Spatial Data Supply Chain Provenance Modelling using Semantic Web Technologies

Spatial Data Supply Chain Provenance Modelling using Semantic Web Technologies

Dedication

I would like to dedicate this thesis to my loving parents …

Declaration

I hereby declare that except where specific reference is made to the work of others, the contents of this dissertation are original and have not been submitted in whole or in part for consideration for any other degree or qualification in this, or any other University. This dissertation is the result of my work and includes nothing which is the outcome of work done in collaboration, except where specifically indicated in the text. This dissertation contains less than words including appendices, bibliography, footnotes, tables and equations and has less than figures.

[Author]

2019

Acknowledgements

And I would like to acknowledge …

List of Figures and Abbreviation

Contents

• Introduction
  • Introduction……………………………………………………………………………………………………
  • Theory…………………………………………………………………………………………………………….
    • Systems of Geospatial Data Sharing……………………………………………………….
    • Semantic web technologies……………………………………………………………………
    • Provenance for spatial analysis……………………………………………………………….
  • Objective…………………………………………………………………………………………………………

• Development of ontologies
  • What is in an ontology……………………………………………………………………………………..
    • Why develop an ontology……………………………………………………………………….
    • Defining classes and a class hierarchy……………………………………………………
    • Determine the domain and scope of the ontology…………………………………..
    • Define the classes and the class hierarchy……………………………………………..
    • Define the properties of classes—slots……………………………………………………
    • Define the facets of the slots………………………………………………………………….
    • Create instances……………………………………………………………………………………
    • System Design………………………………………………………………….

• Data
  • Spatial data source………………………………………………………………………………………….

• Some Experimental Results
  • Publish GIS data……………………………………………………………………………………………..
  • RDF………………………………………………………………………………………………………………..
  • RDF Query………………………………………………………………………………………………………
• Discussion
  • Results Discussion…………………………………………………………………………………………..
  • Suggestions…………………………………………………………………………………………………….
• Conclusions
  • Objective achievement…………………………………………………………………………………….
  • Limitation and suggestion……………………………………………………………………………….
  • Further improvement……………………………………………………………………………………..

References

Abstract:

Geographic Information Systems (GIS) assume a significant job to obtain and convey geospatial learning dependent on spatial information and the utilization of spatial examination, demonstrating, and perception. The affirmation of the legitimacy and nature of spatial information taking care of what’s more, examination remains an extraordinary test, to some degree, due to complex systems are frequently required for collective geospatial critical thinking and basic leadership. These methods, when indicated as information induction work processes, require painstakingly designed parameters and spatiotemporal determinations guided by explicit settings and purposes. The data of spatial information ancestry and related investigation work process is characterized as spatial provenance in this examination.

Provenance, a metadata fragment insinuating the source and the systems grasped to get a particular geo-practical propelled component or thing, is urgent to survey the idea of spatial information and help in imitating and reproducing geospatial structures. Regardless, the heterogeneity and capriciousness of the geospatial forms, which can modify part of the complete substance of datasets, clarify the requirement for depicting geospatial provenance at the dataset, feature and trademark levels. This paper shows the use of W3C provenance, which is a nonexclusive detail to express provenance records, for addressing geospatial data provenance at these different levels. erence in the usage of spatial provenance in GIS applications. As a rule, the building and execution portrayed in the paper show the need, what’s more, feasibility of bringing provenance into GIS.

Chapter 1

Introduction:

Geospatial data has gotten expanding consideration from the standard IT world and become fundamental for different certifiable uses. For example urban arranging, traffic examination and emergency reaction. In the geospatial network, the exchange, sharing and representation of geospatial information chiefly depend on various syntactic benchmarks which shape the present answers for spatial information foundation (SDI). Such norms are chiefly from Open Geospatial Consortium (OGC), and the vast majority of them just certification on a syntactic level, while the semantics and information are said to be inadequate. In this way, we need a path for tending to the semantic difficulties concerning geospatial information and learning [1].

Numerous heritage Geographic Information Systems (GIS) have been created over various periods, for various purposes, with various structures based on various GIS programming. The heritage GIS based on various GIS programming has its exclusive designs of system, information models and database structures of storage. Therefore, databases of geographic-based on these frameworks can’t convey without data transformation. In any case, the transformation of data is costly and tedious and may prompt the similarity issues for some time-critical approach, which need continuous access to diverse data on speedy choices and take instantaneous activities [2]. Although the advancement of the World-Wide-Web (WWW) and numerous Internet GIS provide owner approaches to enable clients to rapidly access, show and inquiry spatial information over the web. This Internet GIS additionally has the confinements of exclusive programming plans, information models and storage of database structures.

Issues recognized by Hakimpour and Timpf [3] with data reconciliation between various frameworks. They portray a few issues identifying with semantic heterogeneity and create answers to conquering these issues utilizing ontologies to make and institutionalize road join. They additionally talk about inter-operability issues between various spatial data-set structures and models and the need to determine semantic heterogeneity for example for similar highlights in various data-sets gathered by various organizations having various definitions. For instance, an element class for Main Street may have various definitions as indicated by their motivation and application in separate offices. A further issue is that information pattern and trait structures and definitions may fluctuate between offices. One arrangement is to formalize the semantics characterized at the area level and get understanding from all gatherings that partake in its utilization. What’s more, the age of an institutionalized metaphysics made for a particular space is likewise conceivable. Here a formalization of ideas can be actualized at a more extensive level with the goal that definitions and understandings of data-sets can be institutionalized.

The significance of investigation into information safeguarding and the need to create provenance information stores for inquiry and reuse of information in a manner that is viable, auspicious and with an abnormal state of client certainty and trust [4]. The provenance of datasets is the wellspring of truth about elements, exercises and individuals, who gather, produce and add to the datasets. If the historical backdrop of datasets has been set up, the ancestry can be followed, possession distinguished and in particular, practices and procedure can be broke down and reused for further experimentation. He features likenesses, clashes and issues with current provenance models and exercises. A study was done on provenance and a scientific classification delivered that portrays the constraints due to there being no actualized provenance measures, no client introduction to provenance data, no devoted stockpiling of provenance data, and no strategies to display provenance data to a client in a reasonable structure. Suriarachchi [4] also underlines meaningfulness issues just as the requirement for improved comprehension of provenance data. He overviewed seven provenance data frameworks, two of which have provenance perception instruments of some structure, four have no representation at all and one creates XML records.

During a geospatial web administrations condition, information is prepared and shared all the time, and regularly by various strategies [5]. This implies it is essential to have a component for distinguishing unique information sources. Geospatial information provenance records the deduction history of a geospatial information item. This is significant for assessing the nature of information items, following work processes, refreshing or replicating logical outcomes, and assessing the unwavering quality and the nature of geospatial information items. As a result, geospatial information provenance is perceived as one of the missing components in present-day Spatial Data Infrastructures (SDIs). Provenance data can expand a client’s comprehension of whether the information is fit for a reason and this thus may build a client’s trust level of the information. Understanding information quality evaluation techniques have been tended to by Liuet al. [6]. They clarify how provenance data is basic in extricating quality parameters and data. Quality measurements can be built dependent on property data and work process models. The more trait data gathered the more datasets can be investigated from a quality point of view. Furthermore, catching work processes distinguishes the total life cycle of an information item and this thusly can be utilized to computerize the quality control evaluation process.

The Cooperative Research Center for Spatial Information (CRCSI) Program 3, Spatial Infrastructures, tries to improve the association, access and utilization of spatial information in Australia and New Zealand [7]. The examination program has grasped progressed Semantic Web Technologies and Artificial Intelligence as methods for improving spatial information supply chains [7].

Systems of Geospatial Data Sharing with Semantic web technologies

A structure of geospatial information frameworks to Geospatial Semantic Web Technologies for varying inheritance GIS is proposed as appeared in the following figure (Fig. 1). For moment remote information access and trade, the cosmology based web administrations are utilized to get to and control geospatial information through the web from heterogeneous databases. This methodology guarantees fundamental conditions for bury operability by utilizing a standard trade component and Geospatial Semantic Web Technologies between different spatial information sources associated over the web.

Fig. 1: Systems of Geospatial Data Sharing with Semantic web technologies

Spatial Analysis

The Spatial Analysis, the kind of topographical examination which tries to clarify examples of human conduct and its spatial articulation as far as science and geometry, that is the examination of location. Models incorporate closest neighbour investigation and Thiessen polygons. A considerable lot of the models are grounded in miniaturized scale financial aspects and anticipate the spatial examples which ought to happen, in, for instance, the development of systems and urban frameworks, given various preconditions, for example, the isotropic plain, development minimization, and benefit augmentation. It depends on the principle that monetary man is in charge of the improvement of the scene, and is along these lines subject to the standard reactions of that idea, for example, the absence of through and through freedom.

A differentiation is made in this course among GIS and spatial investigation. With regards to standard GIS programming, the term investigation alludes to information control and information questioning. With regards to spatial examination, the investigation centres around the factual investigation of examples and hidden procedures or all the more, for the most part, spatial examination tends to the inquiry “what could have been the beginning of the watched spatial example?” It’s an exploratory procedure whereby we endeavour to evaluate the watched example at that point investigate the procedures that may have created the example.

Spatial Data Supply Chains (SDSC)

The cutting edge spatial frameworks must address numerous contemporaneous issues inside the spatial data supply chains (SDSC). An SDSCs comprises of various worth include forms along the chain. At each worth include point in the chain, there might be heterogeneous forms, techniques, models and work processes consolidating to produce, adjust and expend spatial information. The worth include procedures happening in coordinating and preparing numerous informational collections brings up issues about information trust, quality, its qualification for a reason, money and legitimate level. An explanation behind this is these informational collections begun from various sources having had diverse forms executed upon them to touch base at this last item. Knowing how information is gathered and what level of precision was utilized gives understanding concerning what reason the information can be utilized for. The production of a geospatial provenance model that catches these sorts of procedures will empower a capacity to quantify how fit for reason information may be.

A huge amount of the Australian spatial information is gained at the neighbourhood government level; it is then consolidated to frame the State or Territory level data sets and afterwards used to make national-level data sets. Numerous procedures utilized in the spatial information age are manual and undocumented just as certainly requiring human mediation. There is an absence of or no connecting instruments at all between data sets. Numerous variants of informational collections are additionally frequently being utilized which may prompt a mistaken or outdated data set being utilized. There are conditions between the various information at various levels including contrasting arrangements and human mediation. These variables muddle data set mix at various levels.

In the country like Australia numerous associations at the nearby government level, inside state government divisions in various wards just as Commonwealth offices, secure spatial information for explicit territories or focal points autonomous of one another. This prompts information duplication at numerous focuses along the SDSC. Absence of mindfulness or just because no single data set suits different offices’ needs, prompts this duplication.

Spatial data infrastructures (SDI)

The term spatial information foundation was instituted in 1993 by the U.S. National Research Council to mean a structure of advancements, strategies, and institutional courses of action that together encourage the creation, trade, and utilization of geospatial information and related data assets over a data-sharing network. Such a structure can be executed barely to empower the sharing of geospatial data inside an association or all the more extensively for use at a national, local, or worldwide level. In all cases, an SDI will give an institutionally authorized, robotized implies for posting, finding, assessing, and trading geospatial data by taking interest data makers and clients. SDI expands a GIS by guaranteeing geospatial information and guidelines are utilized to make definitive data sets and policies that help it.

The development of SDI (spatial data infrastructures) is intently connected with the endeavours of gathering and creating geo-spatial information, just as the headway of studying and PC innovations. In the previous decades, a lot of geo-spatial information, for example, remote detecting pictures and GPS areas have been gathered by government offices, for example, the U.S. Land Survey (USGS) and the National Oceanic and Atmospheric Administration (NOAA). In the interim, the quick advancement of geographic data frameworks encourages the inference of different information items from the gathered information, for example, topographic maps, land spread information, transportation systems, and hydro-graphic highlights. As location‐based administrations are ending up progressively well known, immense measures of volunteered geographic data (VGI) (Goodchild 2007) has additionally been contributed by the overall population through brilliant cell phones and web-based social networking stages. Furthermore, the component of GIS brings geo-spatial administrations that give information preparing and spatial investigation works in the general Web condition. The huge number of geo-spatial information, administrations, maps, and others, be that as it may, don’t facilitate the utilization of these geospatial assets. On one hand, it is trying to discover and get to these advanced assets which are generally conveyed at various government offices and sites (Li, Wang and Bhatia 2016). Then again, a lot of information redundancies exist, and cash and HR were squandered in copied information accumulation and upkeep endeavours (Rajabifard and Williamson 2001, Maguire and Longley 2005).

An SDI comprises of numerous parts. Notwithstanding the advanced geo-spatial assets, an SDI likewise needs equipment, programming, individuals, associations, guidelines, strategies, and numerous others to work appropriately. Developing an SDI additionally needs compelling correspondences among networks, and exchanges among associations and even nations to arrive at understandings. While an SDI has numerous segments, this section will especially concentrate on geo-portals, metadata, and search capacities, which are three key parts of a common SDI.

Geo portals are ordinarily created utilizing Web‐based advancements and off‐the‐shelf GIS programming bundles. A database the board framework (DBMS) is utilized to store and deal with the metadata of the geospatial assets contained in the SDI. A Web interface, which frequently contains a guide, empowers end clients to collaborate with the framework and to lead look (Figure 2). At the point when a hunt is played out, an HTTP (Hypertext Transmission Protocol) solicitation will be sent to the Web server which has the geo-portal. In the wake of questioning the metadata put away in the database, the geo-portal will at that point send back the outcome to the customer through an HTTP reaction. Geo-portals are regularly intended to be utilized by the two GIS experts and the overall population.

Metadata give documentation on the substance and the generation procedure of geospatial assets. Metadata are regularly called the information about information, and incorporate data, for example, titles, portrayals, information classifications, the areas and time of the information gathering, the information authorities, and the utilized arrange frameworks and guide projections, and the information cleaning and preparing methods. Metadata can likewise be utilized for portraying geo-spatial benefits by giving data about the information and capacities offered by the administrations, the information and yield, the engineers, the advancement time, and others. So, metadata are pretty much all parts of advanced geo-spatial assets.

The search capacities in SDIs are being improved by analysts. One of these enhancements lies in text‐based search: there is a change from a keyword‐based search to the semantic inquiry. Keyword‐based search inspects the coordinating between the catchphrases contribution by the clients and the literary depictions of the geospatial assets. In this way, if a client types in the street, the inquiry capacity won’t have the option to discover the assets named as a road. Semantic inquiry intends to coordinate advanced geospatial assets to client questions dependent on the significance (semantics) of the inquiries, and accordingly can recognize applicable information and administrations even though they are not marked with the precisely same words.

The Australian Spatial Data Infrastructure (ASDI)

The Australian Spatial Data Infrastructure (ASDI) is a national structure for connecting clients with suppliers of spatial data and is comparative in idea to a national interstate or railroad arrange. The ASDI contains the individuals, arrangements and advances important to empower the utilization of spatially referenced information through all degrees of government, the private area, non-benefit associations and the scholarly community.

A few components of the ASDI are now created, including approaches, rules, the Australian Spatial Data Directory (ASDD), national information models, for example, ICSM’s Harmonized Data Model, metadata records and institutional game plans. Since its commencement in the 1990’s the spatial condition has changed fundamentally. On-line frameworks carry a type of the SDI to everybody, independent gadgets that know, think and convey (sensor systems) are getting to be accessible and dynamic foundations are being built up. Web administrations are improving the capacity of uses to get to the SDI. The idea of Virtual Australia has risen and been embraced as a key activity by the CRC-SI together with undertakings to help the improvement of the Australia New Zealand Spatial Market Place (ANZSM).

The Australian Spatial Data Directory (ASDD) has now been supplanted by FIND the Australian Government’s spatial information inventory and related to data.gov.au gives access to a system of open government information. FIND enables you to look and download a wide scope of spatially-referenced datasets made by both the Australian and State and Territory Governments.

To guarantee interoperability, the ASDI should depend on a lot of normal principles and conventions, the utilization of which ought to be commanded and consistence checked and implemented (RF-13). The present recognition that the utilization of models on little undertakings is a superfluous cost should be adjusted with the goal that the long haul advantages emerging from the informational index and business procedure reuse are acquired.

Customarily the ASDI has been characterized with a lot of crucial or system informational indexes, to be specific: Geodetic, Hydrographic, Topography, Cadastral, Names and Addressing (RF-23). While these sorts of informational indexes will stay significant as the ASDI advances, extra information (at different scales and goals) should have the option to be incorporated into the ASDI. The information ought to have the option to be included and evacuated all through the ASDI’s life expectancy. At last information dwelling inside the ASDI ought to be the fundamental and definitive source, with information amendments and updates being made inside the ASDI condition as opposed to the ASDI going about as an information vault holding a duplicate of informational indexes.

Access Services characterize how Users get data and administrations. Access administrations may give a way to get to a whole (or part of an) informational collection or be an increasingly engaged conveyance of a spatially related capacity, for example, a location approval administration.

Provenance for spatial analysis

The provenance utilization in improving information advancement has been recognized in different spaces, including science, natural science, and software engineering [2]. Scientists in these spaces regularly manage information serious issues utilizing a suite of computational methodologies spoke to by Web administrations, work processes to get the infrastructure capacities [2, 6]. Many refined investigation steps are regularly required to acquire attractive answers for complex issues. To deliberately keep up history data of information investigation and to empower the sharing of such data, information provenance data on information deduction heredity and transformational procedures connected to information deduction is coupled with information escalated computational methodologies. Specifically, since these computational methodologies assume a significant job in GIS for geospatial information improvement and scattering, it is basic to fuse provenance into GIS. The utilization of provenance in GIS can be followed back to the information genealogy approach proposed by Lanter [10]. In any case, from that point forward little work has been done to examine the job of provenance GIS being developed and correspondence of geospatial learning. Since the novel attributes of GIS in spatial information dealing with an investigation, extraordinary thought is frequently required for the utilization of provenance to improve the explanatory capacities of GIS.

Information provenance, additionally called information genealogy, records the deduction history of an information item. In the earth science space, geospatial information provenance is significant because it assumes a noteworthy job in information quality and ease of use assessment, information trail tryout, work process replication, and item reproducibility. The age of the geospatial provenance metadata is normally combined with the execution of geo-handling work process. Their cooperative relationship makes them corresponding to one another and guarantees incredible advantage once they are coordinated. Notwithstanding, the heterogeneity of information and figuring assets in the disseminated condition developed under the administration arranged engineering (SOA) carries an incredible test to asset combination. In particular, the issues, for example, the absence of interoperability and similarity among provenance metadata models and provenance and work process, make snags for the coordination of provenance, and geo-preparing work process.

The architecture of a Semantic Service

The system depends on Service Align Architecture and is an accumulation of OGC web administrations, which speak with one another by basic information passing or planning a few exercises. It is made out of four components: specialist organization, administration dealer, administration customer, and cosmology server. Figure 2 below represents the four segments of the proposed structure in the rationale universe. The specialist organization supplies geospatial information; the administration customer enables clients to look and coordinate information from suppliers, and the administration dealer gives a library to accessible administrations. The metaphysics server guarantees the semantic interoperability of the ontologies of administration customers and suppliers.[8]

Fig. 2: Geospatial Semantic Web service component in the logic universe

The Ontology-based Catalogue Service

The online ontologies are put away in an ontology server. Administration customers and suppliers both can enlist their application ontologies to the cosmology server. The ontology server guarantees the inter-operability of the ontologies of administration customers and suppliers. As the ontologies might be made by various networks, therefore heterogeneity issues may emerge when incorporating the data from at least two ontologies. Finding appropriate spatial data in the open and conveyed web condition is a vital errand. The momentum routine with regards to the actualized clearinghouse and geoportal is to look through client intrigued geospatial information and administrations dependent on catchphrases in the metadata. While it is valuable to get the geospatial information by watchwords based quest for metadata, metadata still has semantic heterogeneity issues. Diverse metadata makers may utilize various names for a similar component. Moreover, metadata contains just restricted data to enable clients to look.

Perception is one of the most significant and inescapable application zones of geospatial information and in geographic data frameworks (GIS); it enables clients to investigate, blend, present and break down the hidden geospatial information in an intelligent manner. However, the representation of geospatial information discloses some long-standing difficulties to both the suppliers and clients. One such test is the information combination issue, which can be the coordination between geospatial information and furthermore between geospatial information and information from different areas. At the interim, the perception of geospatial information is additionally learning escalated from a cartographic point of view for both the suppliers and clients. For the suppliers, a wide scope of cartographic hypotheses is required to infer sense-making and cartographically palatable applications; and for the clients, the learning is required to translate the displayed information importantly. Furthermore, now and again the clients need to arrive at an abnormal state of subjective accord with the suppliers to all the more likely see the conveyed data from the representation applications.

The objective of the Study

This research aims to produce a geospatial data provenance model. It will investigate and implement semantic web techniques to aid the user in assessing the results of comprehensive queries by linking provenance features with other information available from spatial systems. The objective is to improve the accessibility and usability of spatial data at the country level, in the first instance, but the techniques created will be generic and applicable to global use.

The administration of cadastral information includes different worth and supply chains. Every ha heterogeneous geo-forms, strategies, models and work processes that join to create, alter and convey spatial information. The coordination and handling of numerous datasets offer to ascend to end client inquiries regarding trust, quality, and readiness for a reason, cash and definitiveness of the information. This is because datasets start from different sources, and distinctive geo-forms are executed to convey the last item. Seeing how information is gathered, prepared, oversaw and scattered gives learning about its history, acceptability and provenance. This thusly expands the convenience of information. [9] This research investigates strategies to catch spatial information provenance and information stream genealogy. The point is to build up a spatial information provenance model for the land organization area utilizing an exhaustive philosophy. Also, this PhD proposition fundamentally researches how the Semantic Web innovations can cultivate better coordination and perception of geospatial information, and thereof help the exceeding of geospatial information, data and learning to different spaces. The extent of the PhD proposal is expansive, and advantages raised by Semantic Web innovations will be shown in a couple of specific situations where some customary geospatial issues are better tackled with the Semantic Web. What’s more, in this structure, ontologies and guidelines are seriously utilized as two primary ideal models for learning portrayal.

Chapter 2

Background Study

According to the Merriam Webster lexicon, provenance is characterized as spot or wellspring of the source where something was made or gathered alongside the records of their exercises throughout proprietorship. Provenance may be utilized to decide genealogy, by following advances engaged with sourcing, moving, and preparing of information [23] or how it touched base in a specific database [3]. By account provenance, it is conceivable to accomplish a sound ancestry record that guarantees reproducibility of significant information change [22]. Myers et al. [15] contend that provenance record should be carefully discoverable, available, and intelligible to empower legitimization and review support on how the information was delivered, and give essential setting data to recreate information examination results. Three noteworthy methodologies have been produced for provenance portrayal: explanation, reversal and virtual information [20]. Explanation portrayal gives adaptability to create profoundly expressive provenance that can be utilized to record data at various granularity levels. The granularity of provenance additionally alluded to as provenance level is commonly delegated coarse or fine-grained. The alluring degree of granularity of provenance relies upon application space necessities, and the expense of accumulation, stockpiling and preparing. The better the granularity of the provenance record the higher the data entropy and related expense. Reversal strategies speak to capacities characterized alongside yield information that is adequate to distinguish information sources. Reversal techniques permit a minimal portrayal of provenance that empowers versatility. Be that as it may, these strategies are not all around relevant and the data gave is restricted to information induction history. In virtual information frameworks, an inventory is kept up for speaking to information induction techniques and inferred information, giving adequate data to re-figure yield as required [7]. The ability of provenance in improving critical thinking drives the utilization of provenance in the space of Geographic Information Science known as GIScience. For instance, Lanter built up a meta-database framework for following the procedure of the work process and a framework (Geolineus) for account Arc/Info GIS activities [10, 11]. Alonso and Hagen [1] showed the plan of a geo-process supported work process framework (Geo-Opera) that permits the following of geospatial examination history. These frameworks or methodologies in the present GIS writing give just heredity data as changes connected to determine information yet don’t record and manage spatial provenance methodically and strongly. Spatial provenance capacity in GIS makes spatial information taking care of and examination mindful of their determination history and, consequently, is custom-fitted to the portrayal and scattering of GIS-based learning.

Other than portrayal, provenance-mindful applications likewise need to guarantee provenance catch, the board, and recovery (Miles et al. 2007). Provenance the board commonly includes the utilization of a database or a Resource Description Framework (RDF) triple store. The recovery of provenance information is through a SQL (Structured Query Language) or SPARQL (Semantic Web Query Language) inquiry in a Semantic Web condition. This part of provenance has been researched widely, e.g., in Buneman et al. (2001), Cui et al. (2000), Wang et al. (2008b), Chebotko et al. (2010). Interestingly, the catch of information provenance, partic

ularly the components that are incorporated into a geo-processing work process is still new to the GIScience people group. To a limited extent, this is because of the multifaceted nature of the spatial investigation and the necessities to create, speak to, and move the provenance in a programmed way. For instance, Tilmes et al. (2013) build up an idea model to introduce the substance and provenance of the national atmosphere appraisal report created by the US Global Change Research Program. This provenance model, utilizing W3C PROV, can be named a post-occasion provenance since the provenance is caught after the appraisal on environmental change is finished. Yue et al. (2011) create formal systems for sharing existing information provenance through an expansion of the Open Geospatial Consortium (OGC) Eb-XML Registry Information Model (Eb-RIM). In any case, provenance catch isn’t tended to in this system. All the more as of late, Di et al. (2013) exhibited a cutting edge answer for catching provenance while the information is being created. The provenance, encoded in the ISO19115 ancestry model, is progressively recorded when a web administration is executed by a Business Process Execution Language (BPEL) motor.

Chapter 3

What is an ontology

The literature of Artificial Intelligence contains numerous meanings of a cosmology; a considerable lot of these repudiate each other. For the motivations behind this guide a cosmology is a formal unequivocal depiction of ideas in a space of talk (classes in some cases called concepts), properties of every idea portraying different highlights and qualities of the idea (slots sometimes called jobs or properties), and limitations on openings (aspects at times called job confinements). A philosophy together with a lot of individual cases of classes establishes an information base. There is an almost negligible difference where the metaphysics closes and the information base begins. Classes are the focal point of general ontologies. Classes depict ideas in the space. For instance, a class of wines speaks to all wines. Explicit wines are occurrences of this class. The Bordeaux wine in the glass before you while you read this record is an occasion of the class of Bordeaux wines. A class can have subclasses that speak to ideas that are more explicit than the superclass. For instance, we can partition the class of all wines into red, white, and rosé wines. On the other hand, we can separate a class of all wines into shining and non-shimmering wines.

In commonsense terms, building up Metaphysics includes:

• defining classes in ontology,
• arranging the classes in an ordered (subclass–superclass) hierarchy,
• defining spaces and depicting permitted values for these slots,
• filling in the qualities for openings for instances.

We would then be able to make a learning base by characterizing individual occasions of these classes filling in explicit space esteem data and extra space limitations.

Fig 3. Ontology life cycle

Why develop ontology

Geo-spatial philosophy improvement and semantic learning revelation tends to the requirement for displaying, investigating and imagining multi-modular data, and is novel in offering coordinated systematic that envelops spatial, transient and topical components of data and information. The extensive capacity to give incorporated examination from different types of data and utilization of express information make this methodology interesting. This likewise includes determination of spatial-transient topical ontologies and populating such ontologies with excellent learning. Such ontologies structure the reason for characterizing the significance of significant relations and terms, for example, close or encompassed by, and empower calculation of spatial-worldly topical closeness estimates we characterize.

Lately the advancement of ontologies unequivocal formal details of the terms in the area and relations among them [9] has been moving from the domain of Artificial-Intelligence research facilities to the work areas of space specialists. Ontologies have turned out to be regular on the World-Wide Web. The ontologies on the Web go from huge scientific classifications classifying Web destinations, to orders of items available to be purchased and their highlights. The WWW Consortium (W3C) is building up the Resource Description Framework [10], a language for encoding learning on Web pages to make it reasonable to electronic operators looking for data. The Defense Advanced Research Projects Agency (DARPA), related to the W3C, is creating DARPA Agent Markup Language (DAML) by expanding RDF with progressively expressive builds went for encouraging operator association on the Web [11]. Numerous controls currently create institutionalized ontologies that area specialists can use to share and explain data in their fields. The drug, for instance, has created enormous, institutionalized, organized vocabularies, for example, SNOMED [8] and the semantic system of the Unified Medical Language System (Humphreys and Lindberg 1993).

Ontologies Development Process

A philosophy characterizes a typical vocabulary for analysts. The information which changes incredibly in size, extension and semantics is conceptualized dependent on an idea metaphysics, which is the accomplishment of the semantic web accordingly the core of the semantic web. Philosophy can be huge or little containing a huge number of terms and connections. These can be made by information portrayal specialists or amateur web clients. Thus metaphysics picked up its significance in research to share data in a specific space in machine decipherable arrangement. Displaying specialists construct straightforward philosophy’s and examples to enter data effectively. Metaphysics together with a lot of individual examples of classes establish an information base. Classes are the focal point of the general treasury. Classes portray ideas in the area. A class can have sub-classes that speak to ideas that are more explicit than the superclass.

Fig 4. iterative steps for ontology development

System design

Here open source tools Protégé is used to evaluate for usability. Operational ontology improvement conditions have concentrated so far on the altering of ontologies in an independent, single-client mode. The developing of the Semantic Web and the Web 2.0 innovations carry another test to the philosophy of proofreader suppliers, the cooperative altering of ontologies in an electronic domain.

The Protege framework is an open-source metaphysics editor and learning base structure created by Stanford Medical Informatics. It utilizes casing based portrayal formalism. The Protege supervisor additionally underpins the altering of RDF and OWL3 ontologies.

Provenance Architecture in GIS

The need of joining provenance into GIS is as a rule progressively perceived, to a limited extent, as a result of the rise of land organization framework and its appropriateness for information escalated geospatial critical thinking. This paper tends to the requirements of provenance in GIS and portrays a provenance-mindful GIS design and its verification of-idea usage. The engineering depends on a conventional provenance model for the administration and utilization of spatial provenance, which permits the fuse of explicit spatial provenance abilities. The utilization cases exhibited are intended to be easy to uncover the capability of utilizing a-spatial, spatial, spatiotemporal provenance information for GIS applications. The spatial locale is presented as an essential reference unit for spatial provenance to digest the thought of both field-and article based spatial information portrayals. The provenance-mindful GIS prototyped in this examination can be stretched out to complex situations to help top to bottom thought of provenance in GIS. Provenance in GIS can speak to causal connections among spatial information and related examination methods. In our provenance-mindful GIS design, provenance abilities are inexactly combined with GIS usage and don’t require any new improvement interior to spatial databases. Our execution sorts out spatial provenance as items in spatial information store, while a-spatial procedure provenance will be spoken to utilizing semantic web innovations. This blend empowers a suite of provenance related tasks, including capacity and recovery. The free coupling procedure is proper to exploit nonexclusive provenance capacities that are stretched out with the capacity to deal with spatial provenance.

An administration arranged three-layered design as shown in the following Figure is created to help GIS to record and inquiry spatial provenance. The Web-administrations level encourages the capacity and inquiry of provenance from GIS applications. The base information level comprises of a semantic information storehouse and a spatial information store. The centre rationale level is made out of a lot of modules and extensions the top and base levels. This engineering gives an extensible and versatile system in which administrations and modules can be included and refreshed without disturbing leaving GIS usefulness. The design is especially appropriate to provenance recording at a coarse-grained level in that recorded provenance in this investigation is controlled at the degree of spatial districts instead of spatial articles. Spatial areas, undifferentiated from the thought of “fields of spatial items”, are a theoretical term to think about both field-and article based portrayals [4, 9]. It ought to be noticed that however provenance is recorded at a coarse-grained level in this examination setting, there does not impede for the engineering to help fine-grain provenance.

                                                Fig 5. Architecture of Provenance

All in all, when recording coarse-grained provenance, we may do without the ability to follow the genuine causality without knowing the subtleties of the calculations or different parts of information, since procedure provenance alone does not generally infer causal connections. For instance, a procedure produces information on the yearly precipitation in Denver does not utilize the day by day downpour aggregates for anyplace in the nation however Denver, regardless of the way that national information is input and is recorded as adding to the yield. If downpour sums for the east coast are observed to be in mistake in national information, the Denver yearly midpoints ought not to be influenced. We will be unable to induce this with coarse-grain provenance and may pointlessly re-process the Denver midpoints. But since we partner spatial districts along inside their provenance, we can alleviate the impact of this issue by regarding Denver as a spatial locale.

Data Model

To record and recover provenance from the semantic and spatial vault requires the improvement of a steady system to gather and store provenance. Thusly, this examination additionally centres on the advancement of a spatial provenance information model. Provenance related to each procedure incorporates properties for its name, parameters, and runtime condition. The causal relationship related to information determination is approved by utilizing begin time and end-time recorded for each procedure. The OPM additionally features the utility of timestamps in getting causality. Notwithstanding partner provenance record for each procedure, each dataset can likewise have a provenance record. This record contains traits that incorporate source, units, designs, transient range, and interchange information area related to it. Qualities (for example spatial degree, units, goals, and highlight) portray spatial provenance perspectives that can be pushed into the spatial information store. An affiliation is then made to interface spatial information to proper passages in the semantic store utilizing remote keys. This affiliation freely couples semantic information and topographically referenced information, which awards the capacity to question provenance dependent on spatial information.

Fig 6. Provenance data model attributes

A few explicit innovations are utilized to build up usage for engineering. Tupelo is utilized as the semantic substance storehouse for putting away, recovering and getting to provenance. Tupelo is middleware that gives a Web get to convention and Java API that interface with an RDF i.e. Resource Description Framework [17] mapping of the Open Provenance Model. RDF has been embraced as the W3C standard for metadata portrayal and is as a rule broadly used to speak to information models of geospatial semantic Web [5]. Provenance information is put away utilizing semantic Web innovations dependent on the RDF and is supported with standard stockpiling advancements (like a database) and RDF stores (like Sesame). In our execution, the connection among procedure and information and traits related with them are spoken to as triples, of the structure (subject, predicate, and article) that relate assets. Right now, MySQL database is utilized as the semantic storehouse for RDF information. The execution is interoperable with other RDF stores and semantic Web advancements.

Dataset feature and attribute-level provenance 

The PROV cosmology archive communicates the PROV-DM utilizing the W3C OWL2 Web Ontology Language (OWL2). It gives a lot of classes, properties and limitations that can be utilized to speak to and exchange provenance data. Utilizing these metaphysics, provenance can be encoded in the Resource Description Framework (RDF). RDF is a standard model for information trade on the web, broadening the connecting structure of the web to utilize URIs to name the connection among things and the two parts of the bargains, typically alluded together as a ‘triple’ (W3C Semantic Web, 2015). Subsequently, the RDF documentation permits portraying, catching and questioning provenance in a conveyed situation. There are a few RDF normal serialization designs; in this paper, we supported the utilization of Notation3 (N3). The utilization of RDF carries us closer to Linked Data (http://linkeddata.org), which permits the sharing of data in a manner that can naturally be perused by PCs and empowers information from various sources to be associated and questioned (24). In the geospatial world, Linked Data permits the setting of connections between various datasets, joining extra depictions to unique information (25) and enhancing the last datasets and maps.

Geographic Markup Language (GML) offers the likelihood to install an ISO record legitimately in an element or a component gathering by utilizing ‘GML:metadata property’ to reference the provenance data. In particular, the ‘xlink:href’, ‘xlink:role’ and ‘xlink:arcrole’ credits were proposed to completely depict the relationship of highlights and ascribes to the provenance components in the dataset-level provenance document. In any case, metadata property was as of late deplored in GML 3.2. Hence, this choice isn’t suggested. Likewise, the likelihood of characterizing a perplexing property type got from ‘AbstractMetadataPropertyType’ was additionally investigated, however, this requires option in the GML pattern, which isn’t constantly conceivable. Sadly, there is an absence of agreement on the best way to execute provenance at the element and property levels.

Provenance is data about substances, exercises and individuals associated with creating a bit of information or a thing, which can be utilized to survey quality, dependability and reliability. PROV characterizes a provenance information model to help the interoperable trade of provenance in heterogeneous situations, for example, the web. The provenance centre structure depends on the meaning of the elements, exercises and operators that are associated with delivering a bit of information or a thing and on how they are connected by characterizing the accompanying four property types: wasGeneratedBy, wasAssociatedBy, was attributed to and utilized

            Fig 7. Geospatial data concept with provenance core element [Guillem Close et al.]

References:

[1] Janowicz, K., Schade, S., Bröring, A., Keßler, C., Maué, P., Stasch, C. (2010) Semantic enablement for spatial data infrastructures. Transactions in GIS 14,111-129.

[2] Zhang,  C.  and  Li,  W.  (2005) The roles of  Web Feature and Web  Map Services in real-time geospatial data sharing for time-critical applications, Cartography and Geographic  Information  Science,  vol.32,  no.  4,  pp.  269-283.

[3] F. Hakimpour, S. Timpf, “Using ontologies for resolution of semantic heterogeneity in GIS,” Paper presented  at the  Proceedings of 4th  AGILE  Conference  on  Geographic  Information  Science, Brno (2001).

[4] I. Suriarachchi, “Survey   of   Provenance   Practices   in   Data   Preservation   Repositories,” https://www.rd-alliance.org/filedepot, [Last Accessed: 29/05/2015]

[5]Lianlian  He;  Peng  Yue;  Liping  Di;  Mingda  Zhang;  Lei  Hu,  “Adding  Geospatial  Data  Provenance  into  SDI—A  Service-Oriented  Approach,”  in Selected  Topics  in  Applied  Earth  Observations and Remote Sensing, IEEE Journal of , vol.8, no.2, pp.926-936, Feb. 2015

[6]T. Liu,., S.  Yanfeng,  D. Smith, A.Terhorst,  “Provenance-Aware  Automated  Data  Quality  Control”,   Paper   presented   at   the   eResearch   Australasia   2011,   Melbourne.   Conference   Material retrieved from http://conference.eresearch.edu.au/eres2011/ presentations/.

[7]G.  West,“Research  Strategy  Spatial  Infrastructure, (Program 3)”.          Updated.     Retrieved: March 2016,     from     www.crcsi.com.au/Resources/   Research/   P3-final-Research-Strategy.aspx

[8] Zhang,  C.,  Li,  W.,  Day,  M.  and  Peng,  Z.R.  (2003)  GML-based  interoperable  geographical  database, Cartography, vol. 32, no. 2, pp.1-16.

[9] Muhammad Azeem Sadiq et. al.,  (2016). Spatial Data Supply Chain Provenance Modelling for Next Generation Spatial Infrastructures Using Semantic Web Technologies. GEOProcessing 2016 : The Eighth International Conference on Advanced Geographic Information Systems, Applications, and Services

[10] Brickley, D. and Guha, R.V. (1999). Resource Description Framework (RDF) Schema Specification. Proposed Recommendation, World Wide Web Consortium: http://www.w3.org/TR/PR-rdf-schema

[11] Hendler, J. and McGuinness, D.L. (2000). The DARPA Agent Markup Language. IEEE Intelligent Systems 16(6): 67-73.

[12] Price, C. and Spackman, K. (2000). SNOMED clinical terms. BJHC&IM-British Journal of Healthcare Computing & Information Management 17(3): 27-31.

[13] Humphreys, B.L. and Lindberg, D.A.B. (1993). The UMLS project: making the conceptual connection between users and the information they need. Bulletin of the Medical Library Association 81(2): 170

[14] D.P. Lanter. Design of a lineage-based meta-database for GIS. Cartography and Geographic Information Systems, 18(4), pp 255-261, 1991

[15] D. P. Lanter. A Lineage Meta-Database Approach Toward Spatial Analytic Database Optimization. Cartography and Geographic Information Systems v.20(2), pp 112-121, 1993.

[16] L. Moreau and I. Foster (Eds.). Provenance and Annotation of Data – International Provenance and Annotation Workshop, IPAW 2006 (Lecture Notes in Computer Science, vol. 4145). Springer: Berlin, 2006.

[17] L. Moreau, J. Freire, J. Futrelle, R. McGrath, J. Myers and P. Paulson. The Open Provenance Model, http://eprints.ecs.soton.ac.uk/14979/1/opm.pdf, 2007.

[18] L. Moreau., P. Groth, S. Miles, J. Vazquez-Salceda, J. Ibbotson, S. Jiang, S. Munroe and O. Rana, A. Schreiber, V. Tan, L. Varga. The provenance of electronic data. Communications of. ACM 51, 4, pp 52-58, Apr. 2008.

[19] J. D. Myers, J. Futrelle, J. Gaynor, J. Plutchak, P. Bajcsy, J. Kastner, K. Kotwani, J.S. Lee, L. Marini, R.E.McGrath and Y. Liu. Embedding Data within Knowledge Spaces. In UK escience 2008, Edinburgh, UK, Sept 2008.

[20] M. Perry, F. Hakimpour, A. Sheth. Analyzing Theme, Space and Time: An Ontology-based Approach. 14th International Symposium on Advances in Geographic Information Systems (ACM-GIS ’06), Arlington, VA, pp. 147-154, Nov, 2006

[21] Resource Description Framework (RDF): Concepts and Abstract Syntax. W3C Recommendation. http://www.w3.org/ TR/rdf-concepts/. Feb, 2004.

[22] A. Sheth and M. Perry. Traveling the Semantic Web through Space, Time and Theme. IEEE Internet Computing, v. 12(2), pp. 81-86, March/April 2008.

[23] S. S. Sahoo, A. Sheth, and C. Henson. Semantic Provenance for eScience: Managing the Deluge of Scientific Data. IEEE Internet Computing, v. 12(4), pp. 46-54, 2008.

[24] Y. Simmhan, B. Plale and D. Gannon. A Survey of Data Provenance in e-science. In SIGMOD Record, v. 34, pp 31-36, 2005.

[25] SPARQL Query Language for RD. W3C Recommendation. http://www.w3.org/TR/rdf-sparql-query/, Jan 2008.

[26] M. Szomszor and L. Moreau. Recording and reasoning over data provenance in Web and grid services. In International Conference on Ontologies, Databases and Applications of Semantics (ODBASE’03), pp 603-620, Nov 2003.

[27] C. Bizer, T. Heath, T. Berners-Lee Linked data—The story so far International Journal on Semantic Web and Information Systems, 5 (3) (2009), pp. 1-22

[28] L.M. Vilches-Blázquez, B. Villazón-Terrazas, O. Corcho, A. Gómez-Pérez Integrating geographical information in the linked digital earth International Journal of Digital Earth, 7 (7) (2014), pp. 554-575, 10.1080/17538947.2013.783127

[29] Hu, Y., K. Janowicz, S. Prasad & S. Gao (2015) Metadata Topic Harmonization and Semantic Search for Linked‐Data‐Driven Geoportals: A Case Study Using ArcGIS Online. TransactionsinGIS,19, 398‐416.

[30] Li, W., V. Bhatia & K. Cao (2015) Intelligent polar cyber infrastructure: enabling semantic search in geospatial metadata catalogue to support polar data discovery. Earth Science Informatics, 8,111‐123.

[31] Li, W., S. Wang & V. Bhatia (2016) Polar Hub: A large‐scale web crawling engine for OGC service discovery in cyber infrastructure. Computers, Environment and Urban Systems, 59, 195‐207.

[32] Li, W., X. Zhou & S. Wu (2016) An Integrated Software Framework to Support Semantic Modeling and Reasoning of Spatiotemporal Change of Geographical Objects: A Use Case of Land Use and Land Cover Change Study. ISPRS International Journal of Geo‐Information, 5, 179.