discussed.The entity-relationship model can be used as a basis for unification of different views of data: the network model, the relational model, and the entity set model. Semantic ambiguities in these models are analyzed. Possible ways to derive their views of data from the entity-relationship model are presented.