Kam, Po-shan. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2000. / Includes bibliographical references (leaves 89-97). / Abstracts in English and Chinese. / Abstract --- p.i / Acknowledgements --- p.ii / Chapter 1 --- Introduction --- p.1 / Chapter 1.1 --- Data Mining --- p.1 / Chapter 1.2 --- Temporal Data Management --- p.2 / Chapter 1.3 --- Temporal reasoning and temporal semantics --- p.3 / Chapter 1.4 --- Temporal Data Mining --- p.5 / Chapter 1.5 --- Motivation --- p.6 / Chapter 1.6 --- Approach --- p.7 / Chapter 1.6.1 --- Focus and Objectives --- p.8 / Chapter 1.6.2 --- Experimental Setup --- p.8 / Chapter 1.7 --- Outline and contributions --- p.9 / Chapter 2 --- Relevant Work --- p.10 / Chapter 2.1 --- Data Mining --- p.10 / Chapter 2.1.1 --- Association Rules --- p.13 / Chapter 2.1.2 --- Classification --- p.15 / Chapter 2.1.3 --- Clustering --- p.16 / Chapter 2.2 --- Sequential Pattern --- p.17 / Chapter 2.2.1 --- Frequent Patterns --- p.18 / Chapter 2.2.2 --- Interesting Patterns --- p.20 / Chapter 2.2.3 --- Granularity --- p.21 / Chapter 2.3 --- Temporal Database --- p.21 / Chapter 2.4 --- Temporal Reasoning --- p.23 / Chapter 2.4.1 --- Natural Language Expression --- p.24 / Chapter 2.4.2 --- Temporal Logic Approach --- p.25 / Chapter 2.5 --- Temporal Data Mining --- p.25 / Chapter 2.5.1 --- Framework --- p.25 / Chapter 2.5.2 --- Temporal Association Rules --- p.26 / Chapter 2.5.3 --- Attribute-Oriented Induction --- p.27 / Chapter 2.5.4 --- Time Series Analysis --- p.27 / Chapter 3 --- Discovering Temporal Patterns for interval-based events --- p.29 / Chapter 3.1 --- Temporal Database --- p.29 / Chapter 3.2 --- Allen's Taxonomy of Temporal Relationships --- p.31 / Chapter 3.3 --- "Mining Temporal Pattern, AppSeq and LinkSeq" --- p.33 / Chapter 3.3.1 --- A1 and A2 temporal pattern --- p.33 / Chapter 3.3.2 --- "Second Temporal Pattern, LinkSeq" --- p.34 / Chapter 3.4 --- Overview of the Framework --- p.35 / Chapter 3.4.1 --- "Mining Temporal Pattern I, AppSeq" --- p.36 / Chapter 3.4.2 --- "Mining Temporal Pattern II, LinkSeq" --- p.36 / Chapter 3.5 --- Summary --- p.37 / Chapter 4 --- "Mining Temporal Pattern I, AppSeq" --- p.38 / Chapter 4.1 --- Problem Statement --- p.38 / Chapter 4.2 --- Mining A1 Temporal Patterns --- p.40 / Chapter 4.2.1 --- Candidate Generation --- p.43 / Chapter 4.2.2 --- Large k-Items Generation --- p.46 / Chapter 4.3 --- Mining A2 Temporal Patterns --- p.48 / Chapter 4.3.1 --- Candidate Generation: --- p.49 / Chapter 4.3.2 --- Generating Large 2k-Items: --- p.51 / Chapter 4.4 --- Modified AppOne and AppTwo --- p.51 / Chapter 4.5 --- Performance Study --- p.53 / Chapter 4.5.1 --- Experimental Setup --- p.53 / Chapter 4.5.2 --- Experimental Results --- p.54 / Chapter 4.5.3 --- Medical Data --- p.58 / Chapter 4.6 --- Summary --- p.60 / Chapter 5 --- "Mining Temporal Pattern II, LinkSeq" --- p.62 / Chapter 5.1 --- Problem Statement --- p.62 / Chapter 5.2 --- "First Method for Mining LinkSeq, LinkApp" --- p.63 / Chapter 5.3 --- "Second Method for Mining LinkSeq, LinkTwo" --- p.64 / Chapter 5.4 --- "Alternative Method for Mining LinkSeq, LinkTree" --- p.65 / Chapter 5.4.1 --- Sequence Tree: Design --- p.65 / Chapter 5.4.2 --- Construction of seq-tree --- p.69 / Chapter 5.4.3 --- Mining LinkSeq using seq-tree --- p.76 / Chapter 5.5 --- Performance Study --- p.82 / Chapter 5.6 --- Discussions --- p.85 / Chapter 5.7 --- Summary --- p.85 / Chapter 6 --- Conclusion and Future Work --- p.87 / Chapter 6.1 --- Conclusion --- p.87 / Chapter 6.2 --- Future Work --- p.88 / Bibliography --- p.97

Thesis (PhD) -- University of South Australia, 1999

Owing to the advances of computer and communication technologies, clustering analysis on moving objects has attracted increasing attention in recent years. An interesting problem is to find the moving clusters composed of objects which move along for a sufficiently long period of time. However, a moving cluster inclines to break after some time because of the goal change in each individual object. In order to identify the set of moving clusters, we propose the formal definition of moving clusters with semantically clear parameters. Based on the definition, we propose delicate approaches to cluster moving objects. The proposed approaches are evaluated using data generated with and without underlying model. We validate our approaches with a through experimental evaluation and comparison.

No description available.

Thesis (Ph. D.)--Hong Kong University of Science and Technology, 2002. / Includes bibliographical references (leaves 208-215). Also available in electronic version. Access restricted to campus users.

Thesis (M. Phil.)--Hong Kong University of Science and Technology, 2003. / Includes bibliographical references (leaves 62-65). Also available in electronic version. Access restricted to campus users.

Pagrindinis šio darbo tikslas yra išnagrinėti judančių objektų indeksavimo duomenų bazėse problemas, siūlomus sprendimus bei palyginti keleto iš jų veiksmingumą. Įvairiais pjūviais buvo lyginami praeities duomenis indeksuojantys R ir iš jo išvesti STR bei TB medžiai. Eksperimentai atlikti naudojant sugeneruotus judančių objektų duomenis. Gauti rezultatai parodė, kad indeksų veiksmingas priklauso nuo tam tikrų sąlygų ir aplinkybių, kuriomis jie naudojami. / Over the past few years, there has been a continuous improvement in the wireless communications and the positioning technologies. As a result, tracking the changing positions of continuously moving objects is becoming increasingly feasible and necessary. Databases that deal with objects that change their location and/or shape over time are called spatio-temporal databases. Traditional database approaches for effective information retrieval cannot be used as the moving objects database is highly dynamic. A need for so called spatio-temporal indexing techniques comes to scene. Mainly, by the problem they are addressed to, indices are divided into two groups: a) indexing the past and b) indexing the current and predicted future positions. Also the have been proposed techniques covering both problems. This work is a survey for well known and used indices. Also there is a performance comparison between several past indexing methods. STR Tree, TB Tree and the predecessor of many indices, the R Tree are compared in various aspects using generated datasets of simulated objects movement.

Thesis (PhD) -- University of South Australia, 1999

JavaScript Object Notation (JSON) is a format for representing data. In this thesis we show how to capture the history of changes to a JSON document. Capturing the history is important in many applications, where not only the current version of a document is required, but all the previous versions. Conceptually the history can be thought of as a sequence of non-temporal JSON documents, one for each instant of time. Each document in the sequence is called a snapshot. Since changes to a document are few and infrequent, the sequence of snapshots largely duplicates a document across many time instants, so the snapshot model is (wildly) inefficient in terms of space needed to represent the history and time taken to navigate within it. A more efficient representation can be achieved by “gluing" the snapshots together to form a temporal model. Data that remains unchanged across snapshots is represented only once in a temporal model. But we show that the temporal model is not a JSON document, and it is important to represent a history as JSON to ensure compatibility with web services and scripting languages that use JSON. So we describe a representational model that captures the information in a temporal model. We implement the representational model in Python and extensively experiment with the model. Our experiments show that the model is efficient.

published_or_final_version / abstract / toc / Computer Science and Information Systems / Master / Master of Philosophy