Transform unstructured e&p information
1 like208 views
The document discusses the need for an efficient content discovery and management solution tailored to the energy and petroleum sector, utilizing machine learning and semantic web technologies. It emphasizes the implementation of an ontology-driven search engine that enhances the retrieval of relevant geoscience and engineering information, allowing users to access personalized content quickly and effectively. Additionally, the document highlights the challenges and methodologies involved in developing a comprehensive ontology for geological data interoperability and intelligent information retrieval.
Transform unstructured e&p information
- 1. You work with E&P information Butyouneedtheknowledgenow! Work on a specific topic What internal and external information on topic is availble? Can not find the unstructured information? Report unstructured information Do not have time to read whole articles, news or journals? Existing systems does not cooperate to give one report? Technology restrict you transfor information to knowledge Existing applications does bring the applicable information as you want it? You know the information is there, but do not know where? Easyto use – A click on a button solution Do you have to be a programmer or expert to find the information? You have to crawl many data repositories before even getting the information?
- 2. You need Content Discovery and Data on Audiences and Content Relevant content across different sourcesinternally and externally, automaticallyon each project blog/page. Understand employee interestsand Increase cost efficiency with dynamic segments. Maximizetraffic and efficiency with intelligentcontentdiscovery and ad targeting.
- 3. A solution for Content Discovery and Data on Audiences and Content A Spark server can uniquely perform capable content discovery services increase traffic for project workers, managers and researchers. A Spark enabled server can precisely target content where other methods fail. A Spark server can provide data that is not available through other means. Text analysis can be able to understand the meaning of content deeper than any other solution. Show project workers, management or researchers relevant content across sites, automatically on each their project/ department or company page. Understand the user’s interests and accurately personalize the service. Search anything with natural language – no engineering skills required to find what the employee is looking for. Quickly create tailored audience segments that is of interest. Understand employee interests in media with unprecedented depth. Offer dynamic creatives that precisely respond to context and employee interests to maximize their attention on topic of interest.
- 4. Apache Spark Apache Spark is a data processing framework that supports buildingprojects in Python and comes with MLlib, distributedmachine learning framework. Creating a recommendationengine using Spark written in Python could be a first entry point for the real intelligentinformationsearch engine for an employee working within specific domains. F.inst. If an employee works within geoscience or engineering domains, a recommendationsearch engine with a domainbased ontology could make it much simpler to retrieve the relevantinformation at any given time and place. Trainingusing a collaborativefiltering model (CFM) could be a plausiblemachine learning (ML) method in this case to provide the appropriateretrievalresults. The recommendationengine creates a recommendationreport based on the result from the CFM.
- 5. Make content discovery effortless and fun through ontology based machine learning Are you interestedin what changes and planningare happeningin your are of interest or work? The service should automaticallyinform the employee whenever there is new information available regarding the topics of interest. Content has been difficult or even impossibleto find with standardInternet search engines. They either fail in understandingthe topics accurately,or can’t locate the right information of that particular subject. Or even worse, the effort has been left for the employeeto crawl through the deep dungeons of hundreds of web sites. Turn the search based traditionalconcept upside down and put the user in the center. Combine altogether millionsof articles and documents, altogetherhundreds of gigabytesof data – enough content to fill several miles of library bookshelves– analyze with Spark server with appropriateontologyof more than a millioninterrelatedconcepts. This includes thorough disambiguationwhere terms such as f.inst. specific geological or engineering terms are matched with their correct meanings.
- 6. What is disambiguation? Here is one example where things can get complicated: What is the difference between Paris (France) and Paris (Tennessee, USA)? Disambiguationrefers to a process of identifying words or phrases that have multiplemeanings and finding out which meaning is correct in its particular context. We are dealing with: • synonymsof geologicalthemes or words • themes or words can have multiplemeanings Spark enabledserver can find out the correct meaning in each particular case so that the service can return results from the relevant theme and topics, removing matches that have the same words with different meaning.
- 7. A system which understands your interests The system analyzes user interests in real time and presents employersrelevantcontent. Both behavioral and contextual contentrecommendationscan be shown on project/ blog or any other relevantsite. More importantlythese recommendationscan be easily integratedto any online service, such as the company/ department or any other relevantsite, through a widget platform. As a result, the employee will get a unified and seamless service conceived by the personallymost relevantinformation from the multitudeof information sources which are deemed to be of importance to him or her. Provide easy access to all kinds of public and/ornon-publicinformation in your work environmentto cater for just the employer need in time and place.
- 8. Use an ”Intelligent Search” engine in-house • Understand the meaning of the search query; and notjust extractthe wordsyou search for. • Possible to find all relevantresults from the scope of your search, notjustthe ones with the specified keywords. You work withTertiary stratigraphic plays inthe NorthSeaarea, andare working in a project identifying traps, risking andranking prospects withinthis stratigraphic interval. Youand your teammember has an Machine learning Searchengine poweredwithSpark searchengine whichyouuse to searchfor information, bothin-house and externally to assistyou inthis work. Entering the searchquery “Howdo I risk a stratigraphic play?” will get youinformationon howto risk sucha play withinthe region you work or interestedinor analogues relevant tothis, dependentuponhowyou usedthe systempreviously. It alsogives you results about Tertiary plays witha stratigraphic elementor solely are stratigraphic. • Disambiguatestopics are sorted out, given the already knowledge aboutthe employee and its interest and worktopics. • Intelligent autocomplete willassist the employee to save time searching information required to create a knowledgeable solution. • The search will relate to both to short phrasesor large blocksof text. • Typing in an web addressin the search the return resultsmatch the contentrequested on the web address.
- 9. Ontology – the keyword to success. Creating a comprehensive purpose built ontologywithin the areas the company is working with. The areas could be G&G, Engineering, economics etc. The limit is to what your scope should be and required. The IntelligentSearch Engine will be required to identify a large amount of concepts and use specified rules. These rules and categories will have to be manuallycrafted by a dedicatedteam of professional linguists deployedwithin the company and they are required to continuouslyupdatein a ontology database.
- 10. Domain Ontology Challengeis the accumulationof large volumes of structured and unstructured data. Important endeavorsto fully exploit the knowledge and information that are included in geological big data and improving the accessibility of large volumes of data.The architecture of the geologicalsurvey information cloud-computing platform (GSICCP) and big-data-relatedtechnologiesenable us to split geologic unstructured data into fragments and extract multi- dimensionalfeatures via geologicaldomain ontology. A geological ontologyis a domain ontologythat describes the knowledge of a geological field.
- 11. Geoscience domain Ontology There are severalchallenges in applying an ontology spaceto achieve geological data interoperability and ability to create an ontology based semantic search engine in the Geological domain. • Modeling versus encoding • Multilinguality • Flexibility and usefulness • Mediation and evolution The goal is to providean overview to be applied in an ontology spacewhere geological data can be searched in a open semantic manner and being interoperational. Use of a RDF/OWL based ontology for various themes within geology seems to be a way to go in order to create easy to use functions for conceptual modeling and encoding for the Semantic Web. With proper extensions, the SKOS is functional for encoding multilingual geoscience thesauriinto a formatthat is compatible with the Semantic Web, and SKOS-based multilingualgeoscience thesauri are efficient for translating online geoscience records into any language that is supported by the thesauri. By using ontologies, innovative applications can be developed to promote geological data interoperability at local, regional and global levels
- 12. Semantics Study of meaning—in language, programming languages, formallogics, and semiotics. Itfocuses on the relationship between signifiers— like words, phrases, signs,and symbols—and whatthey stand for, their denotation. Semantics contrasts with syntax, the study of the combinatorics of units of a language (without referenceto their meaning), and pragmatics, the study of the relationships between the symbols of a language, their meaning, and the users of the language. The Semantic Web takes the solution further. Itinvolves publishing in languages specifically designed for data: ResourceDescription Framework (RDF), Web Ontology Language(OWL), and Extensible Markup Language(XML). HTML describes documents and the links between them. RDF, OWL, and XML, by contrast, can describearbitrary things such as people, meetings, or airplane parts. These technologies are combined in order to providedescriptions that supplement or replace the content of Web documents. Thus, content may manifestitself as descriptive data stored in Web-accessible databases, or as markup within documents (particularly, in Extensible HTML (XHTML) interspersed with XML, or, moreoften, purely in XML, with layoutor rendering cues stored separately). The machine-readable descriptions enable content managers to add meaning to the content, i.e., to describethe structureof the knowledge we haveabout that content. In this way, a machine can process knowledgeitself, instead of text, using processes similar to human deductive reasoning and inference, thereby obtaining moremeaningful results and helping computers to performautomated information gathering and research.
- 13. The Semantic Web Stack The Semantic Web Stack illustrates the architecture of the Semantic Web. The functions and relationships of the components can be summarized as follows: XML provides an elemental syntax for content structure within documents, yet associates no semantics with the meaning of the content contained within. XML is not at present a necessary component of Semantic Web technologies in most cases, as alternative syntaxes exists, such as Turtle. Turtle is a de facto standard, but has not been through a formal standardization process. XML Schema is a language forproviding and restricting the structure and content of elements contained within XML documents. RDF is a simple language forexpressing data models, which refer to objects ("web resources") and their relationships. An RDF-based model can be represented in a variety of syntaxes, e.g., RDF/XML, N3, Turtle, and RDFa. RDF is a fundamental standard of the Semantic Web. RDF Schema extends RDF and is a vocabulary for describing properties and classes of RDF- based resources, with semantics for generalized-hierarchies of such properties and classes. OWL adds more vocabulary for describing properties and classes: among others, relations between classes (e.g. disjointness), cardinality (e.g. "exactly one"), equality, richer typing of properties, characteristics of properties (e.g. symmetry), and enumerated classes. SPARQL is a protocol and query language forsemantic web data sources. RIF is the W3C Rule Interchange Format. It's an XML language forexpressing Web rules that computers can execute. RIF provides multiple versions, called dialects. It includes a RIF Basic Logic Dialect (RIF-BLD) and RIF Production Rules Dialect (RIF PRD).
- 14. Ontolgy driven Semantic search engine concept. REST API Content Analyzer Ontology User Analyzer Content Record Match results User Record Widget Engine The API can be easily used with any language with REST, benefiting both you and your clients. You can access the Search Engine system with smartphone and tablets whenever from wherever! Integrate into the company backend system. Ontology-based technology takes the accuracy and domain of applicability of automatic targeting to new levels. Develop extensive standards-based ontology and a reasoning engine as a foundation of Content and User Analyzer. Concepts and extraction patterns, enables analyzing the topics present in textual content and build a detailed user record based on actual (rather than stated) user interests.
- 15. Om mig Stig-Arne harerfarenhetsomföretagsledare medbetydandeföretagserfarenhetochbakgrundbåde somgeolog,ingenjör,lärare och karriärcoach. Dessutomharhan storerfarenhetavatt träna ochcoacha yrkesverksammamedbehovavattintegrerasig. Stighar arbetatsom enseniorgeologochprojektledare förfleraolje- ochgasbolag,inklusive serviceföretag.Stigharhaftansvarföratt förvalta inomhusochklientbaserade geovetenskapsprojektavolikaslag,alltfrånintroduktionsnivå,renprospekteringtillutvecklingsfasfrånbörjantill slut. Han har genomgåttgeologiskautvärderingarochbidragittill multidisciplinstudierförmöjligheter,prospektochfält,inklusive planeringoch genomförandeavplaneringavhelaprospekterings- ochreservoarexploateringsprogram. Han har långerfarenhetavprojektarbetenavolikaskalaochtyperochvan att vara i kontaktmedkunderföratt svara på frågor somuppstårunder ettprojekt. Stighar fokuspå kunskapsbaseradeprocesserochsystem,kontraktsuppdrag,tillgångsförhandlingar.Stig-Arne harhaftfokuspårumsliganalysmedfokuspåkunskapsbaserade E& P- processeruppnåraffärsmål medhjälpavprediktivanalys. Stigär väl käntför användningavindustriprogramvarainomgeovetenskap,kunskapshantering,affärsanalys,geospatialanalysosv. Han fungerarsomcoach för såväl individersomföretag. https://www.linkedin.com/in/stigarne/ https://plus.google.com/+StigArneKristoffersen https://www.facebook.com/ukranova https://twitter.com/ukranova
- 16. Referenser • Debajyoti Mukhopadhyay, Aritra Banik, Sreemoyee Mukherjee, Jhilik Bhattacharya (2008) A Domain Specific Ontology Based Semantic Web Search Engine • Kitcher, Philip; Salmon, Wesley C. (1989) Scientific Explanation. Minneapolis, MN: University of Minnesota Press. p. 35. • Euzenat, Jerome (2007). Ontology Matching. Springer-Verlag Berlin Heidelberg p. 36 • Nerbonne, J. (1996) ; The Handbook of Contemporary Semantic Theory (ed. Lappin, S.), Blackwell Publishing, Cambridge, MA • Cruse, Alan (2004); Meaning and Language: An introduction to Semantics and Pragmatics, Chapter 1, Oxford Textbooks in Linguistics • Artem Chebotko and Shiyong Lu (2009), "Querying the Semantic Web: An Efficient Approach Using Relational Databases", LAP Lambert Academic Publishing, ISBN 978-3-8383-0264-5. • “OWL Web Ontology Language Overview". World Wide Web Consortium (W3C). February 10, 2004. Retrieved November 26 • "Resource Description Framework (RDF)". World Wide Web Consortium. • Allemang, D., Hendler, J. (2011). "RDF –The basis of the Semantic Web. In: Semantic Web for the Working Ontologist (2nd Ed.)“ • Kuriakose, John (September 2009). "Understanding and Adopting Semantic Web Technology". Cutter IT Journal. Cutter Information Corp. 22 • Timo Honkela, Ville Könönen, Tiina Lindh-Knuutila and Mari-Sanna Paukkeri (2008). "Simulating processes of concept formation and communication". Journal of Economic Methodology. • Geological Time Formalization: an improved formal model for describing time successions and their correlation Michel Perrin, Laura S. Mastella, Olivier Morel, Alexandre Lorenzatti (2011) Earth Science Informatics 4, 2 81-96 • Ontological foundations for petroleum application modeling Ricardo Werlang, Mara Abel, Michel Perrin, Joel Luis Carbonera, Sandro Rama Fiorini (2014) Conference Paper • A geologic timescale ontology and service, Simon Jonathan David Cox, Stephen Miller Richard (2014) Earth Science Informatics • Recent progress on geologic time ontologies and considerations for future works, Xiaogang Ma, Peter Fox (2013) Earth Science Informatics • Ontology-aided annotation, visualization, and generalization of geological time-scale information from online geological map services, Xiaogang Ma, Emmanuel John M. Carranza, Chonglong Wu, Freek D. van der Meer (2012) Computers & Geosciences 40:107-119 • Ontology spectrum for geological data interoperability (2011) Xiaogang Ma, PhD thesis University of Twente