GA approach for finding Rough Set decision rules based on bireducts

Feature selection plays an important role in knowledge discovery and data mining nowadays. In traditional rough set theory, feature selection using reduct - the minimal discerning set of attributes - is an important area. Nevertheless, the original definition of a reduct is restrictive, so in one of the previous research it was proposed to take into account not only the horizontal reduction of information by feature selection, but also a vertical reduction considering suitable subsets of the original set of objects. Following the work mentioned above, a new approach to generate bireducts using a multi--objective genetic algorithm was proposed. Although the genetic algorithms were used to calculate reduct in some previous works, we did not find any work where genetic algorithms were adopted to calculate bireducts. Compared to the works done before in this area, the proposed method has less randomness in generating bireducts. The genetic algorithm system estimated a quality of each bireduct by values of two objective functions as evolution progresses, so consequently a set of bireducts with optimized values of these objectives was obtained. Different fitness evaluation methods and genetic operators, such as crossover and mutation, were applied and the prediction accuracies were compared. Five datasets were used to test the proposed method and two datasets were used to perform a comparison study. Statistical analysis using the one-way ANOVA test was performed to determine the significant difference between the results. The experiment showed that the proposed method was able to reduce the number of bireducts necessary in order to receive a good prediction accuracy. Also, the influence of different genetic operators and fitness evaluation strategies on the prediction accuracy was analyzed. It was shown that the prediction accuracies of the proposed method are comparable with the best results in machine learning literature, and some of them outperformed it.

Geochemistry and petrogenesis of the Proterozoic Pater metavolcanic suite, Spragge, Ontario /

The Pater metavolcanic suite (PVS) was extruded as part O'f the basal Pater Formation of the Huronian Supergroup ca. 2.4 Ga. They Ars classified as wi thin-plate tholeiites associated with an immature ri-fting episode, and are inter layered with associated vol cani clastic and metasedimentary units. Post-solidif ication alteration caused redistribution o-f the alkalies, Sr, Rb, Ba, Cu, and SiO^. Ce, Y, Zr, CFezOs (as total Fe), Al^Os, TiOa, and, PaOa are considered to have remained essentially immobile in least altered samples. Petrogenetic modelling indicates the PVS was derived from the partial melting of two geochemical ly similar sources in the sub-continental lithosphere. Fractionation was characterized by an oli vine-plagioclase assemblage and a sub-volcanic plagioclase-clinopyroxene assemblage. A comparative study indicates that enrichment of the postulated Huronian source cannot be reconciled by Archean contamination. Enrichment is thought to have been caused by hydrous veined metasomatic heterogeneities in the sub-continental lithosphere, generated by an Archean subduct ion event before 2.68 Ga.

NMR studies of the exchange reactions of CH3CN.BX3 with excess CH3CN /|nJoseph Fogelman. -- 260 St. Catharines, Ont. : [s. n.],

Exchange reactions between molecular complexes and excess acid or base are well known and have been extensively surveyed in the literature(l). Since the exchange mechanism will, in some way involve the breaking of the labile donor-acceptor bond, it follows that a discussion of the factors relating to bonding in molecular complexes will be relevant. In general, a strong Lewis base and a strong Lewis acid form a stable adduct provided that certain stereochemical requirements are met. A strong Lewis base has the following characteristics (1),(2) (i) high electron density at the donor site. (ii) a non-bonded electron pair which has a low ionization potential (iii) electron donating substituents at the donor atom site. (iv) facile approach of the site of the Lewis base to the acceptor site as dictated by the steric hindrance of the substituents. Examples of typical Lewis bases are ethers, nitriles, ketones, alcohols, amines and phosphines. For a strong Lewis acid, the following properties are important:( i) low electron density at the acceptor site. (ii) electron withdrawing substituents. (iii) substituents which do not interfere with the close approach of the Lewis base. (iv) availability of a vacant orbital capable of accepting the lone electron pair of the donor atom. Examples of Lewis acids are the group III and IV halides such (M=B, AI, Ga, In) and MX4 - (M=Si, Ge, Sn, Pb). The relative bond strengths of molecular complexes have been investigated by:- (i) (ii) (iii) (iv) (v] (vi) dipole moment measurements (3). shifts of the carbonyl peaks in the IIIR. (4) ,(5), (6) .. NMR chemical shift data (4),(7),(8),(9). D.V. and visible spectrophotometric shifts (10),(11). equilibrium constant data (12), (13). heats of dissociation and heats of reactions (l~), (16), (17), (18), (19). Many experiments have bben carried out on boron trihalides in order to determine their relative acid strengths. Using pyridine, nitrobenzene, acetonitrile and trimethylamine as reference Lewis bases, it was found that the acid strength varied in order:RBx3 > BC1 3 >BF 3 • For the acetonitrile-boron trihalide and trimethylamine boron trihalide complexes in nitrobenzene, an-NMR study (7) showed that the shift to lower field was. greatest for the BB~3 adduct ~n~ smallest for the BF 3 which is in agreement with the acid strengths. If electronegativities of the substituents were the only important effect, and since c~ Br ,one would expect the electron density at the boron nucleus to vary as BF3ganization energy in explaining the formation of the adduct bond (19).

Using genetic algorithms for the single allocation hub location problem

Hub location problem is an NP-hard problem that frequently arises in the design of transportation and distribution systems, postal delivery networks, and airline passenger flow. This work focuses on the Single Allocation Hub Location Problem (SAHLP). Genetic Algorithms (GAs) for the capacitated and uncapacitated variants of the SAHLP based on new chromosome representations and crossover operators are explored. The GAs is tested on two well-known sets of real-world problems with up to 200 nodes. The obtained results are very promising. For most of the test problems the GA obtains improved or best-known solutions and the computational time remains low. The proposed GAs can easily be extended to other variants of location problems arising in network design planning in transportation systems.

Escape and connection : a phenomenological investigation into the meaning of an after-school program for adolescent boys

The purpos e of this phenomenological research is to explore the meaning of a YMCA-sponsored after-school recreation program in the lives of four adolescent boys. Listening to youth voice is impor t ant to the ability of othe r s to design, implement and evaluate high-quality programs tha t facilitate learning opportunities tha t a r e meaningful to participants. Within the context of interviews, task-based activities we r e used to ga the r data. Guided by Creswell's analytic spiral (1998), data wa s analyzed according to van Manen's (1990) thematic analysis and Caeilli's (2000) creative narrative analysis. It wa s found tha t this after-school progr am provided the s e adolescents with the opportunity to escape from the i r monotonous after-school activities and the instability of the i r home and school environments. Also, they we r e connected wi th positive peers, caring adults and the wide r community, opportunities tha t we r e limited in othe r aspects of the i r lives. Methodological issues a r e also discussed.

Multi-Objective Genetic Algorithms for the Single Allocation Hub Location Problem

Hub Location Problems play vital economic roles in transportation and telecommunication networks where goods or people must be efficiently transferred from an origin to a destination point whilst direct origin-destination links are impractical. This work investigates the single allocation hub location problem, and proposes a genetic algorithm (GA) approach for it. The effectiveness of using a single-objective criterion measure for the problem is first explored. Next, a multi-objective GA employing various fitness evaluation strategies such as Pareto ranking, sum of ranks, and weighted sum strategies is presented. The effectiveness of the multi-objective GA is shown by comparison with an Integer Programming strategy, the only other multi-objective approach found in the literature for this problem. Lastly, two new crossover operators are proposed and an empirical study is done using small to large problem instances of the Civil Aeronautics Board (CAB) and Australian Post (AP) data sets.

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.