Dissertation/ Thesis
Achieving database semantic interoperability with a perspective of information flow and formal concepts
| Τίτλος: | Achieving database semantic interoperability with a perspective of information flow and formal concepts |
|---|---|
| Συγγραφείς: | Yang, Guanghui |
| Στοιχεία εκδότη: | University of the West of Scotland, 2016. |
| Έτος έκδοσης: | 2016 |
| Θεματικοί όροι: | 388.3 |
| Περιγραφή: | Databases are ubiquitous and database interoperability and integration becomes a challenge in which how to handle heterogeneity and to achieve semantic matches between database items is a central theme. This study is concerned with the semantic aspect of this topic and looks at the problem of how to identify, align and integrate those databases items for which little syntactical clue of relatedness is available and yet they are semantically related in one way or another. Our work is based on the conviction that concepts over the whole spectrum of data, information and knowledge would have to be mobilized in order to tackle non-trivial tasks in this field and arrive at a generic method. Specifically we address the following research question: how semantic interoperability between relational databases may be achieved with a perspective of data, information and knowledge, and develops an approach to it based on the notions of information flow as defined in the Information channel theory and formal concepts as defined in the theory of Formal concept analysis. Firstly we look at a relational table as an abstracted distributed system, and construct a series of information channels to represent it including a minimal cover of it. These enable us to identify and formalize the relationship between data and information in a relational database and define what we call the data level information content for data, instances, and whole databases. Secondly, we follow Dretske’s idea that knowledge is a belief that is caused or causally sustained by information, and use theory interpretation and formal concept analysis to form human centered knowledge discovery from databases. Knowledge discovered is then accumulated whereby to transform a relational database into a conceptual knowledgebase. Thirdly, we develop a new semantic alignment method with which all the information channels that represent component databases are aligned together. Finally, we construct a database semantic interoperability and integration framework and define an integrated information system for covering the databases involved. This system is capable of supporting any user to achieve semantic interoperability and integration by using their own data level correspondences between databases, and enabling human centered semantic query. Through this study, we find that to tackle problems of relational database semantic interoperability, it is necessary to approach them with concepts across the whole spectrum of data, information and knowledge and relationships between them, and furthermore to develop a sound method these concepts should be formalized and fit together. We have discovered one way to achieve this, namely we can use the Information channel theory and the theory of Formal concept analysis that formalize information flow and formal concepts respectively within a coherent conceptual framework. One result of such an approach is an integrated information system whereby we can achieve semantic interoperability between autonomous databases. |
| Τύπος εγγράφου: | Thesis Or Dissertation |
| Γλώσσα: | English |
| Σύνδεσμος πρόσβασης: | https://ethos.blethoshyku.com/concern/thesis_or_dissertations/847399 |
| Αριθμός Καταχώρησης: | edsble.847399 |
| Βάση Δεδομένων: | British Library EThOS |
καταχωρήστε σχόλιο πρώτοι!