Feature Selection and Classification Using Age Layered Population Structure Genetic Programming

The curse of dimensionality is a major problem in the fields of machine learning, data mining and knowledge discovery. Exhaustive search for the most optimal subset of relevant features from a high dimensional dataset is NP hard. Sub–optimal population based stochastic algorithms such as GP and GA are good choices for searching through large search spaces, and are usually more feasible than exhaustive and deterministic search algorithms. On the other hand, population based stochastic algorithms often suffer from premature convergence on mediocre sub–optimal solutions. The Age Layered Population Structure (ALPS) is a novel metaheuristic for overcoming the problem of premature convergence in evolutionary algorithms, and for improving search in the fitness landscape. The ALPS paradigm uses an age–measure to control breeding and competition between individuals in the population. This thesis uses a modification of the ALPS GP strategy called Feature Selection ALPS (FSALPS) for feature subset selection and classification of varied supervised learning tasks. FSALPS uses a novel frequency count system to rank features in the GP population based on evolved feature frequencies. The ranked features are translated into probabilities, which are used to control evolutionary processes such as terminal–symbol selection for the construction of GP trees/sub-trees. The FSALPS metaheuristic continuously refines the feature subset selection process whiles simultaneously evolving efficient classifiers through a non–converging evolutionary process that favors selection of features with high discrimination of class labels. We investigated and compared the performance of canonical GP, ALPS and FSALPS on high–dimensional benchmark classification datasets, including a hyperspectral image. Using Tukey’s HSD ANOVA test at a 95% confidence interval, ALPS and FSALPS dominated canonical GP in evolving smaller but efficient trees with less bloat expressions. FSALPS significantly outperformed canonical GP and ALPS and some reported feature selection strategies in related literature on dimensionality reduction.

Feature Selection and Classification Using Age Layered Population Structure Genetic Programming

The curse of dimensionality is a major problem in the fields of machine learning, data mining and knowledge discovery. Exhaustive search for the most optimal subset of relevant features from a high dimensional dataset is NP hard. Sub–optimal population based stochastic algorithms such as GP and GA are good choices for searching through large search spaces, and are usually more feasible than exhaustive and determinis- tic search algorithms. On the other hand, population based stochastic algorithms often suffer from premature convergence on mediocre sub–optimal solutions. The Age Layered Population Structure (ALPS) is a novel meta–heuristic for overcoming the problem of premature convergence in evolutionary algorithms, and for improving search in the fitness landscape. The ALPS paradigm uses an age–measure to control breeding and competition between individuals in the population. This thesis uses a modification of the ALPS GP strategy called Feature Selection ALPS (FSALPS) for feature subset selection and classification of varied supervised learning tasks. FSALPS uses a novel frequency count system to rank features in the GP population based on evolved feature frequencies. The ranked features are translated into probabilities, which are used to control evolutionary processes such as terminal–symbol selection for the construction of GP trees/sub-trees. The FSALPS meta–heuristic continuously refines the feature subset selection process whiles simultaneously evolving efficient classifiers through a non–converging evolutionary process that favors selection of features with high discrimination of class labels. We investigated and compared the performance of canonical GP, ALPS and FSALPS on high–dimensional benchmark classification datasets, including a hyperspectral image. Using Tukey’s HSD ANOVA test at a 95% confidence interval, ALPS and FSALPS dominated canonical GP in evolving smaller but efficient trees with less bloat expressions. FSALPS significantly outperformed canonical GP and ALPS and some reported feature selection strategies in related literature on dimensionality reduction.

Perturbing the impulse activity of a single identified neuron augments the formation of long-term memory in a molluscan semi-intact preparation /

The aim of this study was to investigate the neural correlates of operant conditioning in a semi-intact preparation of the pond snail, Lymnaea stagnalis. Lymnaea learns, via operant conditioning, to reduce its aerial respiratory behaviour in response to an aversive tactile stimulus to its open pneumostome. This thesis demonstrates the successful conditioning of na'ive semiintact preparations to show learning in the dish. Furthermore, these conditioned preparations show long-term memory that persists for at least 18 hours. As the neurons that generate this behaviour have been previously identified I can, for the first time, monitor neural activity during both learning and long-term memory consolidation in the same preparation. In particular, I record from the respiratory neuron Right Pedal Dorsal 1 (RPeD 1) which is part of the respiratory central pattern generator. In this study, I demonstrate that preventing RPeDl impulse activity between training sessions reduces the number of sessions needed to produce long-term memory in the present semi-intact preparation.

Self-Regulated Learning and Children At-Risk for Learning Disabilities: Using Motivational Tactics to Support "Reading Rocks!"

This project explored self-regulation among children impacted by leaming disabilities. More specifically, this thesis examined whether a remedial literacy program called Reading Rocks! offered by the Leaming Disabilities Association of Niagara Region, provided participating children opportunities to set goals, develop strategies to meet these goals, and provide intemal and extemal feedback- all processes associated with a model of self-regulated leaming as pioneered by Butler and Winne (1995) and Winne and Hadwin (1999). In this thesis, I triangulate the data through the combination of three different methodologies. Firstly, I describe the various elements of the Reading Rocks! program. Secondly, I analyze the data gathered through three semi-structured interviews with three parents of children that participated in the Reading Rocks! program to demonstrate whether the program provides opportunities for children to self-regulate their learning. Thirdly, I also analyze photographic evidence of the motivational workstation boards created by the tutors and children to further illustrate how Reading Rocks! promotes self-regulatory processes among children.