Domain adaptation reuses the knowledge learned from an existing (source) domain to classify unlabelled data from another related (target) domain. However, the two domains have different data distributions. Common approaches to bridge the two distributions are selecting/reweighting instances, building domain-invariant feature subspaces, or directly building adaptive classifiers. Recent domain adaptation work has shown that combining the above first two approaches before applying the third approach achieves better performance than performing each approach individually. However, most existing instance selection approaches are based on a ranking mechanism, ignore interdependences between instances, and require a pre-defined number of selected instances. Furthermore, adaptive classifiers are sensitive to their parameters which are challenging to optimise due to the lack of target labelled instances. This paper introduces a novel evolutionary instance selection approach for domain adaptation. We propose a compacted representation and an efficient fitness function for Particle Swarm Optimisation to automatically determine the number of selected instances while considering the interdependencies among instances. This paper also proposes to use multiple partial classifiers to build a more reliable and robust adaptive classifier. The results show that evolutionary instance selection selects better instances than the ranking approach. In cooperation with multiple partial classifiers, the proposed algorithm achieves better performance than nine state-of-the-art and well-known domain adaptation approaches.

Particle swarm optimization (PSO) is a heuristic optimization algorithm generally applied to continuous domains. Binary PSO is a form of PSO applied to binary domains but uses the concepts of velocity and momentum from continuous PSO, which leads to its limited performance. In our previous work, we reformulated momentum as a stickiness property and velocity as a flipping probability to develop sticky binary PSO. The initial design provides a good base, but many key factors need to be investigated. In this article, we propose a new algorithm called dynamic sticky binary PSO by developing a dynamic parameter control strategy based on an investigation of exploration and exploitation in the binary search spaces. The proposed algorithm is compared with four state-of-the-art dynamic binary algorithms on two types of binary problems: 1) knapsack and 2) feature selection. The experimental results on the knapsack datasets show that the new velocity and momentum assist sticky binary PSO in evolving better solutions than the benchmark algorithms. On feature selection, the dynamic strategy takes the advantages of these two newly defined movement concepts to help the proposed algorithm to produce smaller feature subsets with higher classification performance. This is the first time in the binary PSO, the four important concepts, that is, velocity, momentum, exploration, and exploitation, are investigated systematically to capture the properties of the binary search spaces to evolve better solutions for binary problems.

Domain adaptation utilizes learned knowledge from an existing domain (source domain) to improve the classification performance of another related, but not identical, domain (target domain). Most existing domain adaptation methods first perform domain alignment, then apply standard classification algorithms. Transfer classifier induction is an emerging domain adaptation approach that incorporates the domain alignment into the process of building an adaptive classifier instead of using a standard classifier. Although transfer classifier induction approaches have achieved promising performance, they are mainly gradient-based approaches which can be trapped at local optima. In this article, we propose a transfer classifier induction algorithm based on evolutionary computation to address the above limitation. Specifically, a novel representation of the transfer classifier is proposed which has much lower dimensionality than the standard representation in existing transfer classifier induction approaches. We also propose a hybrid process to optimize two essential objectives in domain adaptation: 1) the manifold consistency and 2) the domain difference. Particularly, the manifold consistency is used in the main fitness function of the evolutionary search to preserve the intrinsic manifold structure of the data. The domain difference is reduced via a gradient-based local search applied to the top individuals generated by the evolutionary search. The experimental results show that the proposed algorithm can achieve better performance than seven state-of-the-art traditional domain adaptation algorithms and four state-of-the-art deep domain adaptation algorithms.

Feature selection is an important task in machine learning that has two main objectives: 1) reducing dimensionality and 2) improving learning performance. Feature selection can be considered a multiobjective problem. However, it has its problematic characteristics, such as a highly discontinuous Pareto front, imbalance preferences, and partially conflicting objectives. These characteristics are not easy for existing evolutionary multiobjective optimization (EMO) algorithms. We propose a new decomposition approach with two mechanisms (static and dynamic) based on multiple reference points under the multiobjective evolutionary algorithm based on decomposition (MOEA/D) framework to address the above-mentioned difficulties of feature selection. The static mechanism alleviates the dependence of the decomposition on the Pareto front shape and the effect of the discontinuity. The dynamic one is able to detect regions in which the objectives are mostly conflicting, and allocates more computational resources to the detected regions. In comparison with other EMO algorithms on 12 different classification datasets, the proposed decomposition approach finds more diverse feature subsets with better performance in terms of hypervolume and inverted generational distance. The dynamic mechanism successfully identifies conflicting regions and further improves the approximation quality for the Pareto fronts.

In classification, the task of domain adaptation is to learn a classifier to classify target data using unlabeled data from the target domain and labeled data from a related, but not identical, source domain. Transfer classifier induction is a common domain adaptation approach that learns an adaptive classifier directly rather than first adapting the source data. However, most existing transfer classifier induction algorithms are gradient-based, so they can easily get stuck at local optima. Moreover, they usually generate only a single classifier which might fit the source data too well, which results in poor target accuracy. In this paper, we propose a population-based algorithm that can address the above two limitations. The proposed algorithm can re-initialize a population member to a promising region when the member is trapped at local optima. The population-based mechanism allows the proposed algorithm to output a set of classifiers which is more reliable than a single classifier. The experimental results show that the proposed algorithm achieves significantly better target accuracy than four state-of-the-art and well-known domain adaptation algorithms on three real-world domain adaptation problems.

Event detecting Wireless Sensor Networks (WSNs) with overlapping sensor nodes generate many redundant packets. Resource constraints in WSNs require minimizing the number of packets transmitted, so we want to eliminate all redundant packets. But this must be done carefully if we need to ensure reliable detection of all events happening across the network. Ensuring reliable transport of data is an important concern in resource-constrained WSNs. In a densely deployed sensor network, the sensing region of the sensor nodes often overlap with one another. Thus, data from sensor nodes that are in close proximity tend to be highly correlated. This introduces the concept of event reliability, where a reliable transfer of event data from each sensing region in a sensor network is sufficient. This paper proposes a distributed approach, i.e. Enhanced Event Reliability Protocol (EERP), that enables reliable transmission of event information to the sink node while minimizing redundant packets from nodes in close vicinity of one another. EERP builds on the spatio-temporal information of the originating nodes and employs an efficient combination of a greedy strategy with the use of multilateration algorithm in a unique way to perform event identification before reliably transmitting the event information towards the sink node. The performance of EERP is evaluated and compared against a base model, ERP and a commonly used scheme (SWIA). Our results show that EERP significantly improves event information delivery and energy efficiency while maintaining good coverage of events throughout the network.

Transfer learning aims to use acquired knowledge from existing (source) domains to improve learning performance on a different but similar (target) domains. Feature-based transfer learning builds a common feature space, which can minimize differences between source and target domains. However, most existing feature-based approaches usually build a common feature space with certain assumptions about the differences between domains. The number of common features needs to be predefined. In this work, we propose a new feature-based transfer learning method using particle swarm optimization (PSO), where a new fitness function is developed to guide PSO to automatically select a number of original features and shift source and target domains to be closer. Classification performance is used in the proposed fitness function to maintain the discriminative ability of selected features in both domains. The use of classification accuracy leads to a minimum number of model assumptions. The proposed algorithm is compared with four state-of-the-art feature-based transfer learning approaches on three well-known real-world problems. The results show that the proposed algorithm is able to extract less than half of the original features with better performance than using all features and outperforms the four benchmark semi-supervised and unsupervised algorithms. This is the first time Evolutionary Computation, especially PSO, is utilized to achieve feature selection for transfer learning.

Wireless mesh networks (WMNs) cannot utilize the full capability of these radio technologies without an effective channel assignment algorithm. Existing metrics for characterising the efficacy of channel assignment algorithms poorly reflect the goodput distribution in WMNs. In this paper, we develop a fitness evaluation algorithm that feeds back several indicators of efficacy to a channel assignment algorithm and evaluate the resulting goodput distribution. To the best of our knowledge, our algorithm is the first to evaluate the fitness of channel assignment algorithms with goodput distribution, average goodput, and fairness index. Simulation results show that our evaluation algorithm provides accurate prediction of goodput distribution, average goodput, and fairness index with errors below 8%.

Feature selection is an important task in machine learning, which aims to reduce the dataset dimensionality while at least maintaining the classification performance. Particle Swarm Optimisation (PSO) has been widely applied to feature selection because of its effectiveness and efficiency. However, since feature selection is a challenging task with a complex search space, PSO easily gets stuck at local optima. This paper aims to improve the PSO's searching ability by applying genetic operators such as crossover and mutation to assist the swarm to explore the search space better. The proposed genetic operators are specifically designed for feature selection, which not only improve the quality of current feature subsets but also make the search smoother. The proposed algorithm, called CMPSO, is tested and compared with three recent PSO based feature selection algorithms. Experimental results on eight datasets show that CMPSO can adapt with different numbers of features to evolve small feature subsets, which achieve similar or better classification performance than using all features and the three PSO based algorithms. The analysis on evolutionary processes shows that genetic operators assist CMPSO to evolve better solutions than the original PSO.

Many industrial and research datasets suffer from an unavoidable issue of missing values. One of the most common approaches to solving classification with incomplete data is to use an imputation method to fill missing values with plausible values before applying classification algorithms. Multiple imputation is a powerful approach to estimating missing values, but it is very expensive to use multiple imputation to estimate missing values for a single instance that needs to be classified. Genetic programming (GP) has been widely used to construct classifiers for complete data, but it seldom has been used for incomplete data. This paper proposes an approach to combining multiple imputation and GP to evolve classifiers for incomplete data. The proposed method uses multiple imputation to provide a high quality training data. It also searches for common patterns of missing values, and uses GP to build a classifier for each pattern of missing values. Therefore, the proposed method generates a set of classifiers that can be used to directly classify any new incomplete instance without requiring imputation. Experimental results show that the proposed method not only can be faster than other common methods for classification with incomplete data but also can achieve better classification accuracy.

Missing values are an unavoidable problem in many real-world datasets. Dealing with incomplete data is an crucial requirement for classification because inadequate treatment of missing values often causes large classification error. Feature construction has been successfully applied to improve classification with complete data, but it has been seldom applied to incomplete data. Genetic programming-based multiple feature construction (GPMFC) is a current encouraging feature construction method which uses genetic programming to evolve new multiple features from original features for classification tasks. GPMFC can improve the accuracy and reduce the complexity of many decision trees and rule-based classifiers; however, it cannot directly work with incomplete data. This paper proposes IGPMFC which is extended from GPMFC to tackle with incomplete data. IGPMFC uses genetic programming with interval functions to directly evolve multiple features for classification with incomplete data. Experimental results reveal that not only IGPMFC can substantially improve the accuracy, but also can reduce the complexity of learnt classifiers facing with incomplete data.

Missing values are a common issue in many industrial and real-world datasets. Coping with datasets containing missing values is an important requirement for classification because inadequate treatment of missing values may result in large errors on classification. Genetic programming (GP) has been successfully used to evolve classifiers, but it has been applied mainly to complete data. This paper proposes IGP, a GP method for directly evolving classifiers for missing data. In order to directly evolve classifiers for missing data, IGP uses interval functions as the GP function set and builds a set of classifiers for each classification problem. Experiments on 10 benchmark datasets compared IGP with five other classifiers on classification performance. Experimental results showed that, in most cases, IGP achieves significantly better classification accuracy than the other methods.

We developed a mobile learning application grounded on Transactional Distance Theory. The aim is to engage learners and decrease their feelings of isolation and emptiness when learners and instructor are physically separated. This study was launched in an effort to understand our target learners and provide an indication towards the practicality, possibility, and appropriateness of such theory-based design. The application provides text, video, and recorded audio as media, and includes chat function, game-based learning, and electronic assignment. This paper explores the method and findings of a survey study targeted at first year Computer Engineering and Computer Science student learners at the Victoria University of Wellington, New Zealand. Our survey results indicated that the learners had a positive attitude towards mobile learning, and they had a lot of experience using the provided media and functions. The theoretical-design was deemed to be practically appropriate for our learners. However, more encouragement and promotions would be needed in order to increase the application’s usage and recognition. We performed statistical analysis on the results and clustered the responses to form a persona which will be used in the next stage of this application’s development process.

An opportunity arose to secure a share of a multimillion dollar initiative to train additional ICT Postgraduate students. However, local and national ICT industries did not want such students, with the greatest demand being for conventional 4 year Bachelor of Engineering or 3 year Bachelor of Science graduates. This paper outlines a strategy that developed a suite of programmes that passed nine quality assurance stages and won a competitive bid to be selected as one of three preferred suppliers. Further, the programmes so developed addressed industry needs, and were able to target both domestic and international students.

The advance of mobile technology has transformed our society and altered many facets of our daily lives. Furthermore, its use within the context of education are becoming more critical as more educational institutions deliver their courses through the use of mobile technology and engage with their students via distance learning tools. However, mobile learning applications only have a few design guidelines that are grounded in learning principle to ensure their effectiveness. We applied principles from the Transactional Distance Theory (TDT) to formulate design guidelines specifically for designing effective mobile learning applications for computer science students. To evaluate the appropriateness and effectiveness of these guidelines in a mobile learning context, we developed application prototypes that teach computer related topics based on these guidelines, which will then be evaluated for their effectiveness on student engagement.

This paper introduces an aspect based model for analysing search queries, where queries are represented as aspects or concepts instead of “bag of words” or strings. A search query may consist of multiple aspects, while some aspects are covered well in the search results and some others are underrepresented. We think the underrepresented aspects are the main reason for the retrieval of irrelevant documents. This paper introduces novel algorithms that identify query aspects and identify underrepresented aspects. This model has many applications and this paper focuses on three of them: query difficulty prediction, query expansion and interactive query expansion. The main idea is that a hard query is a query with multiple aspects and some of the aspects are underrepresented, and a query can be improved or expanded by adding terms that are semantically related to the underrepresented aspects. Our experiments show that our aspect based methods significantly outperform existing methods.

Missing values are a common problem in many real world databases. A common way to cope with this problem is to use imputation methods to fill missing values with plausible values. Genetic programming-based multiple feature construction (GPMFC) is a filter approach to multiple feature construction for classifiers using Genetic programming. The GPMFC algorithm has been demonstrated to improve classification performance in decision tree and rule-based classifiers for complete data, but it has not been tested on imputed data. This paper studies the effect of GPMFC on classification accuracy with imputed data and how the choice of different imputation methods (mean imputation, hot deck imputation, Knn imputation, EM imputation and MICE imputation) affects classifiers using constructed features. Results show that GPMFC improves classification performance for datasets with a small amount of missing values. The combination of GPMFC and MICE imputation, in most cases, enhances classification performance for datasets with varying amounts of missing values and obtains the best classification accuracy.

Clustering ensembles is a clustering technique which derives a better clustering solution from a set of candidate clustering solutions. Clustering ensemble methods have to address two distinct but interlinked problems: Generating multiple candidate solutions from the data and producing a final clustering solution. Our recently proposed clustering ensembles method (MMOEA) based on NSGA-II used multiple views to address the first problem and a novel cluster oriented approach to address the second problem. MMOEA used a simple crossover method to explore the search space and three objective functions to determine the quality of a candidate clustering solution. The use of a simple crossover method led to slow convergence and using three objectives in NSGA-II framework is often discouraged. This paper presents a new clustering ensemble method, which introduces new ideas for crossover, mutation, tuning steps and two objective functions (instead of three) in an evolutionary process. The results show that our new method outperforms recent methods for clustering ensembles on different multi-view datasets.

Clustering is one of the fundamental data analysis techniques, which aims to find distinct groups of similar objects and discovers hidden structures in data. A recent clustering approach, clustering ensembles tries to derive an improved clustering solution based on previously generated different candidate clustering solutions. Clustering ensembles have two steps: generating multiple candidate clustering solutions from the data and forming a final clustering solution from previously generated candidate clustering solutions. A problem of the first step is the text representation, where word frequencies are often used as features. Other semantic information of the text such as topics, hypertext, etc are ignored. The problem for the second step is that the current popular median partition approach selects one clustering solution from previously generated candidate clustering solutions. A common clustering ensemble approach uses word frequencies as features to represent text data (documents). However, documents usually contain semantically rich information i.e. words, hypertext, titles, topics etc. The cluster ensemble approach ignores the semantic information of the documents and hence is prone to produce futile groupings of the documents. In this research work, we present a new multi-objective clustering ensemble method based on Strength Pareto Evolutionary Algorithm (SPEA-II). Our method utilizes the semantic information (rich features) to address the first problem of clustering ensembles. The cluster oriented evolutionary approach which derives the final clustering solution by selecting better quality clusters is in the second step of our method to address the second problem. The results show that our new method provides better results than other clustering ensemble methods.

Missing values are a common problem in many real world databases. Inadequate handing of missing data can lead to serious problems in data analysis. A common way to cope with this problem is to use imputation methods to fill missing values with plausible values. This paper proposes GPMI, a multiple imputation method that uses genetic programming as a regression method to estimate missing values. Experiments on eight datasets with six levels of missing values compare GPMI with seven other popular and advanced imputation methods on two measures: the prediction accuracy and the classification accuracy. The results show that, in most cases, GPMI not only achieves better prediction accuracy, but also better classification accuracy than the other imputation methods.

Automatic keyphrase extraction is the challenging task of assigning keyphrases to documents to capture the main topics. It assists many research areas in the field of text mining – indexing, clustering, and summarisation. A landmark research KEA (Keyphrase Extraction Algorithm) formulated the probl em as a supervised machine learning problem and successfully applied a Naïve Bayes model to it. KEA showed great promise but its performance is not satisfactory. Its state-of-art extension KEA++ significantly improved its performance but relies on a domain specific vocabulary which is often not available or incomplete for other domains. We present a novel domain-independent system (DIKEA) which makes three main contributions to this field of research: utilising the largest online knowledge source available, Wikipedia, for keyphrase candidate selection; adding new features including a Wikipedia-based feature, link probability; and further boosting performance by using a multilayer perceptron network. Our experiments showed that DIKEA outperformed KEA++ while keeping the overall solution domain-independent. DIKEA was also tested on a benchmark dataset provided by a workshop on Semantic Evaluation (SemEval-2010), allowing comparisons with the 19 other related systems which participated. Our experiments show that DIKEA ranks first when considering only the top 5 keyphrases extracted from each document, and ranks second overall.

This paper presents an evaluation of the effectiveness of a new sociallyassistive robot, Auti, in encouraging physical and verbal interactions in children with autism. It aims to encourage positive play behaviors such as gentle speaking and touching, with positive reinforcement through movement responses, and to discourage challenging behaviors, such as screaming or hitting through the removal of the reinforcing movements. This study evaluates the design by comparing a fullyinteractive Auti to an activeonly version, which does the same movements but does not respond to the child. Results from 18 participants indicate that the Interactive Auti does encourage positive behaviors more than the Activeonly version. However, further design is needed around addressing problematic behaviors.

Clustering ensembles are a common approach to clustering problem, which combine a collection of clustering into a superior solution. The key issues are how to generate different candidate solutions and how to combine them. Common approach for generating candidate clustering solutions ignores the multiple representations of the data (i.e., multiple views) and the standard approach of simply selecting the best solution from candidate clustering solutions ignores the fact that there may be a set of clusters from different candidate clustering solutions which can form a better clustering solution. This paper presents a new clustering method that exploits multiple views to generate different clustering solutions and then selects a combination of clusters to form a final clustering solution. Our method is based on Nondominated Sorting Genetic Algorithm (NSGA-II), which is a multi-objective optimization approach. Our new method is compared with five existing algorithms on three data sets that have increasing difficulty. The results show that our method significantly outperforms other methods.

For many years computing in New Zealand schools was focused on teaching students how to use computers, and there was little opportunity for students to learn about programming and computer science as formal subjects. In this article we review a series of initiatives that occurred from 2007 to 2009 that led to programming and computer science being made available formally as part of the National Certificate in Educational Achievement (NCEA), the main school-leaving assessment, in 2011. The changes were phased in from 2011 to 2013, and we review this process using the Darmstadt model, including describing the context of the school system, the socio-cultural factors in play before, during and after the changes, the nature of the new standards, the reactions and roles of the various stakeholders, and the teaching materials and methods that developed. The changes occurred very quickly, and we discuss the advantages and disadvantages of having such a rapid process. In all these changes, teachers have emerged as having a central role, as they have been key in instigating and implementing change.

In 2011 New Zealand introduced computer science into high schools after a long period when computing was mainly focussed on training students to be users. The transition was rapid, and teachers had little time to upskill to prepare for the new topics, and yet there was widespread voluntary adoption of the new standards. The role of teachers and the national teachers' organisation in making the change has been pivotal, and this paper reviews the changes from the teachers' perspective. This story is intended to inform those planning similar changes in other countries, and provide a context for the next steps in NZ. The discussion centres around a survey of 91~teachers, which reveals strong intrinsic motivation from teachers to make the changes, a mixture of prior knowledge and skills that teachers shared with each other through peer support and online communication, a low level of confidence as teachers of computer science, and a need for further professional development.

The New Zealand Government is moving towards restricting access to tertiary education and implementing a managed entry scheme. It is therefore important to be able to predict whether a student has a reasonable likelihood of succeeding in tertiary engineering study. New Zealand secondary schools mostly operate on the National Certificate of Educational Achievement (NCEA) model whereby subjects are assessed on the basis of discrete individual modules. The paper compares the NCEA results to first year grades in tertiary engineering subjects obtained from most New Zealand providers of the BE degree to determine whether these NCEA grades can be used as a predictor of success or failure in tertiary engineering programmes. This is the first nation-wide survey of its kind and has yielded surprising results. For example, predicting success based on whether a student has not achieved is more insightful than basing on achievements, which is counter to the basis of many school league tables worldwide.

Computer science became available as a nationally assessed topic in NZ schools for the first time in 2011. We review the introduction of computer science as a formal topic, including the level of adoption, issues that have arisen in the process of introducing it, and work that has been undertaken to address those issues.

Feature construction is an effort to transform the input space of classification problems in order to improve the classification performance. Feature construction is particularly important for classifier inducers that cannot transform their input space intrinsically. This paper proposes GPMFC, a multiple-feature construction system for classification problems using genetic programming (GP). This paper takes a nonwrapper approach by introducing a filter-based measure of goodness for constructed features. The constructed, high-level features are functions of original input features. These functions are evolved by GP using an entropy-based fitness function that maximizes the purity of class intervals. A decomposable objective function is proposed so that the system is able to construct multiple high-level features for each problem. The constructed features are used to transform the original input space to a new space with better separability. Extensive experiments are conducted on a number of benchmark problems and symbolic learning classifiers. The results show that, in most cases, the new approach is highly effective in increasing the classification performance in rule-based and decision tree classifiers. The constructed features help improve the learning performance of symbolic learners. The constructed features, however, may lack intelligibility.

This paper proposes a novel text representation for Web pages written in Vietnamese. This representation is based on an analysis of Vietnamese documents at phonetic level in which each document will be represented as a bag of phonemes. It is designed to capture sound-based information in documents and to be helpful for resolving some non-topic text classification problems including automatic Vietnamese language identification of a document, ancient Vietnamese document detection, author identification, and poem identification. We apply some typical machine learning methods including NB, KNN and SVMs to build text classifiers. The experimental results show a significant improvement in terms of effectiveness and efficiency compared to the traditional syllable based representation in most cases.

Large, unwieldy classes are a significant maintenance problem. Programmers dislike them because the fundamental logic is often obscured, making them hard to understand and modify. This paper proposes a solution - a semi-automatic technique for splitting large classes into smaller, more cohesive ones. The core of the technique is the use of betweenness clustering to identify the best way of partitioning a class. This turned a tedious manual process into a quick and simple semi-automated one in roughly one third of the cases we examined.

This paper analyses the distribution of swapped subtree sizes involved in crossover events in approximations of an optimal crossover operator that allows the root node to be crossed over. The goal is to examine how the offspring search space can be effectively reduced for given parents. It concludes that good crossover events have a strong preference for the roots of the parent programs and for nodes with small subtrees. This paper also quantifies the ability of crossover to optimise offspring fitness, and concludes that this ability is far below what was expected.

The standard tournament selection samples individuals with replacement. The sampling-with-replacement strategy has its advantages but also has issues. One of the commonly recognised issues is that it is possible to have some individuals not sampled at all during the selection phase. The not-sampled issue aggravates the loss of program diversity. However, it is not clear how the issue affects Genetic Programming (GP) search. This paper investigates the importance of the issue. The theoretical and experimental results show that the issue can be solved and the loss of diversity contributed by not-sampled individuals can be minimised. However, doing so does not appears to significantly improve a GP system. Our conclusion is that the not-sampled issue does not seriously affect the selection performance in the standard tournament selection.

Standard tournament selection samples individuals with replacement. The sampling-with-replacement strategy has its advantages but also has issues. One of the commonly recognised issues is that it is possible to have the same individual sampled multiple times in a tournament. Although the impact of this multi-sampled issue on genetic programming is not clear, some researchers believe that it may lower the probability of some good individuals being sampled or selected. One solution is to use an alternative tournament selection (no-replacement tournament selection), which samples individuals in a tournament without replacement. This paper analyses no-replacement tournament selection to investigate the impact of the scheme and the importance of the issue. Theoretical simulations show that when common tournament sizes and population sizes are used, no-replacement tournament selection does not make the selection behaviour significantly different from that in the standard one and that the multi-sampled issue seldom occurs. In general, the issue is not crucial to the selection behaviour of standard tournament selection.

Collaborative filtering recommender systems make predictions based on the preferences of users considered like-minded to the target user (user-based), or the popularities of items similar to the target item (item-based). There have been several approaches of combining user-based and item-based collaborative filtering. However, they are predominantly along the lines of averaging user-based and item-based predictions in a close-to-linear fashion, thus behave like smoothing mechanisms and only work well on sparse datasets. This article proposes a new way of combining user and item based collaborative filtering in a nonlinear fashion. The goal of the approach is to improve recommendation accuracy on regular datasets, by means of a more sensible neighbourhood similarity computation method that guides the user similarity computation using the itemspsila similarities to the item that is being predicted.

Reserve Bank of New Zealand (RBNZ) is one of many inflation-targeting central banks. The effective conduct of monetary policy requires the capacity to make accurate short and medium term predictions about price inflation. The RBNZ's prediction system is very complex, requiring many iterations and significant input from human experts. This paper investigates the capability of Genetic Programming (GP) to predict price inflation over short and medium terms. By using un-preprocessed economic time series over small intervals, the experimental results demonstrate that GP can produce predictions of price inflation with accuracy comparable to the RBNZ's official prediction system, over both short and medium terms.

The standard crossover operator in tree-based Genetic Programming (GP) is problematic in that it is most often destructive. Selecting crossover points with an implicit bias towards the leaves of a program tree aggravates its destructiveness and causes the code bloat problem in GE Therefore, a common view has been developed that adjusting the depth of crossover points to eliminate the bias can improve GP performance, and many attempts have been made to create effective crossover operators according to this view. As there are a large number of possible depth-control strategies, it is very difficult to identify the strategy that provides the most significant improvement in performance. This paper explores depth-control strategies by analysing the depth of crossover points in evolutionary process logs of five different GP systems on problems in three different domains. It concludes that controlling the depth of crossover points is an evolutionary stage dependent and problem dependent task, and obtaining a significant performance improvement is not trivial.

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 ofthe 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.

A common problem in genetic programming search algorithms is destructive crossover in which the offspring of good parents generally has worse performance than the parents. Designing constructive crossover operators and integrating some local search techniques into the breeding process have been suggested as solutions. This paper reports on experiments demonstrating that premature convergence may happen more often when using these techniques in combination with standard parent selection. It shows that modifying the selection pressure in the parent selection process is necessary to obtain a significant performance improvement.

Users attempt to express their search goals through web search queries. When a search goal has multiple components or aspects, documents that represent all the aspects are likely to be more relevant than those that only represent some aspects. Current web search engines often produce result sets whose top ranking documents represent only a subset of the query aspects. By expanding the query using the right keywords, the search engine can find documents that represent more query aspects and performance improves. This paper describes AbraQ, an approach for automatically finding the right keywords to expand the query. AbraQ identifies the aspects in the query, identifies which aspects are underrepresented in the result set of the original query, and finally, for any particularly underrepresented aspect, identifies keywords that would enhance that aspect's representation and automatically expands the query using the best one. The paper presents experiments that show AbraQ significantly increases the precision of hard queries, whereas traditional automatic query expansion techniques have not improved precision. AbraQ also compared favourably against a range of interactive query expansion techniques that require user involvement including clustering, web-log analysis, relevance feedback, and pseudo relevance feedback.

For many hard queries, users spend a lot of time refining their queries to find relevant documents. Many methods help by suggesting refinements, but it is hard for users to choose the best refinement, as the best refinements are often quite obscure. This paper presents Qasp, an approach that overcomes the limitations of other refinement approaches by using query aspects to find different refinements of ambiguous queries. Qasp clusters the refinements so that descriptive refinements occur together with more obscure and potentially better performing refinements, thereby explaining the effect of refinements to the user. Experiments are presented that show Qasp significantly increases the precision of hard queries. The experiments also show that Qasp's clustering method does find meaningful groups of refinements that help users choose good refinements, which would otherwise be overlooked.

Good Predecessor Programs (GPPs) are the ancestors of the best program found in a Genetic Programming (GP) evolution. This paper reports on an investigation into GPPs with the ultimate goal of reducing fitness evaluation cost in tree-based GP systems. A framework is developed for gathering information about GPPs and a series of experiments is conducted on a symbolic regression problem, a binary classification problem, and a multi-class classification program with increasing levels of difficulty in different domains. The analysis of the data shows that during evolution, GPPs typically constitute less than 33% of the total programs evaluated, and may constitute less than 5%. The analysis results further shows that in all evaluated programs, the proportion of GPPs is reduced by increasing tournament size and to a less extent, affected by population size. Problem difficulty seems to have no clear influence on the proportion of GPPs.

Web page clustering methods categorize and organize search results into semantically meaningful clusters that assist users with search refinement; but finding clusters that are semantically meaningful to users is difficult. In this paper, we describe a new Web page clustering algorithm, QDC, which uses the user's query as part of a reliable measure of cluster quality. The new algorithm has five key innovations: a new query directed cluster quality guide that uses the relationship between clusters and the query, an improved cluster merging method that generates semantically coherent clusters by using cluster description similarity in additional to cluster overlap, a new cluster splitting method that fixes the cluster chaining or cluster drifting problem, an improved heuristic for cluster selection that uses the query directed cluster quality guide, and a new method of improving clusters by ranking the pages by relevance to the cluster. We evaluate QDC by comparing its clustering performance against that of four other algorithms on eight data sets (four use full text data and four use snippet data) by using eleven different external evaluation measurements. We also evaluate QDC by informally analysing its real world usability and performance through comparison with six other algorithms on four data sets. QDC provides a substantial performance improvement over other Web page clustering algorithms

Web page clustering is a technology that puts semantically related Web pages into groups and is useful for categorizing, organizing, and refining search results. When clustering using only textual information, suffix tree clustering (STC) outperforms other clustering algorithms by making use of phrases and allowing clusters to overlap. One problem of STC and other similar algorithms is how to select a small set of clusters to display to the user from a very large set of generated clusters. The cluster selection method used in STC is flawed in that it does not handle overlapping clusters appropriately. This paper introduces a new cluster scoring function and a new cluster selection algorithm to overcome the problems with overlapping clusters, which are combined with STC to make a new clustering algorithm ESTC. This paper's experiments show that ESTC significantly outperforms STC and that even with less data ESTC performs similarly to a commercial clustering search engine.

Web clustering assists users of a search engine by presenting search results as clusters of related pages. Many clustering algorithms with different characteristics have been developed: but the lack of a standardized Web clustering evaluation method that can evaluate clusterings with different characteristics has prevented effective comparison of algorithms. The paper solves this by introducing a new structure for defining general ideal clusterings and new measurements for evaluating clusterings with different characteristics by comparing them against the general ideal clustering.

This paper describes a domain independent approach for automatically constructing information extraction patterns for semi-structured web pages. Given a randomly chosen page from a web site of similarly structured pages, the system identifies a region of the page that has a regular "tabular" structure, and then infers an extraction pattern that will match the "rows" of the region and identify the data elements. The approach was tested on three corpora containing a series of tabular web sites from different domains and achieved a success rate of at least 80%. A significant strength of the system is that it can infer extraction patterns from a single training page and does not require any manual labeling of the training page.

We describe a domain independent approach to automatically constructing information extraction patterns for semistructured Web pages. The approach was tested on three corpora containing a series of tabular Web sites from different domains and achieved a success rate of at least 80%. A significant strength of the system is that it can infer extraction patterns from a single training page and does not require any manual labeling of the training page.

This paper documents efforts that the department has made to support women students between 1991 and the 1996. Our major goal has been to reduce the high withdrawal rate of women students in our entry level course in computer science. We describe the approaches that have been taken to address this concern, and present the data which has been collected to track the results of our efforts. Our data suggests that providing a gender neutral content is not enough to ensure that men and women will retain similarly. In this paper we suggest policies which we feel may be beneficial in achieving similar male and female retention rates.

Reconstructing vasculature in three dimensions is a challenging problem. Early approaches concentrated on coronary vasculature in X-ray images, recent work uses magnetic resonance imagery of cerebral vasculature. In both cases a priori information has been used, and often the way this is represented has proven limiting to the scope of applications supported. For example, a particular representation may be useful only for X-ray images. This paper addresses two issues: 1) representing a collection of vasculature and 2) the reconstruction of individual vasculature from images. Our representation learns the variations in branching structures and vessel shapes that occur between individuals. It supports a vascular catalogue containing three-dimensional (3-D) anatomical models. The representation is task independent; here we use it to reconstruct vasculature from images. Our algorithm has four features to which we draw attention: 1) it is not premised wholly upon X-ray images (though that is our focus here); 2) it produces several feasible solutions rather than one; 3) it can generalize from the catalogue to reconstruct instances not yet learned; 4) it exhibits polynomial time complexity, reasonable memory consumption, and is reliable. Both our representation and reconstruction algorithm are new and useful approaches. In support of these claims, we present results gathered from X-rays of both simulated and real vasculature.