How will the occupation feature changed from present? While no one doubts web will be platform for publication any longer as scholars shift from paper to digital content there is an increasing need to provide means supporting production of units self-referential knowledge and helping their release, sharing with others future re-use. These systems should support such functions as purposeful input, suitable forms for publication and management of publicly available resources while at the same time leaving no room for non-disclosure of interests. It is a fact that academic research can now become both scalable and reproducible. Confidence in results will derive from not only knowledge Code itself but also knowing who has used Codes and what they do with them--a transparency unlike any other before Perceived performance has attracted number of influential scientists to this approach. By providing them with secure static URLs for their data (which can then link into any newer version gallery on Web as a live retrieve file from database) we have already begun dealing. It is needed both amateur collector and experienced professional to verify documents--otherwise false data will be spread all around at once. Furthermore one can never know when going about searching through the network just which WYSIWYG editor ought best match your processing needs exactly or whether XML stands up better to future validation than does plain unmarked text for input purposes later. I in this paper we introduce the notion of research objects, semantically rich aggregations of resources which can have some scientific intent or support some research objective. We give several principles that we expect such objects and their associated services to follow.
Research Objects: Online Platform for Exchanging Accessories
Gaurav Sharma, Mandeep Chaudhary
Greater Noida Institute of Technology, Greater Noida U.P.
INTRODUCTION
The methods used to carry out scientific study are changing. Data is manipulated and produced in e-laboratories using techniques including scientific processes, research protocols, standard operating procedures, and algorithms for analysis or simulation. Scientific models and experimental or observational data are usually "born digital," meaning they have no physical equivalent. Traditional scholarly publications are being challenged by this shift to digital content, which is causing a revolution in scientific publishing. Therefore, the usage of electronic publication methods is growing as a result of changes in dissemination channels. The majority of traditional paper publications are linear and understandable by humans rather than machines. However, a scientific result is not always decomposable just because it is published electronically instead of on paper. Additionally, it does not ensure that methods, results, or outputs are reusable.
This is demonstrated by the fact that there are numerous studies that link sleep habits to productivity at work, each of which has a somewhat different design, and there is dispute about whether the overall message distinguishes between cause and effect. Instead of merely reviewing summary data, it would be ideal to take the study-data, background information, and modelling techniques from each paper and combine them into a larger model. Compared to conducting a main study, which measures things directly, doing this correctly is more intellectually challenging. Many study fields share the requirement for comprehensive "meta-science" rather than merely in-depth "mega-science."
Research is becoming more collaborative across all domains, according to studies. The majority of engineering and scientific from the results of prior studies, both in terms of data to use in the current modelling task and in terms of information to analysing. Therefore, we see the need for a framework that makes it easier to share and reuse digital knowledge. The current debate over the availability of information and evidence in climate change research is a contemporary illustration of the importance of open "data publication".
The UK Information Commissioner's Office determined that a university research unit violated the Freedom of Information Act by declining to provide information about its scientists' statements that man-made emissions were driving global warming. Putting aside the specifics of this case, promoting the ethical release of data along with analysis and conclusions would have assisted in preventing the "data silos" that initially necessitated a Freedom of Information request. Without taking away the intellectual property and academic rewards for performing the modelling, there would have been greater transparency regarding the foundation and accuracy of the climate modelling.
Our study is placed within the framework of offer distributed and collaborative venues for e-Science, making it possible to organize and carry out in silicon and hybrid studies—processes that integrate data with computational operations to produce research results. This goes much beyond the idea of an e-laboratory as a conventional laboratory with online equipment or a Laboratory Information Management System to encompass scholars in any environment using dispersed digital resources as their laboratory.
What should researchers publish if not conventional papers and volumes? While digital data exchange is simple, digital scientific knowledge transfer and exchange in collaborative settings has proven to be a challenging task, requiring tacit and quickly evolving expert knowledge, much of which is lost in conventional publication and information exchange methods. Mechanisms that encourage the creation of independent knowledge units and make it easier for them to be published, shared, and reused are, in our opinion, necessary.
The concept of Research Objects—semantically rich collections of resources that provide the previously described "units of knowledge"—is briefly discussed in this work. A search object (RO) offers a container for a systematic collection of resources that are shared both inside and between organizations and generated and consumed by shared services. The ability to "borrow strength" might be advantageous for domains limits. An RO compiles crucial data about research and experiments. This encompasses not just the data utilized and the techniques used to generate and analyse it, but also the individuals engaged in the inquiry. There is now more to an association with a dataset (or service, or result collection, or instrument) than merely citing or referencing that dataset (or service, or result collection). Instead, the association is a link to that dataset (or service, or collection of results) that may be explicitly followed or dereferenced, giving access to the real resource and, consequently, enacting the service, querying or retrieving data, and so forth. Additionally, a RO has extra semantic data that will arrange the resources rather than merely aggregate them. Please take note that this paper does not offer a comprehensive technological solution. While we acknowledge that many aspects of our methodology need to be described in greater detail, we also recognize that several new Research Objects are developing in e-laboratory solutions. Therefore, this article outlines the tenets and characteristics of our approach and is meant to be a first step as well as a position paper or manifesto.
CONTEXT
The motivation and willingness of scientists to share their experimental work and data, both before and after publication, is still up for debate However, the recent application of social networking techniques to the development of e-laboratories has demonstrated that scientists are increasingly ready to share their experimental data and resources and in turn find and reuse resources that have been shared by other scientists.
A significant finding from several projects creating e-laboratories to exchange and consume scientific resources that, scientific inquiries consist of collections of resources—has served as the inspiration for this study. The my Experiment Virtual Research Environment, Sys MO SEEK, Obesity e-Lab and its technology platform Method Box, the Greater Manchester Collaboration for Leadership in Applied Health Research and Care: Systems5, and the National e-Infrastructure for Social Simulation (NEISS) project6 are some of these initiatives. We use the use of "Packs" in one of those projects, my-Experiment, as an inspiring example. Scientists can exchange digital materials related to their research using the my Experiment Virtual Research Environment.
Using Web 2.0 concepts, it offers a social website where scientists may find, publish, and curate scientific procedures originally concentrating on workflows7, the project currently incorporates a number of workflow systems, such as Taverna8. Workflows can be enhanced by packaging them with additional information (such as input data, results, or workflow execution logs).
Figure 1: Resources in my Experiment pack and RO
The concept of "Packs," which are groups of goods that can be shared as a single entity, is used in my Experiment to facilitate this. Figure 1's left, which displays several resources combined into a single my Experiment pack, serves as an illustration of the position.
The pack is now a single item that can be shared or annotated, and it permits simple resource aggregation. Metadata that explains the relationships between the resources within the aggregation is required in addition to the basic aggregation structure in order to support more complex forms of reuse (for example, to rerun an investigation with new data or validate that the results being presented are indeed the results expected). The aggregate has been improved by adding metadata that captures the linkages between the resources, such as the fact that a specific data item was generated by the execution of a specific workflow, as seen on the right in Figure.
By defining Research Objects, we want to achieve this enrichment of the aggregation and the corresponding added value in terms of reuse and sharing. It should be noted that in this specific example, relationships between the constituent elements are expressed as resources rather than their physical representations.
The representation of scientific discourse and the use of Semantic Web tools to express discourse structures are currently of great interest (e.g. see [4]). Vocabularies that enable the description of studies and the resources employed within them are provided by ontologies like EXPO11, OBI, MGED12, and SWAN/SIOC13. A semantic authoring framework called Semantically Annotated LATEX (SALT) aims to provide semantic annotations to scientific articles. Ontologies for externalizing the reasoning and rhetorical patterns contained in a publication's content are defined by SALT.
The community of HyPER15 are concentrated on the infrastructure needed to support relationships, evidence, and hypotheses. Generally speaking, however, this study tends to concentrate on the specifics of the connections between the resources being discussed—what we can refer to as content rather than container. A fundamental language for the description of aggregations is provided in part by OAI's Object Reuse and Exchange (OAI-ORE), which is utilized in my Experiment's RDF export of packs.
In order to capture the additional value of the collection and facilitate reuse through the exchange of a single object—the Research Object—principles and mechanisms that describe the aggregation of resources and adequately describe their contributions to the investigation and their relationships to one another are lacking .In reality, scientists will deal with a variety of content kinds and data that are dispersed across several sites over the course of a study, which includes planning, carrying out tests, analysing data, and disseminating and publishing the results. Individually, these resources may be helpful, but when taken as a whole, they complement and enhance one another and form a scientific investigation. Depending on the field, subject, and specific research being done, these resources may differ significantly.
. Nonetheless, we can see how specific sources make up recognizable components of an inquiry, and they are some of the components that will comprise our Research Objects. Content might consist of Questions: The background of the question, which may include a description of the issue, a summary of other studies, and a hypothesis; Organizational context: Details on the experiment's investigators, government regulations, and ethical approval; Study Design: web services, scripts, and scientific procedures; Data: sources and sets of raw or processed data in a variety of formats, from worksheet and databases to flat files from signal transducers; Findings: SBRML and worksheet Techniques Publications, papers, reports, slide shows, DOIs, and PUBMED IDs are the answers.
The objective of Research Objects is to provide a class of remnant that may represent our digital information and offer a way to share and find resources of scientific and research knowledge. We provide many guidelines that we anticipate Research Objects will adhere to below. These mixed-natured principles encompass both the behaviour that we anticipate Research Objects to display and the functionality that we anticipate Research Objects (and their related services) to support. Both the characteristics of the research object model and the services that will create, use, and work with research objects are influenced by these principles.
Reuse can take many different shapes. Items can be reused exactly as they are, broken down, and then reassembled in somewhat different ways. If they contain processes, it is possible to recreate them or look at earlier instances of the process. In order to clarify the differences between these types of general reuse and pinpoint the specific demands they place on any suggested e-Laboratory infrastructure, we present a number of principal below.
As a result, our research objects must be able to be used again in a different study or research object. In this instance, we refer to the reuse of the Research Object as a whole or as a single unit.
Reusable Reuse of a Research Object can also refer to the reuse of its component elements, such as taking a research and replacing its data or services with different ones. We can locate reusable components and parts combinations by "opening the lid." How these components can be reused is determined by the descriptions of the connections between them and how they are put together.
Repeatable research object should contain enough data so that the original researcher or others can replicate the study, possibly years later. This could involve data access or service execution, creating a need for infrastructure or enactment services. To access any necessary data or services, the user will also need enough privileges.
Reproducible A third party may begin using the same materials and procedures to replicate (or replicate) a result and determine whether an earlier result may be verified. Before repurposing a study, it may also be necessary for the original investigator to duplicate it. This can be viewed as a particular instance of repeatability in which a final or intermediate outcome can be confirmed due to a complete set of information. In order to determine whether we have actually achieved the same results, we establish the demand for some kind of comparability framework during the process of repetition, particularly when reproducing a study.
Repeatable Studies that are automated may involve lengthy procedures that take months to complete or single examinations that occur in moments. In any case, human comprehension of what transpired depends on the ability to repeat the study and analysing certain aspects of it. Thus, replay enables the investigator to "go back and see what happened." Keep in mind that replay not always entail carrying out or enacting procedures or services. Therefore, replay does not always require enactment services, but it does set requirements on metadata that records the provenance of data and results.
Traceability The scientific method relies heavily on provenance and the ability to audit experiments and investigations. To be persuaded of the validity of the results, other parties must be able to audit the procedures followed during an experiment. A research object should give enough information to permit audit of the aggregation as a whole, its component pieces, and any process that it may contain because audit is necessary not just for regulatory considerations but also for the interpretation and reuse of experiment results.
The characteristics or limitations on how we observe Research Objects being utilized or acting are described by the aforementioned principles. We list many features that can help deliver this functionality below.
.
Aggregation Research objects are collections of concepts. Therefore, a mechanism for this aggregate must be provided by a Research Object framework. References to resources are likely to be included in aggregations, but there may also be circumstances in which Research Objects should be able to aggregate both literal data and references to data for efficiency or persistence.
object instance or record by an identification that is guaranteed to be distinct across the system in which it is utilized. Mechanisms must permit reference to both the aggregation's component parts and the object as a whole. Identity necessitates an explanation of equality or equivalency. When should items be regarded as interchangeable or equivalent? For instance, two objects might not be regarded as equivalent in a particular situation, but they might be interchangeable (e.g., either might be employed with the same effects).
Metadata The foundation of our e-laboratory and Research Object framework is the availability of machine-readable and process metadata. In addition to being labelled as separate objects, research items will also have internal structures and relationships described via metadata. From generic "Dublin Core" style annotations to license, attribution, credit, or copyright information to detailed descriptions of provenance or the derivation of results, metadata can describe many aspects of the RO. The RO is transformed from a basic aggregate (like a zip file) to a reusable object by the presence of metadata.
.
Lifecycle There is a time component to the procedures and research that we want to document in the e-laboratory. There are lifecycles that characterize the several stages that an inquiry goes through, and events occur in a specific order.
Information presented in a Research Object must be attributed and proved. By including such information, credit and rewards based on the publication of data and methods—rather than papers, as is currently mostly the case—will be supported. The issue of credit and attribution also emphasizes how important it is that research objects generally incorporate (links or references to) persons in addition to data and methodologies.
Degradation of Understanding with Grace Lastly, we present the idea of "graceful degradation of understanding," which allows Research Object services to consume Research Objects without necessarily comprehending or processing all of their internal structure or content. We believe this principle is crucial for ensuring interoperability between services and will support reuse of Research Objects, especially serendipitous or unexpected reuse. As a result, the research object model must be conceptually extensible. Furthermore, it is important to keep responsibilities and concerns distinct so that services or applications might decide to disregard irrelevant components. Additionally, this highlights the necessity of clearly defining and distinguishing between Research Objects and Research Object Services.
STEREOTYPES
Several "stereo-typical Research Objects" common patterns of resource aggregation have be identified from an analysis of our lab related projects .
For instance, a study may go through several phases, such as ethical approval, data collection, data cleaning, data analysis, peer review, and publication (see Figure 2). Different actions on the object may be possible at each stage of the procedure. Therefore, a logical explanation of the lifecycle of a research object is required. Such a lifecycle must account for the arrangement of ROs after they are published if ROs are to be utilized to support publication. Versioning The issue of versioning arises in parallel with the lifespan dilemma. Research objects are dynamic in that their contents are subject to change. For example, new content may be added to aggregations, or more metadata about the contents or relationships between content may be added. The aggregated resources are subject to change. Versioning is therefore necessary in order to record changes to objects, possibly in conjunction with tools for obtaining objects or aggregated elements at specific historical stages in their lifespan. CRUD actions are necessary for the creation and management of research objects. Other factors to think about are storage and indexing for discovery.
Research on Security It is believed that objects serve as a means of facilitating the exchange of methodologies, data, and experimentation. Issues of access, authentication, accounting, and trust that we can broadly categorize as pertinent to security arise with sharing.
Attribution A precise method for determining the at Location involves more than just providing works in digital formats (such as online PDFs of papers), though electronic publication is undoubtedly a part of the process. Instead, this is about offering aggregations that specifically combine the presentation of a piece of work—the "paper"—with the supporting evidence for the conclusions being provided, such as data sets, experimental results, the processes that produced those results, and so on.
Publication objects should be unchangeable since they are meant to serve as a record of activity. Although variants of a Publication Object can be created, they should be regarded as separate objects. This has to do with the idea of lifetime, and publication events must be precisely described. Publication objects have to be citable. Mechanisms for identifying and referencing research objects are necessary, as previously described, but they must be in an externally usable format for publication.
Since they are essential for offering incentives and rewards for scientific publication, credit and attribution are crucial components of the publication process. Additionally, the Publication Object will employ ontologies to express the reasoning or rhetorical structure of the publication.Work Item When the application goes beyond research, such as in business intelligence or artificial intelligence, where repeatability and repurposing are elements , we have used the terms "work object" and "research object" interchangeably.
Live Object work in progress is represented by a live object. Because the content or condition of their sources may change, they are therefore malleable, necessitating version management. Mixed stewardship may apply to live objects that are potentially controlled by many owners. Thus, there are problems with access control and security.
View/Context Object A view over some data that has previously been made public can be provided via a View or Context object. Here, Research Objects can interact with published or exposed data using Linked Data principles, giving those resources a "Named Graph."
Method Object Methodological research is reported in a Research Object by a Method Object, which also makes the method easily accessible to other Research/Work Objects. This could be a crucial component in promoting methodological integrity and preventing technique translation problems.
Archived Object An aggregate that is "finished," deprecated, or no longer "live" is contained in an Archived Object. As a result, archived objects ought to be unchangeable. For instance, materials utilized in an experiment that has been abandoned can be gathered and documented using an Archived Object. Similar to publication objects, archived objects might not need as much information about things like credit and attribution.
CONCLUSIONS
Reusable, shared research will not be sufficiently supported by traditional paper publication or even electronic publication that adheres to the "paper metaphor." In order to share, exchange, and reuse digital knowledge as decomposable entities, new procedures are required. Research objects, which are semantically rich collections of resources that compile the information, techniques, and personnel involved in (scientific) inquiries, are our answer to this problem.
Scientists will be able to organize and link the resources they use in their work thanks to Research Objects. More transparency will follow, enabling the results to be validated. Increased resource sharing and the reuse of already-existing data sets and techniques should ideally result from this.
We have taken two distinct approaches to the subject. A number are already starting to organize and publish their data using a Research Objects methodology. Simultaneously, we have been considering the potential role of such aggregations in the scientific method, which brings us to the concepts mentioned in Section 3. As promoted by the Semantic Web community, our Research Object view offers an aggregation structure layer that complements the Linked Data worldview [3]. In addition to being resources that can be accessed through linked data principles, ROs will gather linked data sources.
There are undoubtedly unanswered problems and concerns, but we have provided a succinct summary of our purpose and stance below. Rewards, credit, and attribution My-Experiment's credit and attribution scheme is a crucial component. It permits credit to be given for derivative works and was added at the request of domain scientists. Citation counts are no longer sufficient if works are also based on the reuse or repurposing of data and methods, hence a move to RO-based publication would necessitate a comparable re-engineering of reward structures for scientists.
Trustworthiness and Quality All new collaborative platforms that support open science and the quick sharing of experimental and pre-publication data and techniques face the difficulty of trust. How can consumers trust user-generated content, and how can content creators trust users to appropriately consume, understand, and attribute? Research objects are identifiable containers that enable us to calculate and attribute a measure of trust to the item itself. This allows us to apply and expand methodologies for measuring and modelling social trust, confidence in content, and trust based on provenance information.
In conclusion, we think that the Research Objects approach will enable us to carry out scientific research in ways that are effective, supporting larger-scale research by reusing parts of models; efficient, since borrowing a model is less expensive than recreating it; and ethical, since research funded by public funds will benefit not only individual scientists but the entire community.
REFERENCES
e-Labs and Work Objects: Towards Digital Health Economies, J. Ainsworth and I. Buchan. in the infrastructure of communication.
Lecture Notes of the Institute for Computer Sciences, Social System and Application in Europe, pages 206-216.Springer 2009.
N. Boss, G. Olsen, J. Olson, J. Yew, J. Yerkie , E. Dahl, and A. Zimmerman . A taxonomy of collaborations, from communities of practice to shared database .Computer-Mediated Communication Journal, 12(2), 2007.
T Berners-Lee and T. H. Christian Bizer. Link Data: The Current Story. Special Issue on Linked Data, International Journal on Semantic Web and Information Systems (in press), 2009.
Editors: T. Clark, J. S. Luciano, M. S. Marshall, E. Prud’hommeaux, and S. Stephens. Volume 523 of Semantic Web Applications in Scientific Discourse, 2009. CUER Workshop Proceedings, October 2009.
M. Courtot, W. Bug, F. Gibson, A. L. Lister, J. Malone, D. Schober, R. Brinkman, and Ruttenberg, The Biomedical Investigations OWL. OWLED 2008,
C. Goble and D. De Roure. Research objects for data-intensive research are the lessons learned from my experiment. in the 2009 Microsoft e-Science workshop.
R. Stevens, C. Goble , and D. De Roure . The creation and implementation of the my Experiment virtual research environment for workflow social sharing. Computer Systems of the Future, 25:561–567, 2009.
J.Vlissides, R. Johnson, E. Gamma, and R. Helm. Design Patterns: Components of Reusablesoftware that is object-oriented. Series for Professional Computing. 1995 Addison-Wesley.
C. Lagoze, P. Johnston, M. Nelson, R. Sanderson, H. V. de Sompel, and S. Warner. Abstract Data Model: ORE Specification. Open Archives Initiative technical report, 2008.
Nelson, B. Empty archives for data sharing. Nature, 461(7261):160–3, 2009.
My Experiment: ontology for e-research, D.Newman,
s.bechhofer, and D.DE roure .workshop on semantic web application in scientific Discourse, ISWC 2009.
G.Olson , N.BOS, and A.Zimmerman .internet-based scientific collaboration. MIT Press, 2008.