This thesis presents several novel methods to address some of the real world data modelling issues through the use of local and individualised modelling approaches. A set of real world data modelling issues such as modelling evolving processes, defining unique problem subspaces, identifying and dealing with noise, outliers, missing values, imbalanced data and irrelevant features, are reviewed and their impact on the models are analysed. The thesis has made nine major contributions to information science, includes four generic modelling methods, three real world application systems that apply these methods, a comprehensive review of the real world data modelling problems and a data analysis and modelling software. Four novel methods have been developed and published in the course of this study. They are: (1) DyNFIS – Dynamic Neuro-Fuzzy Inference System, (2) MUFIS – A Fuzzy Inference System That Uses Multiple Types of Fuzzy Rules, (3) Integrated Temporal and Spatial Multi-Model System, (4) Personalised Regression Model. DyNFIS addresses the issue of unique problem subspaces by identifying them through a clustering process, creating a fuzzy inference system based on the clusters and applies supervised learning to update the fuzzy rules, both antecedent and consequent part. This puts strong emphasis on the unique problem subspaces and allows easy to understand rules to be extracted from the model, which adds knowledge to the problem. MUFIS takes DyNFIS a step further by integrating a mixture of different types of fuzzy rules together in a single fuzzy inference system. In many real world problems, some problem subspaces were found to be more suitable for one type of fuzzy rule than others and, therefore, by integrating multiple types of fuzzy rules together, a better prediction can be made. The type of fuzzy rule assigned to each unique problem subspace also provides additional understanding of its characteristics. The Integrated Temporal and Spatial Multi-Model System is a different approach to integrating two contrasting views of the problem for better results. The temporal model uses recent data and the spatial model uses historical data to make the prediction. By combining the two through a dynamic contribution adjustment function, the system is able to provide stable yet accurate prediction on real world data modelling problems that have intermittently changing patterns. The personalised regression model is designed for classification problems. With the understanding that real world data modelling problems often involve noisy or irrelevant variables and the number of input vectors in each class may be highly imbalanced, these issues make the definition of unique problem subspaces less accurate. The proposed method uses a model selection system based on an incremental feature selection method to select the best set of features. A global model is then created based on this set of features and then optimised using training input vectors in the test input vector’s vicinity. This approach focus on the definition of the problem space and put emphasis the test input vector’s residing problem subspace. The novel generic prediction methods listed above have been applied to the following three real world data modelling problems: 1. Renal function evaluation which achieved higher accuracy than all other existing methods while allowing easy to understand rules to be extracted from the model for future studies. 2. Milk volume prediction system for Fonterra achieved a 20% improvement over the method currently used by Fonterra. 3. Prognoses system for pregnancy outcome prediction (SCOPE), achieved a more stable and slightly better accuracy than traditional statistical methods. These solutions constitute a contribution to the area of applied information science. In addition to the above contributions, a data analysis software package, NeuCom, was primarily developed by the author prior and during the PhD study to facilitate some of the standard experiments and analysis on various case studies. This is a full featured data analysis and modelling software that is freely available for non-commercial purposes (see Appendix A for more details). In summary, many real world problems consist of many smaller problems. It was found beneficial to acknowledge the existence of these sub-problems and address them through the use of local or personalised models. The rules extracted from the local models also brought about the availability of new knowledge for the researchers and allowed more in-depth study of the sub-problems to be carried out in future research.

This thesis presents several novel methods to address some of the real world data modelling issues through the use of local and individualised modelling approaches. A set of real world data modelling issues such as modelling evolving processes, defining unique problem subspaces, identifying and dealing with noise, outliers, missing values, imbalanced data and irrelevant features, are reviewed and their impact on the models are analysed. The thesis has made nine major contributions to information science, includes four generic modelling methods, three real world application systems that apply these methods, a comprehensive review of the real world data modelling problems and a data analysis and modelling software. Four novel methods have been developed and published in the course of this study. They are: (1) DyNFIS – Dynamic Neuro-Fuzzy Inference System, (2) MUFIS – A Fuzzy Inference System That Uses Multiple Types of Fuzzy Rules, (3) Integrated Temporal and Spatial Multi-Model System, (4) Personalised Regression Model. DyNFIS addresses the issue of unique problem subspaces by identifying them through a clustering process, creating a fuzzy inference system based on the clusters and applies supervised learning to update the fuzzy rules, both antecedent and consequent part. This puts strong emphasis on the unique problem subspaces and allows easy to understand rules to be extracted from the model, which adds knowledge to the problem. MUFIS takes DyNFIS a step further by integrating a mixture of different types of fuzzy rules together in a single fuzzy inference system. In many real world problems, some problem subspaces were found to be more suitable for one type of fuzzy rule than others and, therefore, by integrating multiple types of fuzzy rules together, a better prediction can be made. The type of fuzzy rule assigned to each unique problem subspace also provides additional understanding of its characteristics. The Integrated Temporal and Spatial Multi-Model System is a different approach to integrating two contrasting views of the problem for better results. The temporal model uses recent data and the spatial model uses historical data to make the prediction. By combining the two through a dynamic contribution adjustment function, the system is able to provide stable yet accurate prediction on real world data modelling problems that have intermittently changing patterns. The personalised regression model is designed for classification problems. With the understanding that real world data modelling problems often involve noisy or irrelevant variables and the number of input vectors in each class may be highly imbalanced, these issues make the definition of unique problem subspaces less accurate. The proposed method uses a model selection system based on an incremental feature selection method to select the best set of features. A global model is then created based on this set of features and then optimised using training input vectors in the test input vector’s vicinity. This approach focus on the definition of the problem space and put emphasis the test input vector’s residing problem subspace. The novel generic prediction methods listed above have been applied to the following three real world data modelling problems: 1. Renal function evaluation which achieved higher accuracy than all other existing methods while allowing easy to understand rules to be extracted from the model for future studies. 2. Milk volume prediction system for Fonterra achieved a 20% improvement over the method currently used by Fonterra. 3. Prognoses system for pregnancy outcome prediction (SCOPE), achieved a more stable and slightly better accuracy than traditional statistical methods. These solutions constitute a contribution to the area of applied information science. In addition to the above contributions, a data analysis software package, NeuCom, was primarily developed by the author prior and during the PhD study to facilitate some of the standard experiments and analysis on various case studies. This is a full featured data analysis and modelling software that is freely available for non-commercial purposes (see Appendix A for more details). In summary, many real world problems consist of many smaller problems. It was found beneficial to acknowledge the existence of these sub-problems and address them through the use of local or personalised models. The rules extracted from the local models also brought about the availability of new knowledge for the researchers and allowed more in-depth study of the sub-problems to be carried out in future research.

This research aims at finding a foundation for assessment of capabilities and applying the concept in a human resource selection. The research identifies a common ground for assessing individuals’ applied capability in a given job based on literature review of various disciplines in engineering, human sciences and economics. A set of criteria is found to be common and appropriate to be used as the basis of this assessment. Applied Capability is then described in this research as the impact of the person in fulfilling job requirements and also their level of usage from their resources with regards to the identified criteria. In other words how their available resources (abilities, skills, value sets, personal attributes and previous performance records) can be used in completing a job. Translation of the person’s resources and task requirements using the proposed criteria is done through a novel algorithm and two prevalent statistical inference techniques (OLS regression and Fuzzy) are used to estimate quantitative levels of impact and utilisation. A survey on post graduate students is conducted to estimate their applied capabilities in a given job. Moreover, expert academics are surveyed on their views on key applied capability assessment criteria, and how different levels of match between job requirement and person’s resources in those criteria might affect the impact levels. The results from both surveys were mathematically modelled and the predictive ability of the conceptual and mathematical developments were compared and further contrasted with the observed data. The models were tested for robustness using experimental data and the results for both estimation methods in both surveys are close to one another with the regression models being closer to observations. It is believed that this research has provided sound conceptual and mathematical platforms which can satisfactorily predict individuals’ applied capability in a given job. This research has contributed to the current knowledge and practice by a) providing a comparison of capability definitions and uses in different disciplines, b) defining criteria for applied capability assessment, c) developing an algorithm to capture applied capabilities, d) quantification of an existing parallel model and finally e) estimating impact and utilisation indices using mathematical methods.

Cognitive radio is a intelligent technology that helps in resolving the issue of spectrum scarcity. In a spectrum sharing network, where secondary user can communicate simultaneously along with the primary user in the same frequency band, one of the challenges in cognitive radio is to obtain balance between two conflicting goals that are to minimize the interference to the primary users and to improve the performance of the secondary user. In our thesis we have considered a primary link and a secondary link (cognitive link) in a fading channel. To improve the performance of the secondary user by maintaining the Quality of Service (Qos) to the primary user, we considered varying the transmit power of the cognitive user. Efficient utilization of power in any system helps in improving the performance of that system. For this we proposed ANFIS based opportunistic power control strategy with primary user’s SNR and primary user’s channel gain interference as inputs. By using fuzzy inference system, Qos of primary user is adhered and there is no need of complex feedback channel from primary receiver. The simulation results of the proposed strategy shows better performance than the one without power control. Initially we have considered propagation environment without path loss and then extended our concept to the propagation environment with path loss where we have considered relative distance between the links as one of the input parameters.

No description available.

Orientador: José Arnaldo Frutuoso Roveda / Resumo: No Brasil e no mundo é latente a necessidade de fazer a proteção, a ampliação e a promoção do patrimônio geológico, com a finalidade de garantir as próximas gerações acesso a este tipo de patrimônio natural, de modo econômica e ecologicamente sustentável. Diante disso, a Organização das Nações Unidas para a Educação, a Ciência e a Cultura - UNESCO criou a GGN - Global Geoparks Network, ou em sua tradução Rede Global de Geoparques - RGG. A Rede Global de Geoparques tem por função fazer a inclusão de novos parques membros, bem como, fazer a manutenção dos parques membros já existentes. Para que um parque seja postulante a membro do RGG, este deve atender uma série de requisitos, dentre as quais a Geoeducação, além disso, os parques membros de tempos em tempos devem passar por validação destes requisitos. Os parques candidatos possuem a dificuldade de elencar quais são as ações prioritárias num esforço de se tornarem membros da RGG. Como estes parques podem se tornar membros da RGG com o menor esforço. O presente trabalho se tem como objetivo apresenta uma proposta de criação um Sistema de Inferência Fuzzy (SIF) para tomada de decisões em Geoeducação, por parte dos gestores de áreas passíveis de se tornarem membros da RGG. A metodologia utilizada foi a teoria de conjuntos Fuzzy, através da criação de um sistema de inferência para geração de índices de adequação em Geoeducação das áreas candidatas. Foram criados 155 cenários para testar e validar o comportamento do sistema, o mes... (Resumo completo, clicar acesso eletrônico abaixo) / Mestre

The coronavirus (COVID-19) pandemic has caused severe adverse effects on the human life and the global economy affecting all communities and individuals due to its rapid spreading, increase in the number of affected cases and creating severe health issues and death cases worldwide. Since no particular treatment has been acknowledged so far for this disease, prompt detection of COVID-19 is essential to control and halt its chain. In this paper, we introduce an intelligent fuzzy inference system for the primary diagnosis of COVID-19. The system infers the likelihood level of COVID-19 infection based on the symptoms that appear on the patient. This proposed inference system can assist physicians in identifying the disease and help individuals to perform self-diagnosis on their own cases.

Accurate and timely diagnosis of faults is essential for the reliability and security of power grid operation and maintenance. The emergence of big data has enabled the incorporation of a vast amount of information in order to create custom fault datasets and improve the diagnostic capabilities of existing frameworks. Intelligent systems have been successful in incorporating big data to improve diagnostic performance using computational intelligence and machine learning based on fault datasets. Among these systems are fuzzy inference systems with the ability to tackle the ambiguities and uncertainties of a variety of input data such as climate data. This makes these systems a good choice for extracting knowledge from energy big data. In this thesis, qualitative climate information is used to construct a fault dataset. A fuzzy inference system is designed whose parameters are optimized using a single layer artificial neural network. This fault diagnosis framework maps the relationship between fault variables in the fault dataset and fault types in real-time to improve the accuracy and cost efficiency of the framework. / Graduate

Fine-scale resolution digital elevation models (DEMs) created from data collected using high precision instruments have become ubiquitous in fluvial geomorphology. They permit a diverse range of spatially explicit analyses including hydraulic modeling, habitat modeling and geomorphic change detection. Yet, the intercomparison of survey technologies across a diverse range of wadeable stream habitats has not yet been examined. Additionally, we lack an understanding regarding the precision of DEMs derived from ground-based surveys conducted by different, and inherently subjective, observers. This thesis addresses current knowledge gaps with the objectives i) to intercompare survey techniques for characterizing instream topography, and ii) to characterize observer variability in instream topographic surveys. To address objective i, we used total station (TS), real-time kinematic (rtk) GPS, terrestrial laser scanner (TLS), and infrared airborne laser scanning (ALS) topographic data from six sites of varying complexity in the Lemhi River Basin, Idaho. The accuracy of derived bare earth DEMs was evaluated relative to higher precision TS point data. Significant DEM discrepancies between pairwise techniques were calculated using propagated DEM errors thresholded at a 95% confidence interval. Mean discrepancies between TS and rtkGPS DEMs were relatively low (≤ 0.05 m), yet TS data collection time was up to 2.4 times longer than rtkGPS. ALS DEMs had lower accuracy than TS or rtkGPS DEMs, but ALS aerial coverage and floodplain topographic representation was superior to all other techniques. The TLS bare earth DEM accuracy and precision were lower than other techniques as a result of vegetation returns misinterpreted as ground returns. To address objective ii, we used a case study where seven field crews surveyed the same six sites to quantify the magnitude and effect of observer variability on DEMs interpolated from the survey data. We modeled two geomorphic change scenarios and calculated net erosion and deposition volumes at a 95% confidence interval. We observed several large magnitude elevation discrepancies across crews, however many of these i) tended to be highly localized, ii) were due to systematic errors, iii) did not significantly affect DEM-derived metric precision, and iv) can be corrected post-hoc.

No description available.