
S.E.A.L. %2D A Query Language for Entity%2DAssociation Queries.

Towards Automating Class%2DSplitting Using Betweenness Clustering.

Vietnamese Document Representation and Classification.

Text Categorization for Vietnamese Documents.

An analysis of multi%2Dsampled issue and no%2Dreplacement tournament selection.
This thesis presents an analysis of the selection process in tree-based Genetic Programming (GP), covering the optimisation of both parent and offspring selection, and provides a detailed understanding of selection and guidance on how to improve GP search effectively and efficiently. The first part of the thesis provides models and visualisations to analyse selection behaviour in standard tournament selection, clarifies several issues in standard tournament selection, and presents a novel solution to automatically and dynamically optimise parent selection pressure. The fitness evaluation cost of parent selection is then addressed and some cost-saving algorithms introduced. In addition, the feasibility of using good predecessor programs to increase parent selection efficiency is analysed. The second part of the thesis analyses the impact of offspring selection pressure on the overall GP search performance. The fitness evaluation cost of offspring selection is then addressed, with investigation of some heuristics to efficiently locate good offspring by constraining crossover point selection structurally
An analysis of the distribution of swapped subtree sizes in tree%2Dbased genetic programming.
This thesis presents an analysis of the selection process in tree-based Genetic Programming (GP), covering the optimisation of both parent and offspring selection, and provides a detailed understanding of selection and guidance on how to improve GP search effectively and efficiently. The first part of the thesis provides models and visualisations to analyse selection behaviour in standard tournament selection, clarifies several issues in standard tournament selection, and presents a novel solution to automatically and dynamically optimise parent selection pressure. The fitness evaluation cost of parent selection is then addressed and some cost-saving algorithms introduced. In addition, the feasibility of using good predecessor programs to increase parent selection efficiency is analysed. The second part of the thesis analyses the impact of offspring selection pressure on the overall GP search performance. The fitness evaluation cost of offspring selection is then addressed, with investigation of some heuristics to efficiently locate good offspring by constraining crossover point selection structurally
Is the not%2Dsampled issue in tournament selection critical%3F
This thesis presents an analysis of the selection process in tree-based Genetic Programming (GP), covering the optimisation of both parent and offspring selection, and provides a detailed understanding of selection and guidance on how to improve GP search effectively and efficiently. The first part of the thesis provides models and visualisations to analyse selection behaviour in standard tournament selection, clarifies several issues in standard tournament selection, and presents a novel solution to automatically and dynamically optimise parent selection pressure. The fitness evaluation cost of parent selection is then addressed and some cost-saving algorithms introduced. In addition, the feasibility of using good predecessor programs to increase parent selection efficiency is analysed. The second part of the thesis analyses the impact of offspring selection pressure on the overall GP search performance. The fitness evaluation cost of offspring selection is then addressed, with investigation of some heuristics to efficiently locate good offspring by constraining crossover point selection structurally
Genetic Programming for Feature Ranking in Classification Problems.

Iterative Neighbourhood Similarity Computation for Collaborative Filtering.

Genetic Programming for detecting rhythmic stress in spoken English.
This thesis presents an analysis of the selection process in tree-based Genetic Programming (GP), covering the optimisation of both parent and offspring selection, and provides a detailed understanding of selection and guidance on how to improve GP search effectively and efficiently. The first part of the thesis provides models and visualisations to analyse selection behaviour in standard tournament selection, clarifies several issues in standard tournament selection, and presents a novel solution to automatically and dynamically optimise parent selection pressure. The fitness evaluation cost of parent selection is then addressed and some cost-saving algorithms introduced. In addition, the feasibility of using good predecessor programs to increase parent selection efficiency is analysed. The second part of the thesis analyses the impact of offspring selection pressure on the overall GP search performance. The fitness evaluation cost of offspring selection is then addressed, with investigation of some heuristics to efficiently locate good offspring by constraining crossover point selection structurally
Empirical Analysis of GP Tree%2DFragments.

Another investigation on tournament selection modelling and visualisation.
Tournament selection has been widely used and studied in evolutionary algorithms. To supplement the study of tournament selection, this paper provides several models describing the probabilities that a program of a particular rank is sampled and is selected in the standard tournament selection in a simple situation and a complex situation. This paper discovers that, with the same tournament size, trends of sampling probability of a program and selection probability distributions of a population are the same regardless of the population size. This paper also models and investigates an alternative tournament selection method which eliminates one of the drawbacks in the standard tournament selection. Finally, this paper proposes a new fitness evaluation saving algorithm via the use of not-sampled individuals, which is a special property of tournament selection.
Another investigation on tournament selection modelling and visualisation.
Tournament selection has been widely used and studied in evolutionary algorithms. To supplement the study of tournament selection, this paper provides several models describing the probabilities that a program of a particular rank is sampled and is selected in the standard tournament selection in a simple situation and a complex situation. This paper discovers that, with the same tournament size, trends of sampling probability of a program and selection probability distributions of a population are the same regardless of the population size. This paper also models and investigates an alternative tournament selection method which eliminates one of the drawbacks in the standard tournament selection. Finally, this paper proposes a new fitness evaluation saving algorithm via the use of not-sampled individuals, which is a special property of tournament selection.


Genetic Programming for New Zealand CPI Inflation Prediction.
This thesis presents an analysis of the selection process in tree-based Genetic Programming (GP), covering the optimisation of both parent and offspring selection, and provides a detailed understanding of selection and guidance on how to improve GP search effectively and efficiently. The first part of the thesis provides models and visualisations to analyse selection behaviour in standard tournament selection, clarifies several issues in standard tournament selection, and presents a novel solution to automatically and dynamically optimise parent selection pressure. The fitness evaluation cost of parent selection is then addressed and some cost-saving algorithms introduced. In addition, the feasibility of using good predecessor programs to increase parent selection efficiency is analysed. The second part of the thesis analyses the impact of offspring selection pressure on the overall GP search performance. The fitness evaluation cost of offspring selection is then addressed, with investigation of some heuristics to efficiently locate good offspring by constraining crossover point selection structurally
An analysis of depth of crossover points in tree%2Dbased Genetic Programming.
This thesis presents an analysis of the selection process in tree-based Genetic Programming (GP), covering the optimisation of both parent and offspring selection, and provides a detailed understanding of selection and guidance on how to improve GP search effectively and efficiently. The first part of the thesis provides models and visualisations to analyse selection behaviour in standard tournament selection, clarifies several issues in standard tournament selection, and presents a novel solution to automatically and dynamically optimise parent selection pressure. The fitness evaluation cost of parent selection is then addressed and some cost-saving algorithms introduced. In addition, the feasibility of using good predecessor programs to increase parent selection efficiency is analysed. The second part of the thesis analyses the impact of offspring selection pressure on the overall GP search performance. The fitness evaluation cost of offspring selection is then addressed, with investigation of some heuristics to efficiently locate good offspring by constraining crossover point selection structurally
Exploiting underrepresented query aspects for automatic query expansion.

QC4 %2D A Clustering Evaluation Method.

Understanding Query Aspects with applications to Interactive Query Expansion.

Genetic Programming for Automatic Stress Detection in Spoken English.
This thesis presents an analysis of the selection process in tree-based Genetic Programming (GP), covering the optimisation of both parent and offspring selection, and provides a detailed understanding of selection and guidance on how to improve GP search effectively and efficiently. The first part of the thesis provides models and visualisations to analyse selection behaviour in standard tournament selection, clarifies several issues in standard tournament selection, and presents a novel solution to automatically and dynamically optimise parent selection pressure. The fitness evaluation cost of parent selection is then addressed and some cost-saving algorithms introduced. In addition, the feasibility of using good predecessor programs to increase parent selection efficiency is analysed. The second part of the thesis analyses the impact of offspring selection pressure on the overall GP search performance. The fitness evaluation cost of offspring selection is then addressed, with investigation of some heuristics to efficiently locate good offspring by constraining crossover point selection structurally
Population Clustering in Genetic Programming.
This thesis presents an analysis of the selection process in tree-based Genetic Programming (GP), covering the optimisation of both parent and offspring selection, and provides a detailed understanding of selection and guidance on how to improve GP search effectively and efficiently. The first part of the thesis provides models and visualisations to analyse selection behaviour in standard tournament selection, clarifies several issues in standard tournament selection, and presents a novel solution to automatically and dynamically optimise parent selection pressure. The fitness evaluation cost of parent selection is then addressed and some cost-saving algorithms introduced. In addition, the feasibility of using good predecessor programs to increase parent selection efficiency is analysed. The second part of the thesis analyses the impact of offspring selection pressure on the overall GP search performance. The fitness evaluation cost of offspring selection is then addressed, with investigation of some heuristics to efficiently locate good offspring by constraining crossover point selection structurally
Genetic Programming for Automatic Stress Detection in Spoken English.
This thesis presents an analysis of the selection process in tree-based Genetic Programming (GP), covering the optimisation of both parent and offspring selection, and provides a detailed understanding of selection and guidance on how to improve GP search effectively and efficiently. The first part of the thesis provides models and visualisations to analyse selection behaviour in standard tournament selection, clarifies several issues in standard tournament selection, and presents a novel solution to automatically and dynamically optimise parent selection pressure. The fitness evaluation cost of parent selection is then addressed and some cost-saving algorithms introduced. In addition, the feasibility of using good predecessor programs to increase parent selection efficiency is analysed. The second part of the thesis analyses the impact of offspring selection pressure on the overall GP search performance. The fitness evaluation cost of offspring selection is then addressed, with investigation of some heuristics to efficiently locate good offspring by constraining crossover point selection structurally
Automatic Selection Pressure Control in Genetic Programming.
This thesis presents an analysis of the selection process in tree-based Genetic Programming (GP), covering the optimisation of both parent and offspring selection, and provides a detailed understanding of selection and guidance on how to improve GP search effectively and efficiently. The first part of the thesis provides models and visualisations to analyse selection behaviour in standard tournament selection, clarifies several issues in standard tournament selection, and presents a novel solution to automatically and dynamically optimise parent selection pressure. The fitness evaluation cost of parent selection is then addressed and some cost-saving algorithms introduced. In addition, the feasibility of using good predecessor programs to increase parent selection efficiency is analysed. The second part of the thesis analyses the impact of offspring selection pressure on the overall GP search performance. The fitness evaluation cost of offspring selection is then addressed, with investigation of some heuristics to efficiently locate good offspring by constraining crossover point selection structurally
Query Directed Web Page Clustering.

Improving Web Clustering by Cluster Selection.

Standardized Evaluation Method for Web Clustering Results.

Learning Models for English Speech Recognition.

Detecting Stress in Spoken English using Decision Trees and Support Vector Machines.
This paper describes an approach to the detection of stress in spoken New Zealand English. After identifying the vowel segments of the speech signal, the approach extracts two different sets of features --- prosodic features and vowel quality features --- from the vowel segments. These features are then normalised and scaled to obtain speaker independent feature values that can be used to classify each vowel segment as stressed or unstressed. We used Decision Trees (C4.5) and Support Vector Machines (LIBSVM) to learn stress-detecting classifiers with various combinations of the features. The approach was evaluated on 60 adult female utterances with 703 vowels and a maximum accuracy of 84.72% was achieved. The results showed that a combination of features derived from duration and amplitude achieved the best performance but the vowel quality features also achieved quite reasonable results.
Approximately Repetitive Structure Detection for Wrapper Induction.
In recent years, much work has been invested into automatically learning wrappers for information extraction from HTML tables and lists. Our research has focused on a system that can learn a wrapper from a single unlabelled page. An essential step is to locate the tabular data within the page. This is not trivial when the structures of data tuples are similar but not identical. In this paper we describe an algorithm that can automatically detect approximate repetitive structures within one sequence. The algorithm does not rely on any domain knowledge or HTML heuristics and it can be used in detecting repetitive patterns and hence to learn wrappers from a single unlabeled tabular page. Author Information Xiaoying Gao and Peter Andreae are academic staff members in computer science. Richard
A Two Phase Genetic Programming Approach to Object Detection.

Approximately Repetitive Structure Detection for Wrapper Induction.
In recent years, much work has been invested into automatically learning wrappers for information extraction from HTML tables and lists. Our research has focused on a system that can learn a wrapper from a single unlabelled page. An essential step is to locate the tabular data within the page. This is not trivial when the structures of data tuples are similar but not identical. In this paper we describe an algorithm that can automatically detect approximate repetitive structures within one sequence. The algorithm does not rely on any domain knowledge or HTML heuristics and it can be used in detecting repetitive patterns and hence to learn wrappers from a single unlabeled tabular page. Author Information Xiaoying Gao and Peter Andreae are academic staff members in computer science. Richard
Automatic Pattern Construction For Web Information Extraction.

Pixel Statistics and False Alarm Area in Genetic Programming for Object Detection.

Learning Information Extraction Patterns from Tabular Web Pages without Manual Labelling.

Interpreting Microarray Expression Data Using Text Annotating the Genes.
MicqSqSq  expression data is being generated by the gigabyte all over the world with 13 undoubted  exponentialinconent  tocS8q  Annotatedgenomic  data is also rapidly 14 pouring  intopublic  databases. Our goal is to develop automated ways  ofcSW)T8qS  15 these twosourc( of information toproduc insight into the operation ofcSW) under 16  variousciousS:(9:  Ourapproac is to  usemac(q((S58q(9qS tecSc(S5  to identify 17  cScqSqSqScc  of genes that are up-regulated or down-regulated in  apartic57: micti  18 array experiment. We seek models that are  (a)acS8(#97  (b) easy to interpret, and (c  19 stable to small variations in the training data. This paper explores  thee#ecq9qS57T  20 two standard  macrdS77T:)S579  algorithms for this task: Na# ve Bayes (based on prob- 21 ability) and PFOIL (based on building rules). Although we do  notantic)TS5  using our 22 learned models topredic expression levels of genes, wecS8 the task in  apredic57W  23 framework, and evaluate the quality of the models in terms of  theirpredic8:(  power on 24 genes held out from the training. The paper reports on experiments using acngS E. coli 25  micqSqSq data,disc7W8q8  the strengths and weaknesses of the two algorithms and 26 demonstrating the trade-o#s betweenacweenS8  cweenS8q:##S5:)q8  and stability.
Interpreting microarray expression data using text annotating the genes.
MicqSqSq  expression data is being generated by the gigabyte all over the world with 13 undoubted  exponentialinconent  tocS8q  Annotatedgenomic  data is also rapidly 14 pouring  intopublic  databases. Our goal is to develop automated ways  ofcSW)T8qS  15 these twosourc( of information toproduc insight into the operation ofcSW) under 16  variousciousS:(9:  Ourapproac is to  usemac(q((S58q(9qS tecSc(S5  to identify 17  cScqSqSqScc  of genes that are up-regulated or down-regulated in  apartic57: micti  18 array experiment. We seek models that are  (a)acS8(#97  (b) easy to interpret, and (c  19 stable to small variations in the training data. This paper explores  thee#ecq9qS57T  20 two standard  macrdS77T:)S579  algorithms for this task: Na# ve Bayes (based on prob- 21 ability) and PFOIL (based on building rules). Although we do  notantic)TS5  using our 22 learned models topredic expression levels of genes, wecS8 the task in  apredic57W  23 framework, and evaluate the quality of the models in terms of  theirpredic8:(  power on 24 genes held out from the training. The paper reports on experiments using acngS E. coli 25  micqSqSq data,disc7W8q8  the strengths and weaknesses of the two algorithms and 26 demonstrating the trade-o#s betweenacweenS8  cweenS8q:##S5:)q8  and stability.


Reconstruction of Vascular Networks Using Three%2DDimensional Models.

Reconstruction of Blood Vessel Networks from X%2DRay Projections and a Vascular Catalogue.

Reconstruction of blood vessel networks from a few perspective projections.

Information Transfer and Area%2DTime Tradeoffs for VLSI Multiplication.

