LOGIN TO YOUR ACCOUNT

Username
Password
Remember Me
Or use your Academic/Social account:

Congratulations!

You have just completed your registration at OpenAire.

Before you can login to the site, you will need to activate your account. An e-mail will be sent to you with the proper instructions.

Important!

Please note that this site is currently undergoing Beta testing.
Any new content you create is not guaranteed to be present to the final version of the site upon release.

Thank you for your patience,
OpenAire Dev Team.

Close This Message

CREATE AN ACCOUNT

Name:
Username:
Password:
Verify Password:
E-mail:
Verify E-mail:
*All Fields Are Required.
Please Verify You Are Human:

OpenAIRE is about to release its new face with lots of new content and services.
During September, you may notice downtime in services, while some functionalities (e.g. user registration, login, validation, claiming) will be temporarily disabled.
We apologize for the inconvenience, please stay tuned!
For further information please contact helpdesk[at]openaire.eu

fbtwitterlinkedinvimeoflicker grey 14rssslideshare1
Wang, Lizhen
Languages: English
Types: Doctoral thesis
Subjects: Q1, QA75
The technical progress in computerized spatial data acquisition and storage results\ud in the growth of vast spatial databases. Faced with large amounts of increasing spatial\ud data, a terminal user has more difficulty in understanding them without the helpful\ud knowledge from spatial databases. Thus, spatial data mining has been brought under\ud the umbrella of data mining and is attracting more attention.\ud Spatial data mining presents challenges. Differing from usual data, spatial data includes\ud not only positional data and attribute data, but also spatial relationships among\ud spatial events. Further, the instances of spatial events are embedded in a continuous\ud space and share a variety of spatial relationships, so the mining of spatial patterns demands\ud new techniques.\ud In this thesis, several contributions were made. Some new techniques were proposed,\ud i.e., fuzzy co-location mining, CPI-tree (Co-location Pattern Instance Tree),\ud maximal co-location patterns mining, AOI-ags (Attribute-Oriented Induction based on Attributes’\ud Generalization Sequences), and fuzzy association prediction. Three algorithms\ud were put forward on co-location patterns mining: the fuzzy co-location mining algorithm,\ud the CPI-tree based co-location mining algorithm (CPI-tree algorithm) and the orderclique-\ud based maximal prevalence co-location mining algorithm (order-clique-based algorithm).\ud An attribute-oriented induction algorithm based on attributes’ generalization sequences\ud (AOI-ags algorithm) is further given, which unified the attribute thresholds and\ud the tuple thresholds. On the two real-world databases with time-series data, a fuzzy association\ud prediction algorithm is designed. Also a cell-based spatial object fusion algorithm\ud is proposed. Two fuzzy clustering methods using domain knowledge were proposed:\ud Natural Method and Graph-Based Method, both of which were controlled by a\ud threshold. The threshold was confirmed by polynomial regression. Finally, a prototype\ud system on spatial co-location patterns’ mining was developed, and shows the relative\ud efficiencies of the co-location techniques proposed\ud The techniques presented in the thesis focus on improving the feasibility, usefulness,\ud effectiveness, and scalability of related algorithm. In the design of fuzzy co-location\ud Abstract\ud mining algorithm, a new data structure, the binary partition tree, used to improve the\ud process of fuzzy equivalence partitioning, was proposed. A prefix-based approach to\ud partition the prevalent event set search space into subsets, where each sub-problem can\ud be solved in main-memory, was also presented. The scalability of CPI-tree algorithm is\ud guaranteed since it does not require expensive spatial joins or instance joins for identifying\ud co-location table instances. In the order-clique-based algorithm, the co-location table\ud instances do not need be stored after computing the Pi value of corresponding colocation,\ud which dramatically reduces the executive time and space of mining maximal colocations.\ud Some technologies, for example, partitions, equivalence partition trees, prune\ud optimization strategies and interestingness, were used to improve the efficiency of the\ud AOI-ags algorithm. To implement the fuzzy association prediction algorithm, the “growing\ud window” and the proximity computation pruning were introduced to reduce both I/O and\ud CPU costs in computing the fuzzy semantic proximity between time-series.\ud For new techniques and algorithms, theoretical analysis and experimental results\ud on synthetic data sets and real-world datasets were presented and discussed in the thesis.
  • The results below are discovered through our pilot algorithms. Let us know how we are doing!

    • 2.3.4 Generating Co-location Rules…………………………………………………30 2.4 Analysis for Discovering Fuzzy Co-location………………………………………..30
    • 2.4.1 Completeness and Correctness……………………………………………30
    • 2.4.2 Computational Complexity Analysis…………………………………...31 2.5 Experimental Evaluation………………………………………………………………32
    • 2.5.1 Performance Study……………………………………………………………..33
    • 2.5.2 Experiments on a Real Data Set………………………………………………35 2.6 Summary……………………………………………………………………………….39 Chapter 3. A New Join-less Approach for Identifying Co-location Pattern Table Instances……………………………………………………………………………………….....40
    • 3.1 Overview………………………………………………………………………………..40
    • 3.1.1 Basic Concepts………………………………………………………………….41
    • 3.1.2 Problem Definition………………………………………………………………42
    • 3.1.3 Background for Mining Co-location Patterns………………………………...43
    • 3.1.4 Motivation……….. ……………………………………………………………...44
    • 3.1.5 Organization of the Chapter……………………………………………………45
    • 3.2 Co-location-Pattern Tree (CPI-tree): Design and Construction…………………..45
    • 3.2.1 CPI-tree…………………………………………………………………………..45
    • 3.2.2 Complexity and Completeness of CPI-tree…………………………………..48
    • 3.3 Generating Complete Table-Instance Using CPI-tree ........................................48
    • 3.3.1 Principles of Table-Instance Generation from a CPI-tree…………………..49
    • 3.3.2 Table-Instance Generation Algorithm………………………………………...50
    • 3.4 Some Optimization Strategies………………………………………………………..52
    • 3.4.1 Pruning Strategies………………………………………………………………52
    • 3.4.2 Optimization by Reducing the Depth of CPI-tree……………………………53
    • 3.5 Experimental Results………………………………………………………………….55
    • 3.6 Summary……………………………………………………………………………….57
    • 4.3.2 Algorithms………………………………………………………………………..70 4.4 Algorithm and Analysis for Mining Maximal Ordered Prevalence Co-locations...71
    • 4.4.1 Algorithms………………………………………………………………………..72
    • 4.4.2 Analysis…………………………………………………………………………..73 4.5 Performance Study……………………………………………………………………74 4.6 Summary……………………………………………………………………………….78
    • 5.3.3 Equivalence Partition Trees and Calculatingπ Ai,gi ………………….…...…87
    • 5.3.4 Algorithms……………………………………………………………….……….88 5.4 Interestingness of Attributes' Generalization Sequences…………………………90 5.5 Analysis…………………………………………………………………………………91
    • 5.5.1 Completeness and Correctness……………………………………………….91
    • 5.5.2 Computational Complexity……………………………………………………..92 5.6 Performance Evaluation and Applications………………………………………….92
    • 5.6.1 Evaluation Using Synthetic Datasets…………………………………………92
    • 5.6.2 Applications in a Real Dataset………………………………………………...93 5.7 Summary……………………………………………………………………………….94 Chapter 6. Fuzzy Data Mining Prediction Technologies and Applications……………….96
    • 6.1 Overview………………………………………………………………………………..96
    • 6.2 Preparing the Data for Prediction…………………………………………………....97
    • 6.2.1 Preparing the Data for Predicting the Shovel Cable Lifespan……………..97
    • 6.2.2 Preparing the Data for Predicting Plant Species in an Ecological Environ-
    • ment ………………………………….... …………………………………........ ………98
    • 6.3 Initial Data Exploration - IDE…………………………………................................99
    • 6.3.1 Comparing the Similarity of Two Time-Series……………………………….99
    • 6.3.2 Fuzzy Equivalence Partition for the Set of Time-Series…………………..101
    • 6.3.3 An Example………………………………….... ……………………………...102
    • 6.4 Mining Prediction………………………………….... ………………………………103
    • 6.4.1 Degree of Fuzzy Association………………………………….....................103
    • 6.4.2 Superposition of the Degrees of Fuzzy Association……………………….105
    • 6.4.3 An Example…………………………………................................................106 6.5 Algorithms………………………………….... ………………………………….......108
    • 6.5.1 IDE Algorithm………………………………….... ……………………………108
    • 6.5.2 Mining Prediction Algorithm………………………………….......................109
    • 6.5.3 Analysis of Algorithm Complexity…………………………………...............110 6.6 Results of Experiments………………………………….......................................111
    • 6.6.1 Estimating Error Rates…………………………………...............................112
    • 6.6.2 Quality………………………………….... …………………………………....113
    • 6.6.3 Performance of the Algorithm…………………………………....................113 6.7 Summary………………………………….... ………………………………….........114 Chapter 7. A Cell-Based Spatial Object Fusion Method………………………………….115
    • 7.1 Overview………………………………….... ………………………………….........115
    • 7.2 Basic Definitions and Measurements…………………………………..................116
    • 7.3 A Cell-Based Method Finding Fusion Seta………………………………….........117
    • 7.3.1 The Method………………………………….... ………………………………118
    • 7.3.2 The Algorithm………………………………….... ……………………………121
    • 7.3.3 Complexity Analysis………………………………….... …………………….121
    • 7.4 Testing the Method………………………………….... …………………………….122
    • 7.5 Summary………………………………….... ………………………………….........125 Chapter 8. A Fuzzy Clustering Method Based on Domain Knowledge…………………127
    • 8.1 Overview………………………………….... ………………………………….........127
    • 8.2 Basic Concepts and Methods………………………………….... ………………..128
    • 8.2.1 Basic Concepts ………………………………….........................................128
    • 8.2.2 Fuzzy Clustering Using Matrix Method ………………………………….... 130
    • 8.3 New Algorithms for Fuzzy Clustering…………………………………..................131
    • 8.3.1 Natural Method (NM) …………………………………................................131
    • 8.3.2 Graph-based Method (GBM) ………………………………….... …….……132
    • 8.3.3 Confirming the Threshold λ …………………………………......................133
    • 8.4 Algorithm analysis………………………………….... ……………………………..134
    • 8.4.1 Correctness Analysis………………………………….... …………………...134
    • 8.4.2 Time Complexity………………………………….... ………………………..135
    • 8.5 Experiments………………………………….... …………………………………....135
    • 8.5.1 Evaluation Using Synthetic Datasets………………………………….........136
    • 8.5.2 Evaluation Using Real Datasets…………………………………................137 8.6 Summary………………………………….... ………………………………….........138 Chapter 9. A Visual Spatial Co-location Patterns' Mining Prototype System (SCPMiner)………………. ………………………………….... …………………………….139
    • 9.1 Overview………………………………….... ………………………………….........139
    • 9.2 Analysis and Design of SCPMiner ………………………………….... …………..139
    • 9.3 Implementation of SCPMiner ………………………………………………………141
    • 9.3.1 Co-location Data Management (CDM) ……………………………………..141
    • 9.3.2 Co-location Patterns Mining (CPM) …………………………………………143
    • 9.3.3 Co-location Mining Analyzing (CMA) ……………………………………….144
    • 9.3.4 Co-location Patterns Applying (CPA) ……………………………………….146
    • 9.4 Summary……………………………………………………………………………...149 [1] Wang, L., Lu, J., Yip, J. (2007) AOG-ags Algorithms and Applications. In: Proceed-
    • plications (ADMA 2007), Springer-Verlag, Berlin, LNAI 4632, pp. 323-334,
    • 2007.8 [2] Wang, L., Lu, J., Yip, J. (2007) An Effective Approach to Predicting Plant Species in
    • an Ecological Environment, In: Proceedings of the 2007 international Confer-
    • ence on Information and Knowledge Engineering (IKE'07), Las Vegas Ne-
    • vada, USA, June 25-28, 2007, pp. 245-250 [3] Wang, L., Bao, y., Lu, J., Yip, J. (2008) A New Join-less Approach for Co-
    • location Pattern Mining, In: Proceedings of the IEEE 8th International Confer-
    • ence on Computer and Information Technology (CIT2008), Syney, Australia 8-
    • 11 July 2008 (Accepted) [4] Wang, L., Lu, J., Yip, J. (2008) An Order-Clique-Based Approach for Mining Maximal
    • Co-location Patterns, University of Huddersfield, Poster, 2008 [5] Wang, L., Lu, J., Yip, J. Discovering Co-location Patterns from Fuzzy Spatial Data
    • December, 2007. [6] Wang, L., Zhou, L., Lu, J., Yip, J. An Efficient Approach for Mining Maximal Co-
    • ences, Elsevier, April, 2008. [7] Wang, L., Bao, Y., Lu, J., Yip, J. A Visual Spatial Co-location Patterns' Mining Proto-
    • 2008), Hangzhou, China, Sept. 22-24, 2008, April, 2008.
  • No related research data.
  • No similar publications.

Share - Bookmark

Cite this article

Cookies make it easier for us to provide you with our services. With the usage of our services you permit us to use cookies.
More information Ok