China
Africa
Explore chapters and articles related to this topic
Functional Architecture for Knowledge Semantics
Published in Denise Bedford, Knowledge Architectures, 2020
The most common forms of ontologies are upper ontologies, domain ontologies, interface ontologies, and process ontologies. An upper ontology is not unlike an object model and an object class – it tells us about the nature of primary, secondary, and tertiary categories and their attributes. A domain ontology elaborates on the concepts relevant to a particular topic, context, or area of interest. An interface ontology elaborates and relates the concepts pertinent to the juncture of two disciplines. Finally, a process ontology elaborates on the inputs, outputs, constraints, and process concepts associated with an operation or function. In a nutshell, any of the semantic structures described earlier could be represented as and referred to as an ontology. From our experience, ontologies are described because of the encoding methods used to represent them – the RDF schema or languages used to represent them.
Reverse Engineering the Organizational Processes: A Multiformalism Approach
Published in Wasim Ahmed Khan, Ghulam Abbas, Khalid Rahman, Ghulam Hussain, Cedric Aimal Edwin, Functional Reverse Engineering of Machine Tools, 2019
Modern enterprises employ a host of different technologies to undertake their missions. The advancements in communication and sensor technologies, and availability of affordable computational and storage resources, have offered unprecedented opportunities for enterprises to develop new, faster, and more efficient ways to undertake their traditional and new missions made possible by the convergence of technologies. With the use of these technologies and accessible computing power, the underlying process model, employed by the enterprises, becomes the key enabler of these missions. The performance of an organization, as a result, depends not only upon the systems, algorithms, and resources employed but also on the way these entities are utilized in a process workflow model. An underlying process model that utilizes the resources in an efficient and intelligent manner easily outperforms another that might have access to the same resources but employs an inefficient process model. In large-scale, complex enterprises, e.g., manufacturing, aviation, banking, command and control, etc., these underlying processes are so vital and important to the overall enterprise’s yield or performance that their details are closely guarded and are not available outside an organization. In this paper, we present an overview of an approach developed for reverse engineering an enterprise’s underlying process model in the form of an executable workflow. The proposed approach can also be used to develop solution workflows for new decision and/or inference problems. The approach is based on the premise that domain knowledge, which includes domain-specific data, concepts, their structure, relationships, and employed algorithms, together with a multiformalism integration platform are means to understanding and recreating an enterprise’s process model. A domain ontology with concepts, their properties, and mutual relationships, allows reasoning about data available or generated by an enterprise. A process ontology with relevant algorithms and computational entities allows reasoning about the way these entities can be put together to solve a problem that the individual entities cannot solve on their own. This ontology identifies the possible workflows that can be instantiated with the available data and/or computational entities employed by an enterprise. Finally, a multimodeling integration platform allows implementation of the instantiated workflow incorporating desperate computational entities and execution of the integrated workflow for evaluation (e.g., sensitivity analysis, performance, etc.) purposes.
Enriching the functionally graded materials (FGM) ontology for digital manufacturing
Published in International Journal of Production Research, 2021
Munira Mohd Ali, Ruoyu Yang, Binbin Zhang, Francesco Furini, Rahul Rai, J. Neil Otte, Barry Smith
The Product Life Cycle Ontologies3 are developed through the CHAMP project and funded by DMDII (Otte et al. 2019). The Product Life Cycle Ontologies descend from BFO and CCO and are designed to assist in managing discordant manufacturing data sources that concern the whole life of a product in the market from design through manufacture, test, use, maintenance, and finally disposal for all phases. The ontologies include six mid-level ontologies, including a Manufacturing Process Ontology. In fact, the Additive Manufacturing Ontology (AMO) (Ali et al. 2019) is developed by extending the Manufacturing Process Ontology from The Product Life Cycle Ontologies.
Knowledge on-demand: a function of the future spatial knowledge infrastructure
Published in Journal of Spatial Science, 2021
Lesley M. Arnold, David A. McMeekin, Ivana Ivánová, Kylie Armstrong
Semantic Web enabled data can be queried once it is accessible on the Web and a current best practice for this is via APIs. Query processing has eight steps that constitute the automation of query services (Figure 7). The SKI architecture for processing a query is shown in Figure 8. Query: End user specifies a query through a query interface;Profile: End user and computer usage profiles are used to better understand the end user’s context;Interrupt: Decompose the query, e.g. using Natural Language Processing, and then search for data that potentially contain the answer to the user’s question or that could be used in constructing the answer;Retrieve: Semantic queries operate on the Web of Data and use rules and RDF triples to process the relationships between information and infer answers. Semantic queries retrieve both explicit and implicitly derived information based on the syntactic, semantic and structural information contained in data (Reed et al. 2016). Spatial filtering is then applied, further increasing the relevance of results by extracting locations from the search terms and looking at similar locations;Process: Responding to a complex end user query automatically, requires the chaining of resources, such as Web services and various data sources, to form an orchestrated query processing workflow. For example, ‘What are the chances of flooding in my area?’ requires a series of geoprocesses to be executed. The actual rules that control the workflow and execute and run the processes are managed with a process ontology;Portray: Visualise query results, for instance as map, image, chart, text, table, video;Rank and Rate: Communicate fitness-for-purpose with rankings and ratings. Rankings make it possible for a user to evaluate complex information according to certain criteria. This is the approach commonly used by Internet search engines. Ratings are assigned to communicate trustworthiness of information to the end user e.g. accuracy, completeness, timeliness, cost, provenance; andDeliver: knowledge as a single response or as a list of ranked results from which the user can choose.