Common Data Representation Research Articles

Agreeing to Disagree: Translating Representations to Uncover a Unified Representation for Social Robot Actions

Researchers and designers of social robots often approach robot control system design from a single perspective; such as designing autonomous robots, teleoperated robots, or robots programmed by an end-user. While each design approach presents a tradeoff between some advantages and limitations, there is an opportunity to integrate these approaches where people benefit from the best-fit approach for their use case. In this work, we propose integrating these seemingly distinct robot control approaches to uncover a common data representation of social actions defining social expression by a robot. We demonstrate the value of integrating an authoring system, teleoperation interface, and robot planning system by integrating instances of these systems for robot storytelling. By relying on an integrated system, domain experts can define behaviors through end-user interfaces that teleoperators and autonomous robot programmers can use directly thus providing a cohesive expert-driven robot system.

A heuristic algorithm solving the mutual-exclusivity-sorting problem.

Binary (or Boolean) matrices provide a common effective data representation adopted in several domains of computational biology, especially for investigating cancer and other human diseases. For instance, they are used to summarize genetic aberrations-copy number alterations or mutations-observed in cancer patient cohorts, effectively highlighting combinatorial relations among them. One of these is the tendency for two or more genes not to be co-mutated in the same sample or patient, i.e. a mutual-exclusivity trend. Exploiting this principle has allowed identifying new cancer driver protein-interaction networks and has been proposed to design effective combinatorial anti-cancer therapies rationally. Several tools exist to identify and statistically assess mutual-exclusive cancer-driver genomic events. However, these tools need to be equipped with robust/efficient methods to sort rows and columns of a binary matrix to visually highlight possible mutual-exclusivity trends. Here, we formalize the mutual-exclusivity-sorting problem and present MutExMatSorting: an R package implementing a computationally efficient algorithm able to sort rows and columns of a binary matrix to highlight mutual-exclusivity patterns. Particularly, our algorithm minimizes the extent of collective vertical overlap between consecutive non-zero entries across rows while maximizing the number of adjacent non-zero entries in the same row. Here, we demonstrate that existing tools for mutual-exclusivity analysis are suboptimal according to these criteria and are outperformed by MutExMatSorting. https://github.com/AleVin1995/MutExMatSorting. Supplementary data are available at Bioinformatics online.

Fast Self-Guided Multi-View Subspace Clustering.

Multi-view subspace clustering is an important topic in cluster analysis. Its aim is to utilize the complementary information conveyed by multiple views of objects to be clustered. Recently, view-shared anchor learning based multi-view clustering methods have been developed to speed up the learning of common data representation. Although widely applied to large-scale scenarios, most of the existing approaches are still faced with two limitations. First, they do not pay sufficient consideration on the negative impact caused by certain noisy views with unclear clustering structures. Second, many of them only focus on the multi-view consistency, yet are incapable of capturing the cross-view diversity. As a result, the learned complementary features may be inaccurate and adversely affect clustering performance. To solve these two challenging issues, we propose a Fast Self-guided Multi-view Subspace Clustering (FSMSC) algorithm which skillfully integrates the view-shared anchor learning and global-guided-local self-guidance learning into a unified model. Such an integration is inspired by the observation that the view with clean clustering structures will play a more crucial role in grouping the clusters when the features of all views are concatenated. Specifically, we first learn a locally-consistent data representation shared by all views in the local learning module, then we learn a globally-discriminative data representation from multi-view concatenated features in the global learning module. Afterwards, a feature selection matrix constrained by the l2,1 -norm is designed to construct a guidance from global learning to local learning. In this way, the multi-view consistent and diverse information can be simultaneously utilized and the negative impact caused by noisy views can be overcame to some extent. Extensive experiments on different datasets demonstrate the effectiveness of our proposed fast self-guided learning model, and its promising performance compared to both, the state-of-the-art non-deep and deep multi-view clustering algorithms. The code of this paper is available at https://github.com/chenzhe207/FSMSC.

CALIBRATION AND VALIDATION OF A STEREO CAMERA SYSTEM AUGMENTED WITH A LONG-WAVE INFRARED MODULE TO MONITOR ULTRASONIC WELDING OF THERMOPLASTICS

Abstract. Ultrasonic welding of thermoplastics has become an important industrial manufacturing process in the fields of aerospace and transportation. High quality standards are demanded and a reliable quality assessment routine is fundamental. Most on-site and off-site approaches are time consuming or require additional active illumination. Especially temperature models have proven to be good indicators for welding quality. We present a methodology to measure the surface temperature in real-time and visualize it simultaneously as a 3D model. Through augmenting a stereo camera system with an additional passive thermal infrared camera, we are able to map the heat data of multiple successive welds of large, free-form structures into a common 3D data representation. A challenging calibration approach is used to derive the inner and exterior orientation for the trifocal camera system. Geometric and radiometric improvements for an aluminium chessboard allow the usage of wide-angle optics for the thermal infrared camera. Consequently, we verify the quality of each camera by means of their resolving power. Therefore, a Siemens star test pattern is used for the thermal camera as well. We demonstrate the effectiveness of our methodology on a robot-guided ultrasonic welding tool.

A survey on datasets for fairness‐aware machine learning

AbstractAs decision‐making increasingly relies on machine learning (ML) and (big) data, the issue of fairness in data‐driven artificial intelligence systems is receiving increasing attention from both research and industry. A large variety of fairness‐aware ML solutions have been proposed which involve fairness‐related interventions in the data, learning algorithms, and/or model outputs. However, a vital part of proposing new approaches is evaluating them empirically on benchmark datasets that represent realistic and diverse settings. Therefore, in this paper, we overview real‐world datasets used for fairness‐aware ML. We focus on tabular data as the most common data representation for fairness‐aware ML. We start our analysis by identifying relationships between the different attributes, particularly with respect to protected attributes and class attribute, using a Bayesian network. For a deeper understanding of bias in the datasets, we investigate interesting relationships using exploratory analysis.This article is categorized under: Commercial, Legal, and Ethical Issues > Fairness in Data Mining Fundamental Concepts of Data and Knowledge > Data Concepts Technologies > Data Preprocessing

LSLPCT: An Enhanced Local Semantic Learning Transformer for 3-D Point Cloud Analysis

The 3D point cloud is a common 3D data representation that has received increasing attention for remote sensing applications. However, processing 3D point cloud semantics, especially local semantic information, has always been a challenge and has attracted much attention. In this paper, we propose a novel enhanced local semantic learning transformer for 3D point cloud analysis, which aims to enhance the transformer awareness of local semantic features to handle complex point cloud tasks. First, we propose a novel transformer framework, the local semantic learning point cloud transformer (LSLPCT), which not only learns 3D point clouds the global information, but also enhances the perception of local semantic information end-to-end. Second, we design an efficient local semantic learning self-attention mechanism, namely LSL-SA, which can parallelize the perception of global contextual information and the capture of finer-grained local semantic features. Third, our proposed LSL-SA is easy to implement and can integrate existing transformers and CNN-based networks for processing various point cloud tasks. Numerous experiments in different types of point cloud tasks have been conducted, and our method performs better or is competitive with other state-of-the-art methods.

CorpusDB: Software for Analysis, Storage, and Manipulation of Sound Corpora

CorpusDB is a system for representing sound files and associated analysis metadata in a structured format. The formats and conventions used in conjunction with the database allow for representation of sound files and their processed variants; multiple, overlapping, hierarchical relationships between sound files and segments thereof; and connections between sounds, their transformations, and analysis metadata. The software described in this paper is a parallel implementation consisting of SuperCollider classes, Python classes, and a common data representation of corpora that allows for seamless sharing of data between the two complementary environments. Code examples and listings of multi-step algorithms are included that demonstrate the kinds of operations possible within this system.

Towards a Semantic Data Harmonization Federated Infrastructure.

Data integration is an increasing need in medical informatics projects like the EU Precise4Q project, in which multidisciplinary semantically and syntactically heterogeneous data across several institutions needs to be integrated. Besides, data sharing agreements often allow a virtual data integration only, because data cannot leave the source repository. We propose a data harmonization infrastructure in which data is virtually integrated by sharing a semantically rich common data representation that allows their homogeneous querying. This common data model integrates content from well-known biomedical ontologies like SNOMED CT by using the BTL2 upper level ontology, and is imported into a graph database. We successfully integrated three datasets and made some test queries showing the feasibility of the approach.

Igniting Harmonized Digital Clinical Quality Measurement through Terminology, CQL, and FHIR.

Background Electronic clinical quality measures (eCQMs) seek to quantify the adherence of health care to evidence-based standards. This requires a high level of consistency to reduce the effort of data collection and ensure comparisons are valid. Yet, there is considerable variability in local data capture, in the use of data standards and in implemented documentation processes, so organizations struggle to implement quality measures and extract data reliably for comparison across patients, providers, and systems. Objective In this paper, we discuss opportunities for harmonization within and across eCQMs; specifically, at the level of the measure concept, the logical clauses or phrases, the data elements, and the codes and value sets. Methods The authors, experts in measure development, quality assurance, standards and implementation, reviewed measure structure and content to describe the state of the art for measure analysis and harmonization. Our review resulted in the identification of four measure component levels for harmonization. We provide examples for harmonization of each of the four measure components based on experience with current quality measurement programs including the Centers for Medicare and Medicaid Services eCQM programs. Results In general, there are significant issues with lack of harmonization across measure concepts, logical phrases, and data elements. This magnifies implementation problems, confuses users, and requires more elaborate data mapping and maintenance. Conclusion Comparisons using semantically equivalent data are needed to accurately measure performance and reduce workflow interruptions with the aim of reducing evidence-based care gaps. It comes as no surprise that electronic health record designed for purposes other than quality improvement and used within a fragmented care delivery system would benefit greatly from common data representation, measure harmony, and consistency. We suggest that by enabling measure authors and implementers to deliver consistent electronic quality measure content in four key areas; the industry can improve quality measurement.

Representing molecular and materials data for unsupervised machine learning

Statistical analysis and machine learning can help us understand and predict the collective properties and performance of ensembles of molecules and nanostructures, while accounting for all the complexity and diversity of real world specimens. Combining data-driven techniques with robust and reliable simulation methods can provide insights that cannot be made any other way. However, not all statistical and machine learning methods are right for all occasions; testing, validation and perhaps some trial and error are needed. Domain knowledge alone is not sufficient to choose the right algorithms. Data representation methods that are best suited to machine learning are not necessarily scientifically intuitive. The best descriptors are not always the structural features or physiochemical properties that we are aiming to control, and the way our data is distributed can be as important as what it contains. In this review, we discuss the differences, advantages and disadvantages of some of the common data representation, reduction and classification methods applicable to molecular and materials modelling. Focussing on unsupervised methods, we highlight features of these algorithms that determine their suitability and can inform choices of which learning method to use and how to effectively prepare data. A case study is also provided to demonstrate how testing can be undertaken, and how methods can be combined.

Improving patient safety reporting with the common formats: Common data representation for Patient Safety Organizations

Medical errors and patient safety issues remain a significant problem for the healthcare industry in the United States. The Institute of Medicine report To Err is Human reported that there were as many as 98,000 deaths per year due to medical error as of 1999. Many authors and government officials believe that the first step on the path to improvement in patient safety is more comprehensive collection and analysis of patient safety events. The belief is that this will enable safety improvements based on data showing the nature and frequency of events that occur, and the effectiveness of interventions. This systematization of healthcare practice can be a step in the right direction toward a value based, safety conscious and effective healthcare system. To help standardize this reporting and analysis, AHRQ created Common Formats for Patient Safety data collection and reporting. This manuscript describes the development of patient safety reporting and learning through the Patient Safety Organizations (PSO)s and the Common Formats and gives readers an overview of how the system is expected to function and the breadth of development of the Common Formats to date.

USING LATTICE TOPOLOGY INFORMATION TO INVESTIGATE PERSISTENT SCATTERERS AT FACADES IN URBAN AREAS

Abstract. Modern spaceborne SAR sensors like TerraSAR-X offer ground resolution of up to one meter in range and azimuth direction. Buildings, roads, bridges, and other man-made structures appear in such data often as regular patterns of strong and temporally stable points (Persistent Scatterer, PS). As one step in the process of unveiling what object structure actually causes the PS (i.e., physical nature) we compare those regular structures in SAR data to their correspondences in optical imagery. We use lattices as a common data representation for visible facades. By exploiting the topology information given by the lattices we can complete gaps in the structures which is one step towards the understanding of the complex scattering characteristics of distinct facade objects.

Impact of the Use of Object Request Broker Middleware for Inter-Component Communications in C6416 Digital Signal Processor Based Software Communications Architecture Radio Systems

Problem statement: This study presents an in-depth analysis of the performance of Software Communications Architecture (SCA) component-based waveform applications in terms of inter-component communications. The main limitation with SCA, in the context of embedded systems, is the additional cost introduced by the use of Object Request Broker (ORB) middleware. The ORB middleware handles the interaction between components and objects in SCA distributed environment. This interaction should be highly efficient, due to the real time nature of SCA systems and transparent to the application programmer. Approach: We can achieve high efficiency in SCA systems by enhancing the Inter-Process Communications (IPC) mechanisms in Operating systems (OS) micro kernels, while we achieve transparency through Interface Definition Language (IDL). Different encoding mechanisms like “External Data Representation (XDR), Network Data Representation (NDR) and Common Data Representation (CDR) facilitate inter-component communication transparently and efficiently”. Marshalling procedures format data from the local machine representation to common network representations. A most common encoding mechanism for Common Object Request Broker Architecture (CORBA) systems is CDR representation. Measurements have been performed with ORBExpress DSP as a CORBA distribution and Open Source SCA Implementation Embedded (OSSIE) for SCA implementation. In order to perform these measurements we proposed two metrics for profiling the ORB that are invocation and marshalling. In addition, we propose three elements of data types to evaluate the performance of ORB middleware that are, Basic, Array and Sequence data types. Results: The CORBA bus is really the part, which brings an overhead to the SCA radio systems. This overhead is due to method invocations that have been carried out by ORB middleware. Conclusion: Performance benchmarks of ORBExpress DSP middleware show that, although using CORBA for inter-component communications introduces delays and overheads, the overall effect can be reduced by sending packets of data instead of basic data type elements.

Social computing: an intersection of recommender systems, trust/reputation systems, and social networks

Computational applications now go beyond personal computing, facilitating collaboration and social interactions. Social computing is an area of information technology concerned with the intersection of human and social studies connected by computer networks. The primary goal of this article is to provide a brief survey of three popular social computing services: recommender systems, trust/reputation systems, and social networks. We approach these services from a data representation perspective and discuss two of their main challenges: network sparsity and coldstart problems. We also present a novel graph model, which provides an abstract taxonomy and a common data representation model for the three services. We are mainly motivated by the power of graph theory in data representation and analysis for social computing services. Through this model, we believe that it becomes clearer that data from different contexts can be related such that new solutions can be explored; thus, it may provide illumination for the aforementioned problems and stimulate new research.

Exact OBDD Bounds for Some Fundamental Functions

Ordered binary decision diagrams (OBDDs) are nowadays one of the most common dynamic data structures or representation types for Boolean functions. Among the many areas of application are verification, model checking, computer aided design, relational algebra, and symbolic graph algorithms. Although many exponential lower bounds on the OBDD size of Boolean functions are known, there are only few functions where the OBDD size is asymptotically known exactly. In this paper the exact OBDD sizes of the fundamental functions multiplexer and addition of n-bit numbers are determined.

A relational–XML data warehouse for data aggregation with SQL and XQuery

AbstractIntegrating information from multiple data sources is becoming increasingly important for enterprises that partner with other companies for e‐commerce. However, companies have their internal business applications deployed on diverse platforms and no standard solution for integrating information from these sources exists. To support business intelligence query activities, it is useful to build a data warehouse on top of middleware that aggregates the data obtained from various heterogeneous database systems. Online analytical processing (OLAP) can then be used to provide fast access to materialized views from the data warehouse. Since extensible markup language (XML) documents are a common data representation standard on the Internet and relational tables are commonly used for production data, OLAP must handle both relational and XML data. SQL and XQuery can be used to process the materialized relational and XML data cubes created from the aggregated data. This paper shows how to handle the two kinds of data cubes from a relational–XML data warehouse using extract, transformation and loading. Copyright © 2008 John Wiley & Sons, Ltd.

Bridging the language gap in scientific computing: the Chasm approach

AbstractChasm is a toolkit providing seamless language interoperability between Fortran 95 and C++. Language interoperability is important to scientific programmers because scientific applications are predominantly written in Fortran, while software tools are mostly written in C++. Two design features differentiate Chasm from other related tools. First, we avoid the common‐denominator type systems and programming models found in most Interface Definition Language (IDL)‐based interoperability systems. Chasm uses the intermediate representation generated by a compiler front‐end for each supported language as its source of interface information instead of an IDL. Second, bridging code is generated for each pairwise language binding, removing the need for a common intermediate data representation and multiple levels of indirection between the caller and callee. These features make Chasm a simple system that performs well, requires minimal user intervention and, in most instances, bridging code generation can be performed automatically. Chasm is also easily extensible and highly portable. Copyright © 2005 John Wiley & Sons, Ltd.

COMVIS: A Communication Framework for Computer Vision

We describe a general approach to integrate the information produced by different visual modules with the goal of generating a quantitative 3D reconstruction of the observed scene and to estimate the reconstruction errors. The integration is achieved in two steps. Firstly, several different visual modules analyze the scene in terms of a common data representation: planar patches are used by different visual modules to communicate and represent the 3D structure of the scene. We show how it is possible to use this simple data structure to share and integrate information from different visual modalities, and how it can support the necessities of the great majority of different visual modules known in literature. Secondly, we devise a communication scheme able to merge and improve the description of the scene in terms of planar patches. The applications of state-of-the-art algorithms allows to fuse information affected by an unknown grade of correlation and still guarantee conservative error estimates. Tests on real and synthetic scene show that our system produces a consistent and marked improvement over the results of single visual modules, with error reduction up to a factor of ten and with typical reduction of a factor 2–4.

Asymptotically Optimal Bounds for OBDDs and the Solution of Some Basic OBDD Problems

Ordered binary decision diagrams (OBDDs) are nowadays the most common dynamic data structure or representation type for Boolean functions. Among the many areas of application are verification, model checking, and computer aided design. For many functions it is easy to estimate the OBDD size but asymptotically optimal bounds are only known in simple situations. In this paper, methods for proving asymptotically optimal bounds are presented and applied to the solution of some basic problems concerning OBDDs. The largest size increase by a synthesis step of π-OBDDs followed by an optimal reordering is determined as well as the largest ratio of the size of deterministic finite automata, quasi-reduced OBDDs, and zero-suppressed BDDs compared to the size of OBDDs. Moreover, the worst case OBDD size of functions with a given number of 1-inputs is investigated.

Hybrid Rendering of Multidimensional Image Data

The most important rendering methods applied in medical imaging are surface and volume rendering techniques. Each approach has its own advantages and limitations: Fast surface-oriented methods are able to support real-time interaction and manipulation. The underlying representation, however, is dependent on intensive image processing to extract the object surfaces. In contrast, volume visualization is not necessarily based on extensive image processing and interpretation. No data reduction to geometric primitives, such as polygons, is required. Therefore, the process of volume rendering is currently not operating in real time. In order to provide the radiological diagnosis with additional information as well as to enable simulation and preoperative treatment planning we developed a new hybrid rendering method which combines the advantages of surface and volume presentation, and minimizes the limitations of these approaches. We developed a common data representation method for both techniques. A preprocessing module enables the construction of a data volume by interpolation as well as the calculation of object surfaces by semiautomatic image interpretation and surface construction. The hybrid rendering system is based on transparency and texture mapping features. It is embedded in a user-friendly open system which enables the support of new application fields such as virtual reality and stereolithography. The efficiency of our new method is described for 3-D subtraction angiography and the visualization of morpho-functional relationships.