Fuzzy ART: An Adaptive Resonance Algorithm for Rapid, Stable Classification of Analog Patterns

The Fuzzy ART system introduced herein incorporates computations from fuzzy set theory into ART 1. For example, the intersection (n) operator used in ART 1 learning is replaced by the MIN operator (A) of fuzzy set theory. Fuzzy ART reduces to ART 1 in response to binary input vectors, but can also learn stable categories in response to analog input vectors. In particular, the MIN operator reduces to the intersection operator in the binary case. Learning is stable because all adaptive weights can only decrease in time. A preprocessing step, called complement coding, uses on-cell and off-cell responses to prevent category proliferation. Complement coding normalizes input vectors while preserving the amplitudes of individual feature activations.

Fuzzy ART: Fast Stable Learning and Categorization of Analog Patterns by an Adaptive Resonance System

A Fuzzy ART model capable of rapid stable learning of recognition categories in response to arbitrary sequences of analog or binary input patterns is described. Fuzzy ART incorporates computations from fuzzy set theory into the ART 1 neural network, which learns to categorize only binary input patterns. The generalization to learning both analog and binary input patterns is achieved by replacing appearances of the intersection operator (n) in AHT 1 by the MIN operator (Λ) of fuzzy set theory. The MIN operator reduces to the intersection operator in the binary case. Category proliferation is prevented by normalizing input vectors at a preprocessing stage. A normalization procedure called complement coding leads to a symmetric theory in which the MIN operator (Λ) and the MAX operator (v) of fuzzy set theory play complementary roles. Complement coding uses on-cells and off-cells to represent the input pattern, and preserves individual feature amplitudes while normalizing the total on-cell/off-cell vector. Learning is stable because all adaptive weights can only decrease in time. Decreasing weights correspond to increasing sizes of category "boxes". Smaller vigilance values lead to larger category boxes. Learning stops when the input space is covered by boxes. With fast learning and a finite input set of arbitrary size and composition, learning stabilizes after just one presentation of each input pattern. A fast-commit slow-recode option combines fast learning with a forgetting rule that buffers system memory against noise. Using this option, rare events can be rapidly learned, yet previously learned memories are not rapidly erased in response to statistically unreliable input fluctuations.

Fuzzy ARTMAP: A Neural Network Architecture for Incremental Supervised Learning of Analog Multidimensional Maps

A new neural network architecture is introduced for incremental supervised learning of recognition categories and multidimensional maps in response to arbitrary sequences of analog or binary input vectors. The architecture, called Fuzzy ARTMAP, achieves a synthesis of fuzzy logic and Adaptive Resonance Theory (ART) neural networks by exploiting a close formal similarity between the computations of fuzzy subsethood and ART category choice, resonance, and learning. Fuzzy ARTMAP also realizes a new Minimax Learning Rule that conjointly minimizes predictive error and maximizes code compression, or generalization. This is achieved by a match tracking process that increases the ART vigilance parameter by the minimum amount needed to correct a predictive error. As a result, the system automatically learns a minimal number of recognition categories, or "hidden units", to met accuracy criteria. Category proliferation is prevented by normalizing input vectors at a preprocessing stage. A normalization procedure called complement coding leads to a symmetric theory in which the MIN operator (Λ) and the MAX operator (v) of fuzzy logic play complementary roles. Complement coding uses on-cells and off-cells to represent the input pattern, and preserves individual feature amplitudes while normalizing the total on-cell/off-cell vector. Learning is stable because all adaptive weights can only decrease in time. Decreasing weights correspond to increasing sizes of category "boxes". Smaller vigilance values lead to larger category boxes. Improved prediction is achieved by training the system several times using different orderings of the input set. This voting strategy can also be used to assign probability estimates to competing predictions given small, noisy, or incomplete training sets. Four classes of simulations illustrate Fuzzy ARTMAP performance as compared to benchmark back propagation and genetic algorithm systems. These simulations include (i) finding points inside vs. outside a circle; (ii) learning to tell two spirals apart; (iii) incremental approximation of a piecewise continuous function; and (iv) a letter recognition database. The Fuzzy ARTMAP system is also compared to Salzberg's NGE system and to Simpson's FMMC system.

A Neural Network Realization of Fuzzy ART

A neural network realization of the fuzzy Adaptive Resonance Theory (ART) algorithm is described. Fuzzy ART is capable of rapid stable learning of recognition categories in response to arbitrary sequences of analog or binary input patterns. Fuzzy ART incorporates computations from fuzzy set theory into the ART 1 neural network, which learns to categorize only binary input patterns, thus enabling the network to learn both analog and binary input patterns. In the neural network realization of fuzzy ART, signal transduction obeys a path capacity rule. Category choice is determined by a combination of bottom-up signals and learned category biases. Top-down signals impose upper bounds on feature node activations.

Living long-term with acquired brain injury in Ireland: towards a counter discourse

Irish literature on Acquired Brain Injury (ABI) is very scant and is mainly deficits and/or needs based. The focus is generally on how to manage the short term needs of the younger population with ABI. The starting position of my thesis is that people living long-term with ABI are important participants in developing knowledge about this social phenomenon, living with ABI while accepting that their brain injury does not determine them. Six mature adults with ABI and their six significant others participated in this longitudinal study. Using a narrative approach in interviews, over twenty months, five repeat individual interviews with each of the twelve participants was held. From this I gained an understanding of their lived experiences, their life-world and their experiences of our local public ABI/disability services, systems and discourse. Along with this new empirical data, theoretical developments from occupational therapy, occupational science, sociology, and disability studies were also used within a meta-narrative informed by critical theory and critical realism to develop a synthesis of this study. Social analysis of their narratives co-constructed with me, allowed me generate nuanced insights into tendencies and social processes that impacted and continues to impact on their everyday-everynight living. I discuss in some depth here, the relational attitudinal, structural, occupational and environmental supports, barriers or discrimination that they face(d) in their search for social participation and community inclusion. Personal recognition of the disabled participants by their family, friends and/or local community, was generally enhanced after much suffering, social supports, slow recovery, and with some form of meaningful occupational engagement. This engagement was generally linked with pre-injury interests or habits, while Time itself became both a major aid and a need. The present local ABI discourse seldom includes advocacy and inclusion in everyday/every night local events, yet most participants sought both peer-support or collective recognition, and social/community inclusion to help develop their own counter-discourse to the dominant ABI discourse. This thesis aims to give a broad social explanation on aspects of their social becoming, 'self-sameness' and social participation, and the status of the disabled participants wanting to live 'the slow life'. Tensions and dialectical issues involved in moving from the category of a person in coma, to person with a disability, to being a citizen should not demote the need for special services. While individualized short-term neuro-rehabilitation is necessary, it is not sufficient. Along with the participants, this researcher asks that community health and/or social care planners and service-providers rethink how ABI is understood and represented, and how people with ABI are included in their local communities

Experiences and desires of people with tetraplegia living with and without electronic aids to daily living: an Irish focus group study

This qualitative descriptive study explores the lived experience for persons with a high cervical spinal cord injury who have Electronic Aids to Daily Living (EADLs), and for persons who have no EADLs. Fifteen people with cervical spinal cord injuries attended four focus groups. Data analysis uncovered a novel framework of several themes that were organised into three categories: experiences, desires and meanings of living with EADL. Users’ and non users’ groups revealed homogenous themes. Experiences and desires are explored further in this paper. Themes within the category of experiences included: EADL devices, supply support and training, abandonment, mouthsticks and powered wheelchairs. Desires included: simple stuff, reliability, aesthetics and voice activation. Findings offer valuable personal insights about life with EADL to be considered by all involved with EADL.

Innovation in Ireland's 'high-technology' businesses: the roles of interaction and proximity

This thesis explores the drivers of innovation in Irish high-technology businesses and estimates, in particular, the relative importance of interaction with external businesses and other organisations as a source of knowledge for innovation at the business-level. The thesis also examines the extent to which interaction for innovation in these businesses occurs on a local or regional basis. The study uses original survey data of 184 businesses in the Chemical and Pharmaceutical, Information and Communications Technology and Engineering and Electronic Devices sectors. The study considers both product and process innovation at the level of the business and develops new measures of innovation output. For the first time in an Irish study, the incidence and frequency of interaction is measured for each of a range of agents, other group companies, suppliers, customers, competitors, academic-based researchers and innovation-supporting agencies. The geographic proximity between the business and each of the most important of each of each category of agent is measured using average one-way driving distance, which is the first time such a measure has been used in an Irish study of innovation. Utilising econometric estimation techniques, it is found that interaction with customers, suppliers and innovation-supporting agencies is positively associated with innovation in Irish high-technology businesses. Surprisingly, however, interaction with academic-based researchers is found to have a negative effect on innovation output at the business-level. While interaction generally emerges as a positive influence on business innovation, there is little evidence that this occurs at a local or regional level. Furthermore, there is little support for the presence of localisation economies for high-technology sectors, though some tentative evidence of urbanisation economies. This has important implications for Irish regional, enterprise and innovation policy, which has emphasised the development of clusters of internationally competitive businesses. The thesis brings into question the suitability of a cluster-driven network based approach to business development and competitiveness in an Irish context.

Alcohol income and health: a complicated but desirable mix

The aim of this thesis is to examine if a difference exists in income for different categories of drinkers in Ireland using the 2007 Slán data set. The possible impact of alcohol consumption on health status and health care utilisation is also examined. Potential endogeneity and selection bias is accounted for throughout. Endogeneity is where an independent variable included in the model is determined within the context of the model (Chenhall and Moers, 2007). An endogenous relationship between income and alcohol and between health and alcohol is accounted for by the use of separate income equations and separate health status equations for each category of drinker similar to what was done in previous studies into the effects of alcohol on earnings (Hamilton and Hamilton, 1997; Barrett, 2002). Sample selection bias arises when a sector selection is non-random due to individuals choosing a particular sector because of their personal characteristics (Heckman, 1979; Zhang, 2004). In relation to alcohol consumption, selection bias may arise as people may select into a particular drinker group due to the fact that they know that by doing so it will not have a negative effect on their income or health (Hamilton and Hamilton, 1997; Di Pietro and Pedace, 2008; Barrett, 2002). Selection bias of alcohol consumption is accounted for by using the Multinomial Logit OLS Two Step Estimate as proposed by Lee (1982), which is an extension of the Heckman Probit OLS Two Step Estimate. Alcohol status as an ordered variable is examined and possible methods of estimation accounting for this ordinality while also accounting for selection bias are looked at. Limited Information Methods and Full Information Methods of estimation of simultaneous equations are assessed and compared. Findings show that in Ireland moderate drinkers have a higher income compared with abstainers or heavy drinkers. Some studies such as Barrett (2002) argue that this is as a consequence of alcohol improving ones health, which in turn can influence ones productivity which may ultimately be reflected in earnings, due to the fact that previous studies have found that moderate levels of alcohol consumption are beneficial towards ones health status. This study goes on to examine the relationship between health status and alcohol consumption and whether the correlation between income and the consumption of alcohol is similar in terms of sign and magnitude to the correlation between health status and the consumption of alcohol. Results indicate that moderate drinkers have a higher income than non or heavy drinkers, with the weekly household income of moderate drinkers being €660.10, non drinkers being €546.75 and heavy drinkers being €449.99. Moderate Drinkers also report having a better health status than non drinkers and a slightly better health status than heavy drinkers. More non-drinkers report poor health than either moderate or heavy drinkers. As part of the analysis into the effect of alcohol consumption on income and on health status, the relationship between other socio economic variables such as gender, age, education among others, with income, health and alcohol status is examined.

Corporeal prisons: dynamics of body and mise-en-scène in three films by Paul Schrader

This thesis focuses on the complex relationship between representations of the human body and the formal processes of mise-en-scène in three consecutive films by the writer-director Paul Schrader: American Gigolo (1980), Cat People (1982) and Mishima: A Life in Four Chapters (1985). While Schrader’s work has typically been critiqued under the broad category of masculinity in crisis (and often as a subset of the films of his more famous long-time collaborator, Martin Scorsese), I focus on a fiveyear early period of his filmography when he sought to explore his key themes of bodily crisis, fragmentation and alienation through an unusually intense focus upon the expressive potential of film form, specifically via the combined elements of colour, lighting, camerawork and production design. By approaching these three films as corporeal character studies of troubled figures whose emotional and psychosexual neurosis is experienced in and through the body, I will locate Schrader’s filmmaking process and style within the thematic and aesthetic contexts of both his own early film criticism and the European and Japanese art cinemas that he claims as his primary influence. In doing so, I will establish Schrader’s position as a director whose literary and theological background differentiated him from his peers of the postclassical Hollywood generation, and who thus continually sought to develop his own visual literacy through his relationship with the camera and his collaborations with more overtly style-oriented film artists. But instead of merely focusing on mise-en-scène to gain a formalist appreciation of these films, I mobilise stylistic analysis as a new critical approach towards the problematic discourses of identity and embodiment that have haunted Schrader’s career from the beginning. In particular, I argue that paying closer attention to Schrader’s formal choices sheds new light on how these films – which he approached as exercises in style – repeatedly deal with the volatile and unavoidably body-oriented categories of race, gender and sexuality. In the process, I argue that a formalist attentiveness to mise-en-scène can also provide valuable cultural insights into Schrader’s oeuvre.

Proteomic approaches to environmental stress in mussel Mytilus edulis due to emerging classes of anthropogenic pollutants

Anthropogenic pollutant chemicals pose a major threat to aquatic organisms. There is a need for more research on emerging categories of environmental chemicals such as nanomaterials, endocrine disruptors and pharmaceuticals. Proteomics offers options and advantages for early warning of alterations in environmental quality by detecting sub-lethal changes in sentinel species such as the mussel, Mytilus edulis. This thesis aimed to compare the potential of traditional biomarkers (such as enzyme activity measurement) and newer redox proteomic approaches. Environmental proteomics, especially a redox proteomics toolbox, may be a novel way to study pollutant effects on organisms which can also yield information on risks to human health. In particular, it can probe subtle biochemical changes at sub-lethal concentrations and thus offer novel insights to toxicity mechanisms. In the first instance, the present research involved a field-study in three stations in Cork Harbour, Ireland (Haulbowline, Ringaskiddy and Douglas) compared to an outharbour control site in Bantry Bay, Ireland. Then, further research was carried out to detect effects of anthropogenic pollution on selected chemicals. Diclofenac is an example of veterinary and human pharmaceuticals, an emerging category of chemical pollutants, with potential to cause serious toxicity to non-target organisms. A second chemical used for this study was copper which is a key source of contamination in marine ecosystems. Thirdly, bisphenol A is a major anthropogenic chemical mainly used in polycarbonate plastics manufacturing that is widespread in the environment. It is also suspected to be an endocrine disruptor. Effects on the gill, the principal feeding organ of mussels, were investigated in particular. Effects on digestive gland were also investigated to compare different outcomes from each tissue. Across the three anthropogenic chemicals studied (diclofenac, copper and bisphenol A), only diclofenac exposure did not show any significant difference towards glutathione transferase (GST) responses. Meanwhile, copper and bisphenol A significantly increased GST in gill. Glutathione reductase (GR) enzyme analysis revealed that all three chemicals have significant responses in gill. Catalase activity showed significant differences in digestive gland exposed to diclofenac and gills exposed to bisphenol A. This study focused then on application of redox proteomics; the study of the oxidative modification of proteins, to M. edulis. Thiol proteins were labelled with 5-iodoacetamidofluorescein prior to one-dimensional and two-dimensional electrophoresis. This clearly revealed some similarities on a portion of the redox proteome across chemical exposures indicating where toxicity mechanism may be common and where effects are unique to a single treatment. This thesis documents that proteomics is a robust tool to provide valuable insights into possible mechanisms of toxicity of anthropogenic contaminants in M. edulis. It is concluded that future research should focus on gill tissue, on protein thiols and on key individual proteins discovered in this study such as calreticulin and arginine kinase which have not previously been considered as biomarkers in aquatic toxicology prior to this study.

Religion, the study of religion and other products of transnational and colonial imaginings

The category of ‘religion’ as contemporary scholarship has demonstrated is a fairly recent innovation, dating back only a few hundred years in Western thought, and ‘world religions’ as we think of it and as we teach it is an even more recent category, emerging out of European colonialism. Thus the academic study of religion is both the product and, at times, the agent of colonial modes of knowledge. And yet, it is perhaps because ‘religion’ continues to be invented and reinvented through connections across cultures that investigating the work of religious ideas and practices offers such fruitful possibilities for understanding the work of culture and power. This article investigates religion and the study of religion as a mode of anti-colonial practice, seeking to understand how each have the potential to cross boundaries, build bridges and produce critical insights into assumptions and worldviews too often taken for granted.

Monstrative acts and becoming-monster: on identity, bodies, and the feminine other

Using two examples of literary monsters, the Creature in Mary Shelley’s Frankenstein (1818), and Grendel’s Mother in Beowulf, this thesis demonstrates the bearing fictional identities have on “real” bodies, through an examination of two further literary texts, David Henry Hwang’s play, M. Butterfly (1986) and J. M. Coetzee’s novel, Disgrace (1999). Western definitions of Being have historically divided body and mind, favouring the mind as formative of subjective experience and denigrating the body as secondary and impure. This thesis demonstrates that this mind/body binary is symptomatic of the masculine ontological imperative to disown the body and its effects on Being, simultaneously ridding itself of the feminine it believes is its irrational opposite. Using recent feminist reviews of the canon, which emphasise the body’s importance to ontology and demonstrate the conceptual association between the feminine and the corporeal, this thesis links performative identity practices to theories of monstrosity, explaining how fictional qualities adhere to monstrous bodies by proposing a new theoretical category, the “monstrative.” The monstrative is a performative force that makes the Other into a living sign of Otherness; however, unlike earlier theories of Othering, the monstrative accounts for the Other’s being other to herself. This thesis also attempts to read the misrepresented body of the Other as a possible site for more empowered identity performances, where the monstrous “I” is interpreted as a potentially positive model for identity practice, through the conceptualisation of identity as a process of Becoming rather than Being. The transferal from a noun to a verb not only emphasises the performativity of identity, but also suggests fluidity and multiplicity in identity practice, which always already indicates a monstrosity at work. Thus, while monstrative acts constitute bodies as monstrous, Becoming-monster is an empathetic response to the Other’s monstrosity.

Novel co-crystals of the nutraceutical sinapic acid

Sinapic acid (SA) is a nutraceutical with known anti-oxidant, anti-microbial, anti-inflammatory, anti-cancer, and anti-anxiety properties. Novel co-crystals of SA were prepared with co-formers belonging to the category of GRAS [isonicotinic acid (INC), nicotinamide (NIA)], non-GRAS [4-pyridinecarbonitrile (PYC)], and active pharmaceutical ingredients (APIs) [6-propyl-2-thiouracil (PTU)] list of compounds. Structural study based on the X-ray crystal structures revealed the intermolecular hydrogen-bonded interactions and molecular packing. The crystal structure of sinapic acid shows the anticipated acid-acid homodimer along with discrete hydrogen bonds between the acid carbonyl and the phenolic moiety. The robust acid-acid homodimer appears to be very stable and is retained in the structures of two co-crystals (SA[middle dot]NIA and SA[middle dot]PYC). In these cases, co-crystallization occurs via intermolecular phenol O-H[three dots, centered]Naromatic hydrogen bonds between the co-formers. In the SA[middle dot]PTU[middle dot]2MeCN co-crystal the acid-acid homodimer gives way to the anticipated acid-amide heterodimer, with the phenolic moiety of SA hydrogen-bonded to acetonitrile. Attempts at obtaining the desolvated co-crystal led to lattice breakdown, thus highlighting the importance of acetonitrile in the formation of the co-crystal. Among the co-crystals examined, SA[middle dot]INC (5 weeks), SA[middle dot]NIA (8 weeks) and SA[middle dot]PYC (5 weeks) were found to be stable under accelerated humidity conditions (40 [degree]C, 75% RH), whereas SA[middle dot]PTU[middle dot]2MeCN decomposed after one week into individual components due to solvent loss.

Examining Relationships between Culture, Creativity and Business Stage in an Emerging Market: A Categorical Data Analysis of Vietnam’s Data Set

In this article, we offer a new way of exploring relationships between three different dimensions of a business operation, namely the stage of business development, the methods of creativity and the major cultural values. Although separately, each of these has gained enormous attention from the management research community, evidenced by a large volume of research studies, there have been not many studies that attempt to describe the logic that connect these three important aspects of a business; let alone empirical evidences that support any significant relationships among these variables. The paper also provides a data set and an empirical investigation on that data set, using a categorical data analysis, to conclude that examinations of these possible relationships are meaningful and possible for seemingly unquantifiable information. The results also show that the most significant category among all creativity methods employed in Vietnamese enterprises is the “creative disciplines” rule in the “entrepreneurial phase,” while in general creative disciplines have played a critical role in explaining the structure of our data sample, for both stages of development in our consideration.

STUDIES AND PERFORMANCES OF TRANSCRIPTIONS FOR CELLO FROM THE VIOLIN REPERTOIRE

The potential of the violoncello as a solo instrument was recognized and supported by cellists such as Luigi Boccherini (1743-1805), Luis Duport (1749-1819), Auguste Franchomme (1808-1884), and Alfredo Piatti (1822-1901). These pioneers composed technically demanding etudes, exercises, and caprices for the cello that were comparable to those already present in the violin literature. Even so, in the late nineteenth century and early twentieth century, considerably fewer substantial works were brought forth for the cello as compared with the violin. Consequently, many cellists such as Luigi Silva (1903-1961), Gregor Piatigorsky (1903-1976), Pierre Fournier (1906-1986), and Janos Starker (b. 1924) selected notable pieces from the violin repertoire and transcribed these for the cello. Some composers themselves actually adapted for the cello their own works originally written for the violin. Johannes Brahms with his Violin Sonata Op. 78, Igor Stravinsky with his Suite Italienne, and Béla Bartók with his First Rhapsody all belong to this category. Adaptations such as these further raised awareness among composers and performers of the possibilities of the cello as an independent and expressive instrument. Thus, many composers from the early 1900s to the present were encouraged to write increasing numbers of more soloistic and demanding works for cello. Herein, I explore the repertoire of cello transcriptions in order to analyze the differences between the original and transcribed versions and the challenges found therein. The performer may attempt to recreate the effect originally intended for the violin or, more daringly, may strive to search for alternate presentations of the music more suitable and expressive of the cello's own character. The project includes two recitals of the following transcribed works presented at the University of Maryland College Park, School of Music: Sonata in A by César Franck, transcribed by Jules Delsart, Variations on a Theme from Rossini by Nicolo Paganini, transcribed by Fournier, Suite Italienne by Igor Stravinsky, transcribed with the help of Piatigorsky, Sonatina Op. 137, No. 1 by Franz Schubert, transcribed by Starker, First Rhapsody by Béla Bartók and Sonata, Op. 108 by Johannes Brahms, transcribed by Hsiao-mei Sun.