Search

As the number of artificial intelligence (AI) applications increases rapidly and more people will be affected by AI’s decisions, there are real needs for novel AI systems that can deliver both accuracy and explanations. To address these needs, this paper proposes a new approach called eXplainable Mapping Analytical Process (XMAP). Different from existing works in explainable AI, XMAP is highly modularised and the interpretability for each step can be easily obtained and visualised. A number of core algorithms are developed in XMAP to capture the distributions and topological structures of data, define contexts that emerged from data, and build effective representations for classification tasks. The experiments show that XMAP can provide useful and interpretable insights across analytical steps. For the binary classification task, its predictive performance is very competitive as compared to advanced machine learning algorithms in the literature. In some large datasets, XMAP can even outperform black-box algorithms without losing its interpretability.

© 2019 The Royal Society of New Zealand. Artificial intelligence (AI) emphasises the creation of intelligent machines/systems that function like humans. AI has been applied to many real-world applications. Machine learning is a branch of AI based on the idea that systems can learn from data, identify hidden patterns, and make decisions with little/minimal human intervention. Evolutionary computation is an umbrella of population-based intelligent/learning algorithms inspired by nature, where New Zealand has a good international reputation. This paper provides a review on evolutionary machine learning, i.e. evolutionary computation techniques for major machine learning tasks such as classification, regression and clustering, and emerging topics including combinatorial optimisation, computer vision, deep learning, transfer learning, and ensemble learning. The paper also provides a brief review of evolutionary learning applications, such as supply chain and manufacturing for milk/dairy, wine and seafood industries, which are important to New Zealand. Finally, the paper presents current issues with future perspectives in evolutionary machine learning.This is an Accepted Manuscript of an article published by Taylor & Francis in 'Journal of the Royal Society of New Zealand' on 2019-04-03, available online: https://www.tandfonline.com/10.1080/03036758.2019.1609052.

Recent research has shown that Particle Swarm Optimisation is a promising approach to feature selection. However, applying it on high-dimensional data with thousands to tens of thousands of features is still challenging because of the large search space. While filter approaches are time efficient and scalable for high-dimensional data, they usually obtain lower classification accuracy than wrapper approaches. On the other hand, wrapper methods require a longer running time than filter methods due to the learning algorithm involved in fitness evaluation. This paper proposes a new strategy of combining filter and wrapper approaches in a single evolutionary process in order to achieve smaller feature subsets with better classification performance in a shorter time. A new local search heuristic using symmetric uncertainty is proposed to refine the solutions found by PSO and a new hybrid fitness function is used to better evaluate candidate solutions. The proposed method is examined and compared with three recent PSO based methods on eight high-dimensional problems of varying difficulty. The results show that the new hybrid PSO is more effective and efficient than the other methods.© 2016 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.

This paper aims to produce classifiers that are not only accurate but also interpretable to decision makers. The classifiers are represented in the form of risk scores, i.e. simple linear classifiers where coefficient vectors are sparse and bounded integer vectors which are then optimised by a novel and scalable discrete particle swarm optimisation algorithm. In contrast to past studies which usually use particle swarm optimisation as a pre-processing step, the proposed algorithm incorporates particle swarm optimisation into the classification process. A penalty-based fitness function and a local search heuristic based on symmetric uncertainty are developed to efficiently identify classifiers with high classification performance and a preferred model size or complexity. Experiments with 10 benchmark datasets show that the proposed swarm-based algorithm is a strong candidate to develop effective linear classifiers. Comparisons with other interpretable machine learning algorithms that produce rule-based and tree-based classifiers also demonstrate the competitiveness of the proposed algorithm. Further analyses also confirm the interpretability of the produced classifiers. Finally, the proposed algorithm shows excellent speed-up via parallelisation, which gives it a great advantage when coping with large scale problems.

Growing big data has posed a great challenge for machine learning algorithms. To cope with big data, the algorithm has to be both efficient and accurate. Although evolutionary computation has been successfully applied to many complex machine learning tasks, its ability to handle big data is limited. In this paper, we proposed a dynamic self-organising swarm algorithm to learn an effective set of prototypes for big high-dimensional datasets in an unsupervised manner. The novelties of this new algorithm are the energy-based fitness function, the adaptive topological neighbourhood, the growing/shrinking capability, and the efficient learning scheme. Experiments with well-known datasets show that the proposed algorithm can maintain a very compact set of prototypes and achieve competitive predictive performance as compared to other algorithms in the literature. The analyses also show that prototypes generated by the proposed algorithms have a stronger separatability compared to those from other prototype generation algorithms.

Feature space is an important factor influencing the performance of any machine learning algorithm including classification methods. Feature selection aims to remove irrelevant and redundant features that may negatively affect the learning process especially on high-dimensional data, which usually suffers from the curse of dimensionality. Feature ranking is one of the most scalable feature selection approaches to high-dimensional problems, but most of them fail to automatically determine the number of selected features as well as detect redundancy between features. Particle swarm optimisation (PSO) is a population-based algorithm which has shown to be effective in addressing these limitations. However, its performance on high-dimensional data is still limited due to the large search space and high computation cost. This study proposes the first adaptive multi-swarm optimisation (AMSO) method for feature selection that can automatically select a feature subset of high-dimensional data more effectively and efficiently than the compared methods. The subswarms are automatically and dynamically changed based on their performance during the evolutionary process. Experiments on ten high-dimensional datasets of varying difficulties have shown that AMSO is more effective and more efficient than the compared PSO-based and traditional feature selection methods in most cases.

Falls are a major global health problem that may result in long-term health issues, disabilities, and even death (~650,000 fatalities each year). Each year approximately 37.3 million elderly worldwide experience a fall event that is severe enough to require medical attention. This research proposes a system for activity assessment and fall detection intended for in-home applications using the Internet of Things (IoT) for real-time detection of falls events and potentially fall prevention using a low-cost, wearable sensor system. While the proposed system uses an integrated MEMS sensor (i.e. accelerometer, gyroscope) for biomechanical monitoring and event detection, the developed algorithms can also be deployed onto other smart devices. The system communicates periodically with a cloud server system for uploading, archiving, and analyzing sensor data. Data is transferred to the cloud. Once data is received, the cloud server can provide alerts (i.e. automated calls, SMS, EMAIL) to formal and informal caregivers as well as emergency services when falls or other emergencies occur. Additionally, individuals and their caregivers can access and review personalized activity information to assess health, wellness and independence. Preliminary tests using healthy subjects performing Activity Daily Living (ADL) is quite promising, with specificity to detect fall of 100% in about 200 h normal activities in real-world setting. This research will also present practical challenges to deployment of the proposed system in real-world settings including usability, performance, and feedback from end-users.

Feature selection is an essential step in classification tasks with a large number of features, such as in gene expression data. Recent research has shown that particle swarm optimisation PSO is a promising approach to feature selection. However, it also has potential limitation to get stuck into local optima, especially for gene selection problems with a huge search space. Therefore, we developed a PSO algorithm PSO-LSRG with a fast “local search” combined with a gbest resetting mechanism as a way to improve the performance of PSO for feature selection. Furthermore, since many existing PSO-based feature selection approaches on the gene expression data have feature selection bias, i.e. no unseen test data is used, 2 sets of experiments on 10 gene expression datasets were designed: with and without feature selection bias. As compared to standard PSO, PSO with gbest resetting only, and PSO with local search only, PSO-LSRG obtained a substantial dimensionality reduction and a significant improvement on the classification performance in both sets of experiments. PSO-LSRG outperforms the other three algorithms when feature selection bias exists. When there is no feature selection bias, PSO-LSRG selects the smallest number of features in all cases, but the classification performance is slightly worse in a few cases, which may be caused by the overfitting problem. This shows that feature selection bias should be avoided when designing a feature selection algorithm to ensure its generalisation ability on unseen data.

Classification on high-dimensional data with thousands to tens of thousands of dimensions is a challenging task due to the high dimensionality and the quality of the feature set. The problem can be addressed by using feature selection to choose only informative features or feature construction to create new high-level features. Genetic programming (GP) using a tree-based representation can be used for both feature construction and implicit feature selection. This work presents a comprehensive study to investigate the use of GP for feature construction and selection on high-dimensional classification problems. Different combinations of the constructed and/or selected features are tested and compared on seven high-dimensional gene expression problems, and different classification algorithms are used to evaluate their performance. The results show that the constructed and/or selected feature sets can significantly reduce the dimensionality and maintain or even increase the classification accuracy in most cases. The cases with overfitting occurred are analysed via the distribution of features. Further analysis is also performed to show why the constructed feature can achieve promising classification performance. This is a post-peer-review, pre-copyedit version of an article published in 'Memetic Computing'. The final authenticated version is available online at: https://doi.org/10.1007/s12293-015-0173-y. The following terms of use apply: https://www.springer.com/gp/open-access/publication-policies/aam-terms-of-use.

The notion of cyberspace became impossible to separate from the notions of cyber threat and cyberattack. Since cyberattacks are getting easier to run, they are also becoming more serious threats from the economic damage perspective. Consequently, we are evident of a continuous adversarial relationship between the attackers trying to mount as powerful as possible attacks and defenders trying to stop the attackers in their goals. To defend against such attacks, defenders have at their disposal a plethora of techniques but they are often falling behind the attackers due to the fact that they need to protect the whole system while the attacker needs to find only a single weakness to exploit. In this paper, we consider one type of a cyberattack – network intrusion – and investigate how to use feature construction via genetic programming in order to improve the intrusion detection accuracy. The obtained results show that feature construction offers improvements in a number of tested scenarios and therefore should be considered as an important step in defense efforts. Such improvements are especially apparent in scenario with the highly unbalanced data, which also represents the most interesting case from the defensive perspective.

Genetic Programming (GP) has shown promise in feature construction where high-level features are formed by combining original features using predefined functions or operators. Multiple feature construction methods have been proposed for high-dimensional data with thousands of features. Results of these methods show that several constructed features can maintain or even improve the discriminating ability of the original feature set. However, some particular features may have better ability than other features to distinguish instances of one class from other classes. Therefore, it may be more difficult to construct a better discriminating feature when combing features that are relevant to different classes. In this study, we propose a new GP-based feature construction method called CDFC that constructs multiple features, each of which focuses on distinguishing one class from other classes. We propose a new representation for class-dependent feature construction and a new fitness function to better evaluate the constructed feature set. Results on eight datasets with varying difficulties showed that the features constructed by CDFC can improve the discriminating ability of thousands of original features in most cases. Results also showed that CFDC is more effective and efficient than the hybrid MGPFC method which was shown to have better performance than standard GP to feature construction.

Feature construction and feature selection are common pre-processing techniques to obtain smaller but better discriminating feature sets than the original ones. These two techniques are essential in high-dimensional data with thousands or tens of thousands of features where there may exist many irrelevant and redundant features. Genetic programming (GP) is a powerful technique that has shown promising results in feature construction and feature selection. However, constructing multiple features for high-dimensional data is still challenging due to its large search space. In this paper, we propose a GP-based method that simultaneously performs multiple feature construction and feature selection to automatically transform high-dimensional datasets into much smaller ones. Experiment results on six datasets show that the size of the generated feature set is less than 4% of the original feature set size and it significantly improves the performance of K-Nearest Neighbour, Naive Bayes and Decision Tree algorithms on 15 out of 18 comparisons. Compared with the single feature construction method using GP, the proposed method has better performance on half cases and similar on the other half. Comparisons between the constructed features, the selected features and the combination of both constructed and selected features by the propose method reveal different preferences of the three learning algorithms on these feature sets.

In this paper, we discuss a framework for bridging the gap between security and medical Large Data Analysis (LDA) with functional- biomarkers. Unsupervised Learning for individual e-IQ & IQ relying on memory eliciting (i.e. scent, grandmother images) and IQ baseline profiles could further enhance the ability to uniquely identify and properly diagnose individuals. Sub-threshold changes in a common/probable biomedical biomarker (disorders) means that an individual remains healthy, while a martingale would require further investigation and more measurements taken to determine credibility. Empirical measurements of human actions can discover anomalies hidden in data, which point to biomarkers revealed through stimulus response. We review the approach for forming a single-user baseline having 1-d devices and a scale-invariant representation for N users each (i) having N*d(i) total devices. Such a fractal representation of human-centric data provides self-similar levels information and relationships which are useful for diagnosis and identification causality anywhere from a mental disorder to a DNA match. Biomarkers from biomedical devices offer a robust way to collect data. Biometrics could be envisioned as enhanced and personalized biomedical devices (e.g. typing fist), but used for security. As long as the devices have a shared context origin, useful information can be found by coupling the sensors. In the case of the electroencephalogram (EEG), known patterns have emerged in low frequency Delta Theta Alpha Beta-Gamma (DTAB-G) waves when an individual views a familiar picture in the visual cortex which is shown on EEGs as a sharp peak. Using brainwaves as a functional biomarker for security can lead the industry to create more secure sessions by allowing not only passwords but also visual stimuli and/or keystrokes coupled with EEG to capture and stay informed about real time user e-IQ/IQ data changes. This holistic Computer Science (CS) Knowledge Discovery in Databases, Data Mining (KDD, DM) approach seeks to merge the fields having a shared data origin - biomarkers revealed through stimulus response.

Feature selection is a process of selecting a subset of relevant features from a large number of original features to achieve similar or better classification performance and improve the computation efficiency. As an important data pre-processing technique, research into feature selection has been carried out over the past four decades. Determining an optimal feature subset is a complicated problem. Due to the limitations of conventional methods, evolutionary computation EC has been proposed to solve feature selection problems. Particle swarm optimisation PSO is an EC technique which recently has caught much interest from researchers in the field. This paper presents a review of PSO for feature selection in classification. After describing the background of feature selection and PSO, recent work involving PSO for feature selection is reviewed. Current issues and challenges are also presented for future research.