Data clustering: a fundamental method in data science and management

The four Magnetospheric Multiscale (MMS) spacecraft will collect a combined volume of approximately 100 gigabits per day of particle and field data. On average, only 4 gigabits of that volume can be transmitted to the ground. To maximize the scientific value of each transmitted data segment, MMS has developed the Science Operations Center (SOC) to manage science operations, instrument operations, and selection, downlink, distribution, and archiving of MMS science data sets. The SOC is managed by the Laboratory for Atmospheric and Space Physics (LASP) in Boulder, Colorado and serves as the primary point of contact for community participation in the mission. MMS instrument teams conduct their operations through the SOC, and utilize the SOC's Science Data Center (SOC) for data management and distribution. The SOC provides a single mission data archive for the housekeeping and science data, calibration data, ephemerides, attitude and other ancillary data needed to support the scientific use and interpretation. All levels of data products will reside at and be publicly disseminated from the SDC. Documentation and metadata describing data products, algorithms, instrument calibrations, validation, and data quality will be provided. Arguably, the most important innovation developed by the SOC is the MMS burst data management and selection system. With nested automation and 'Scientist-in-the-Loop' (SITL) processes, these systems are designed to maximize the value of the burst data by prioritizing the data segments selected for transmission to the ground. This paper describes the MMS science operations approach, processes and data systems, including the burst system and the SITL concept.

Scientific data management has recently become a critical research issue due to computerization and automation of scientific research procedures and the resulting explosion of electronic data. For proper management of scientific data, adequate data types must be provided for representing the semantics of scientific data directly. Issues such as large data volume and data evolution are common among scientific applications, so traditional database support, such as storage management, are also beneficial to them. Current approaches such as standardized data formats and adoption of traditional databases do not accommodate the data support requirements of existing scientific applications well. Object-oriented databases, on the other hand, seem to hold promise, by combining flexible data models with traditional database support. In this dissertation, we constructed an experimental platform for exploring the design space for a scientific data management system. With this experimental approach, we could dynamically examine possible architectures for data support against actual instances of scientific data and typical operations executed on them. As many of the dynamic features of existing scientific applications, such as data access patterns, are yet to be discovered, the platform also provides an opportunity to explore such features. Based on the observed potential of object-oriented databases for scientific data management, the GemStone object-oriented database management system was chosen for the baseline data management architecture of the platform. Among scientific applications where better data support is desired, a scientific persistent language and data analysis environment called NewS and applications using it were selected as a target for the study. Connecting NewS with GemStone provided a cost-effective experimental platform where we could investigate a variety of scientific applications with single implementation. We incorporated GemStone into the NewS environment in such a manner that we could use existing NewS applications without modifications for experiments. We describe the design and implementation of our platform, GemStone-based NewS, and experiments performed on the platform. At the end, we describe primary contributions of the work, and assess whether or not our approach was a productive initial step toward improved scientific data management.

_ This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper SPE 217665, “Well Construction in the Era of Big Data: It’s Not Data Analytics, It’s Engineering With Data,” by Gregory S. Payette, SPE, ExxonMobil; Jeffrey R. Bailey, SPE, consultant; and Paul E. Pastusek, SPE, ExxonMobil, et al. The paper has not been peer reviewed. _ The authors write that data science captures value in well construction when data-analysis methods such as machine learning are underpinned by first principles derived from physics and engineering and supported by deep domain expertise. The authors provide a summary of important data pertaining to the well-construction process and discuss high-level areas where data science can add value to well construction through analysis of such data. Introduction The authors write that it is important for organizations to grow data science and physics-based modeling solutions together; well-construction data science should be what they term an “engineering with data” endeavor. A large portion of the complete paper is dedicated to discussing ways that organizations can improve their abilities to derive value from data-science efforts. Most discussion focuses on steps that data science teams can take today. However, the provided commentary on data management and governance is forward-looking. The paper concludes by examining several historical data-science case studies for well construction. Each example highlights the need to connect data and some physical or engineering process to deliver value through data science. Improving the Effect of Data Science in Well Construction Data Management and Governance. A step that can be taken to enable better holistic well-construction data management is for data collectors and creators to organize their data using data models that center around what the authors term “data entities” that are common and stable across data collectors. Common well-construction data entities include items such as Wells, Wellbores, Hole Sections, Surveys, Casing Strings, Drillstrings and Bottomhole Assemblies (BHAs), Tubing Strings, and Downhole Completion Equipment. These data entities are capitalized in this paper to denote their roles as data entities. The industry is accustomed to organizing data around entities. Two common types of databases particularly relevant to data-science efforts commonly used by operators include the following: - Well Information Management Databases (WIMD)—These databases are used to document information about the life cycle of wells. - Wellsite Information Transfer Standard Markup Language (WITSML) Databases: WITSML databases are primarily used to store time- and depth-series sensor data collected during drilling and completion activities. For the purposes of their paper, the authors do not include other data entities that can be captured through WITSML data repositories. Even though WIMD and WITSML databases share common data entities, such entities typically are not linked to one another through formal relationships at the original database levels. This creates a barrier for data-science projects that require data from both source systems.

In this chapter we present a case study to demonstrate how the current state-of-the-art Genetic Programming (GP) fairs as a tool for the emerging field of Data Science. Data Science refers to the practice of extracting knowledge from data, often Big Data, to glean insights useful for predicting business, political or societal outcomes. Data Science tools are important to the practice as they allow Data Scientists to be productive and accurate. GP has many features that make it amenable as a tool for Data Science, but GP is not widely considered as a Data Science method as of yet. Thus, we performed a real-world comparison of GP with a popular Data Science method to understand its strengths and weaknesses. GP proved to find equally strong solutions, leveraged the new Big Data infrastructure, and was able to provide several benefits like direct feature importance and solution confidence. GP lacked the ability to quickly build and test models, required much more intensive computing power, and, due to its lack of commercial maturity, created some challenges for productization as well as integration with data management and visualization capabilities. The lessons learned leads to several recommendations that provide a path for future research to focus on key areas to improve GP as a Data Science tool.

Scientific research processes are accompanied by burgeoning amounts of research data that serves as findings of the research and provides a platform for the results to be verifiable. Scientific researchers are under pressure to make their data openly available to other researchers for use. Access to scientific data and its curation has become a dominant theme in scientific policy discussions. Besides the funder requirements, researchers are making their data openly available for a number of reasons that include, raising the profiles of their research outputs, to increase their citations, to validate their research outputs and for their authenticity and integrity as researchers. However, not all researchers possess the skills to curate their own data and make it openly available for use in an open environment that is FAIR (Findable, Accessible, Interoperable and Reusable). The primary objective of this paper is to raise awareness of the strategic role of data scientists in the scientific research process. The study seeks to contribute towards a growing and emergent discourse of data science in HEIs. It also seeks to contribute towards an educational model for data scientists in HEIs in South Africa. In-order to realise these objectives the paper asks the following research questions:How are the emergent roles of data scientists conceptualised in HEIs in South Africa?What is the strategic role of a data scientist in the scientific research process?.This study analysed data science programs offered by universities in South Africa to understand how the role of data scientists is conceptualised and to delineate the scope of this emergent field. A qualitative content analysis survey method is used to analyse courses offered by HEIs in South Africa. Preliminary findings from the research revealed that data scientists are drawn from fields like mathematics, statistics, information science, and computer science. The practical jobs of data scientists as conceptualised by the majority of universities employs theories drawn from the information science, computer science and statistics. The data scientists are expected to work closely with the researchers where the research is done. Most universities are offering data science programs at post-graduate level as professional courses for learners coming in from diverse backgrounds. The study recommends that the next generation of data scientists in a South African context should possess a cocktail of skills that enable them to conduct data analysis and enquiry to enable others to work with digital data. Presentation at the UCT E-Research Conference 2019: DataScience4D: Explore the practice, principles and opportunities of data science for development15 – 18 April 2019 held at the University of Cape Town

'Research through innovation' is the current demand echoing throughout the healthcare industry, healthcare institutions tend to invest heavily in technology. Data Science being the major disruptor across industries is being incepted through establishment of innovation and R&D centers within their respective organizations. Data Science has become a critical component for the healthcare industry, supporting innovative approaches towards advanced clinical practice, clinical research and corporate management, serving to build an intelligent enterprise. Every healthcare institution maintains a good number of technical staffs with IT, Software, data management, BI and analytical capabilities, aiding the institutions to manage report and publish its data in some or the other way, grossly covering most aspects of data science knowingly or unknowingly. Setting up a new entity within the organization by recruitment of staff with Data Science based skill sets would be the first thought to strike the management, which in contrast would end up as disaster when it comes to understanding the organizational culture, processes, infrastructure, platforms, data etc. Hence in order to setup a data science hub, regrouping or realigning some of the existing institutional resources is crucial. With this approach, the Data Science hub would carry out three primary functions. The "Project Management & Data Sourcing", the "Data Management & General Analytics" and "Advanced Analytics". Current resources can be reorganized within the first two functions, further; it would be about establishing an advanced analytics group within the hub which would perform the Machine learning and AI functions.

Most of the domains of science are facing an explosion in the data that they have to collect and process in order to conduct research. This is true for both: scientific domains dealing with experimental data, e.g. biology, sociology, astronomy etc, but also scientific domains dealing with simulation data, e.g. seismology, physics etc. To maximize the potential outcome of scientific data analysis, respective data management applications need to fulfil the following coarse tasks: fast on-demand data processing, and effective storage and consolidation of diverse data collections. These two tasks are in general hard to realize because of: (a) the big data size, (b) the diversity of data formats, (c) their conceptual dependencies, (d) disperse data locations, and (e) intensive and systematic nature of scientific queries. We present the characteristics of big scientific data collections and their necessities in terms of data management. Based on this discussion, we discuss the structure of a framework for the processing and consolidation of heterogeneous scientific data collections. Such a framework aims to mediate between the user and a set of available data management technologies, such as relational DBMSs, key-value stores and column stores, in order to efficiently direct data management operations (insertions, updates) and especially requests (queries) to the appropriate data management application. The framework aims to distribute, dissect, and schedule data management actions, as well as integrate results, in a way that reduces response time. This entails the accommodation of methods for the selective parallelism and serialization depending on partial results and response times. Also, this entails the accommodation of methods for the gradual alteration of data formats and storage, e.g. storage of semi-structured data or raw data in files into relational databases. Furthermore, we discuss the processing of scientific query bulks or workflows with the possibility to retrieve early partial results and calibrate query parameters.

Public science is critical to the economy and to society. However, much of the beneficial impact of scientific research only occurs when scientific knowledge is disseminated broadly and is used by others. This thesis examines the emerging policy, law, and practice of facilitating open access to scientific research data. One particular focus is to examine the open data policies recently introduced by research funders and publishers, and the potential in these for driving the practice of open scientific data into the future. This thesis identifies five major stumbling blocks to sustainable open scientific data. Firstly, the prevailing ‘mindset’ that facilitating open access to data is analogous to facilitating open access to publications and, therefore, research data can easily be shared, with research funders and librarians effectively leading the process. Secondly, the unclear meaning of the term ‘data’, which causes confusion among stakeholders. Thirdly, ‘misunderstood incentives’ for data sharing and the additional inputs required from researchers. Fourthly, ‘data privacy’—an issue that only applies to selected research datasets, and yet appears to dominate the discussion about open research data. Finally, there is ‘copyright law’, which poses challenges at different stages of data release and reuse. In this thesis, I argue that the above problems can be addressed using a staged model for open scientific data. I draw specifically on the practice with open scientific data at CERN (the European Organization for Nuclear Research) and the practice of sharing clinical trial data to argue that open data can be shared at various stages of processing and diversification. This model is supplemented by recommendations proposing changes to existing open data mandates and the introduction of a text and data mining exemption into Australian copyright law.

BepiColombo is an interdisciplinary cornerstone mission to planet Mercury. It is a joint mission between ESA and JAXA consisting of 2 complementary spacecraft, the ESA Mercury Planetary Orbiter (MPO) and the JAXA Mercury Magnetospheric Orbiter (MMO). The two spacecraft will be independently operated and controlled from the MPO Operations Ground Segment (OGS) located at the European Space Operations Centre (ESOC) and the MMO Operations Centre at ISAS/JAXA Sagamihara Space Operations Centre (SSOC). The science operations for MMO will be prepared by the MMO Science Operations Centre at SSOC while the MPO operations by the MPO Science Ground Segment (SGS), located at the European Space Astronomy Centre (ESAC) in Madrid, Spain. The BepiColombo MPO Science Ground Segment (SGS) combines the functionalities of a Science Operations Centre, covering the science operation planning and generation of consolidated payload command requests, and the science data handling and archiving tasks. BepiColombo’s MPO scientific payload comprises 11 instruments that will investigate the Mercury planet interior, surface composition and morphology, the intrinsic magnetic field and the composition of the exosphere and the coupling between all of these fields. The BepiColombo MPO has a large on board SSMM (Solid State Mass Memory) in order to store engineering and science data with a total capacity of 384 Gbit. The telemetry science data packets are stored in the SSMM packet stores based on Process Identifiers (PIDs). The instruments have been specified to generate lowand/or high-priority science data with different PIDs, such that this data can be stored in different packet stores. The BepiColombo MPO has two different radio frequency bands available for data return. The X-band channel is predictable in terms of planned data downlink and will be therefore used to return near real-time data needed on ground to check the spacecraft status/health and highpriority science data. In addition, a Ka-band downlink channel is available, where due to weather uncertainties larger data downlink latencies might arise. Thus, it is foreseen to return low-priority science data using the Kaband via file transfer with automatic retransmission of any lost packets. BepiColombo MPO has several downlink mechanisms in order to get the science data in an efficient way back to ground that combine the large

There are various ideas in everyone's mind. But to convert the ideas into reality, requires initiative and action. Petroleum industry has been using different techniques, science and algorithms over the years to tackle the challenges in order to achieve a business objective. This paper shares the implemented approach to support business challenges by developing a Research and development environment to support the business community. The idea was to find out current business user challenges and support their requirements. IBTIKAR (innovation) lab is the result of this idea which is created to provide required In-House infrastructure (Hardware and Software) support within UAE National Oil Company premises. Research and development are integral part of any company/industry for tackling challenges. Accessing various data types and implementing the concept of Big data has been of a great value. Collaboration between business users, Information technology and Data Management is required to have concrete foundation and to prepare roadmap for the future. The application portfolio in an Oil and Gas upstream varies a lot and now a days, almost all the applications /tool come with a programming interface for integrating Open-Source algorithms. It makes it easier to directly connect the open-source tools to existing applications. The type of data types in upstream varies a lot and since these applications are already developed over the years to handle different kind of data, which further makes data analysis tasks achievable in timely manner. The available algorithms in existing tools and those accessed visa Open-source tools are further used for prediction and classification problems. With the advances in the learning algorithms, the modelling has become easier, and it requires very less coding. The challenges are more towards infrastructure support, as the data types and data size has gradually increased. Machine learning and Deep learning requires High end infrastructure to perform the data science using more CPU's, High End GPU's. In order to adapt to the advancing science and technologies, it makes everyone responsible to learn and implement the data science and Artificial intelligence (AI) concepts in daily tasks and projects to support organization strategy and objectives. In order to get the business community ready for AI, availability of Data science tools and infrastructure is required. This will support and allow the business users of any kind of data to try and test various data types using AI and data science. Following approach was taken from the idea initiation to the execution phase: Discussion with Business users to find out current business challenges and impediments in overcoming those challengesDocument project scopes for the discussed business challenges.Discussion with the respective Infrastructure support team to provide the Hardware and software support along with the required data storage space.Secure budget for developing IBTIKAR labDevelop environment for the research and development of various data science projectsIntegration of Open Source and Free software/s with the existing tools and applicationsAgree and support of the respective roles and responsibilitiesAgree of establishing a collaborative environment between business users, Information technology and data management engineers.To have tools covering the life cycle of the machine learning model (from creation to deployment to monitoring) The business users involved throughout the upstream activities, want to use available open-source algorithms and run various Proof of concepts. In order to get value out of data and to extract insight, it is must that business users are empowered with the required infrastructure, trainings, data science tools and Artificial intelligent algorithms in a collaborative Research and development environment. Configuration and implementation of the IBTIKAR environment has aided the business user community in conducting various Proof of concepts and execute process to handle various business objectives and extract value. IBTIKAR environment has delivered value to the business community by providing required High-end resources, a programming environment, available open-source algorithms and it has improved the level of efficiency, collaboration and people engagement.

A Ciência Aberta é definida como um ecossistema que abriga os diversos movi- mentos em torno do acesso aberto à informação científica. Os dados científicos são a unidade nuclear básica que suporta as descobertas científicas. Nesse sentido, o objetivo deste trabalho é situar os dados científicos de pesquisa no contexto da Ciência Aberta, apresentando as definições de dados, compartilhamento e reuso de dados científicos. O trabalho apresenta um apanhado sobre os desafios quanto à gestão dos dados científicos, à tipologia de dados e às vantagens do comparti- lhamento e reuso. Apresenta os desafios computacionais para a construção de repositórios de dados de pesquisa. Conclui-se que a diversidade, a tipologia e o volume de dados representam os grandes desafios na implantação dos repositó- rios e indicam-se diretórios de pesquisa de repositórios de dados disciplinares e multidisciplinares.

As the country promotes the reform of science and technology project management, the number of science and technology projects has been increasing. At the same time, science and technology management data is growing rapidly with the characteristics of complex correlation and multi-dimensionality. How to effectively store the enormous data and mine the contained value of the data have become a crucial problem for current researches. First, we propose a science and technology management data model to illustrate the internal relation of project, topic, unit, and person, and then define the model by mathematical representation. According to science and technology management data query requirements, we present a data query method of science and technology management based on relational engine, which comprises query parser, query optimizer, and query executor. Furthermore, we design a prototype system to achieve the query requirements of science and technology management data. This research contributes to data query by providing a query method with relational engine, and improves the efficiency of data query in science and technology management.

Research on the Construction of Statistical Data Management Platform under Big Data

International audience

We are pleased to launch the first issue of the New England Journal of Statistics in Data Science (NEJSDS). NEJSDS is the official journal of the New England Statistical Society (NESS) under the leadership of Vice President for Journal and Publication and sponsored by the College of Liberal Arts and Sciences, University of Connecticut. The aims of the journal are to serve as an interface between statistics and other disciplines in data science, to encourage researchers to exchange innovative ideas, and to promote data science methods to the general scientific community. The journal publishes high quality original research, novel applications, and timely review articles in all aspects of data science, including all areas of statistical methodology, methods of machine learning, and artificial intelligence, novel algorithms, computational methods, data management and manipulation, applications of data science methods, among others. We encourage authors to submit collaborative work driven by real life problems posed by researchers, administrators, educators, or other stakeholders, and which require original and innovative solutions from data scientists.