Data mining algorithms analyze and mine databases for discovering implicit, previously unknown and potentially useful knowledge. Frequent pattern mining algorithms discover sets of database items that often occur together. Many of the frequent pattern mining algorithms represent the discovered knowledge in the form of a long textual list containing these sets of frequently co-occurring database items. As the amount of discovered knowledge can be large, it may not be easy for most users to examine and understand such a long textual list of knowledge. In my M.Sc. thesis, I represent both the original database and the discovered knowledge in pictorial form. Specifically, I design a new interactive visualization system for viewing the original transaction data (which are then fed into the frequent pattern mining engine) and for revealing the interesting knowledge discovered from the transaction data in the form of mined patterns.

El objetivo general de esta tesis es el desarrollo de una técnica adaptativa para la extracción de conocimiento en grandes bases de datos. En el análisis de enormes volúmenes de datos resulta de interés contar con técnicas que permitan, primero analizar la información y obtener conocimiento útil en forma de reglas de clasificación y luego adaptar el conocimiento adquirido ante los cambios que ocurran en los datos originales. El aporte de la tesis está centrado en la definición de una técnica adaptativa que permite extraer conocimiento de grandes bases de datos a partir de un modelo dinámico capaz de adaptarse a los cambios de la información, obteniendo así una técnica de minería de datos que sea capaz de generar conocimiento útil, produciendo resultados que sean de provecho al usuario final. Los resultados de esta investigación pueden aplicarse en áreas tales como análisis de suelos, análisis genético, biología, robótica, economía, medicina, detección de fallas en plantas y comunicación de sistemas móviles. <i>(del texto de la contratapa)</i> / Tesis doctoral de la Facultad de Informática (UNLP). Grado alcanzado: Doctor en Ciencias Informáticas. Directores de tesis: Armando De Giusti y Laura Lanzarini. La tesis, presentada en el año 2012, obtuvo el Premio "Dr. Raúl Gallard" en el 2013.

A unified approach to the design and development of distributed control software is presented. This method is the result of a 'tight' integration between a formal method for concurrent systems (CSP) and a structured method for distributed control system (DARTS). The work presented in this thesis does not seek to extend the semantic model of CSP nor to design a specific control algorithm, rather, efforts are made to apply the existing specification and verification techniques to enhance the formality of the well established and case-proven structured counterparts that benefits are captured from both methods. As a methodology is the central aim, the suggested approach is a first step towards a complete unified software development environment, which engineers can follow from organising design ideas to system implementation with proven correctness. The thesis develops a set of parameterised CSP predicates for expressing concurrency and communication together with a corresponding set of generic processes to reflect these specified behaviours. These generic processes are formal building blocks for generating system implementations at different levels of abstraction. Utilisation of DARTS criteria and the parameterised CSP objects frame the refinement strategies. Also, mappings of generic processes to pictorial representations are suggested which enable easy assimilation of the evolving designs. Applicability of the approach is demonstrated through a high level software design of a highperformance robot control system where its suitability is shown via requirement specifications, properties verification and implementation of salient behaviours using generic building blocks. Although verification often means rigorous mathematical reasoning, the thesis presents a proof assistant the Causality Diagram Evaluation Tool to automate the manipulation of CSP processes according to the defined algebraic laws. It is shown to be of value in reasoning with designs and implementations of the robot system. It is found that the analysis facility and the graphical interpretation of communication provided by the tool allow effective analysis and manipulation of early designs. The results derived from specifying essential design details, from transforming highly abstracted implementation models, and from investigation of system behaviours through formal reasoning and simulation conclude that formal methods, in particular CSP, has a niche value in enhancing software reliability at the sub-system level as well as providing a better underpinning to the structured method DARTS. The end product is a method to generate a correct and unambiguous document of the system concerned that is amenable to a direct implementation.

Real-world information containing subjective opinions and judgments has emerged the need to represent complex and imprecise data in databases. Additionally, the challenge of transferring information between databases whose data storage methods are not compatible has been an important research topic. Extensible Markup Language (XML) has the potential to meet these challenges since it has the ability to represent complex and imprecise data. In this thesis, an XML based fuzzy data representation and querying system is designed and implemented. The resulting system enables fuzzy querying on XML documents by using XQuery, a language used for querying XML documents. In the system, complex and imprecise data are represented using XML combined with the fuzzy representation. In addition to fuzzy querying, the system enables restructuring of XML Schemas by merging of elements of the XML documents. By using this feature of the system, one can generate a new XML Schema and new XML documents from the existing documents according to this new XML Schema. XML data used in the system are retrieved from Internet by Web Services, which can make use of XML&rsquo / s capabilities to transfer data and, XML documents are stored in a native XML database management system.

This is a study of quantitative electroencephalographic (QEEG) comodulation analysis, which is used to assist in identifying regional brain patterns associated with Chronic Fatigue Syndrome (CFS) compared to an EEG normative database. Further, this study investigates EEG patterns in depression which is found to be a highly comorbid condition to CFS. The QEEG comodulation analysis examines spatial-temporal cross-correlation of spectral estimates in the individual resting dominant frequency band. A pattern shown by Sterman and Kaiser (2001) and referred to as the Anterior Posterior Dissociation (APD) discloses a significant reduction in shared functional modulation between frontal and centro-parietal areas of the cortex. Conversely, depressed patients have not shown this pattern of activity but have disclosed a pattern of frontal Hypercomodulation localized to bilateral pre-frontal and frontal cortex. This research investigates these comodulation patterns to determine whether they exist reliably in these populations of interest and whether a clear distinction between two highly comorbid conditions can be made using this metric. Sixteen CFS sufferers and 16 depressed participants, diagnosed by physicians and a psychiatrist respectively were involved in QEEG data collection procedures. Nineteen-channel cap recordings were collected in five conditions: eyes-closed, eyes open, reading task-one, math computations task-two, and a second eyes-closed baseline. Five of the 16 CFS patients showed a clear Anterior Posterior Dissociation pattern for the eyes-closed resting dominant frequency. However, 11 participants did not show this pattern of dysregulation. Examination of the mean 8-12 Hz band spectral magnitudes across three cortical regions (frontal, central and parietal) indicated a trend of higher overall alpha levels in the parietal region in CFS patients who showed the APD pattern compared to those who did not show this pattern. All participants who showed the APD pattern were free of medication, while the majority of those absent of this pattern were using antidepressant medications. For the depressed group, all of which were medication free, 100 % of the depressed group showed a frontal Hypercomodulation pattern. Furthermore, examination of the mean 8-12 Hz band spectral magnitudes across three cortical regions disclosed a trend of high frontal alpha and a left/right asymmetry of greater voltages in the left frontal cortex. Although these samples are small, it is suggested that this method of evaluating the disorder of CFS holds promise. The fact that this pattern is not consistently represented in the CFS sample could be explained by the possibility of subtypes of CFS, or perhaps comorbid conditions. Further, the use of antidepressant medications may mask the pattern by altering the temporal characteristics of the EEG. This study, however, was able to demonstrate that the QEEG was able to parse out the regional cerebral brain differences between CFS and depressed group.

This dissertation challenges an unstudied area in moving objects database domains; predicting (long-term) future locations of moving objects. Moving object prediction enables us to provide a wide range of applications, such as traffic prediction, pre-detection of an aircraft collision, and reporting attractive gas prices for drivers along their routes ahead. Nevertheless, existing location prediction techniques are limited to support such applications since they are generally capable only of short-term predictions. In the real world, many objects exhibit typical movement patterns. This pattern information is able to serve as an important background to tackle the limitations of the existing prediction methods. We aims at offering foundations of pattern-aware prediction for moving objects, rendering more precise prediction results. Specifically, this thesis focuses on three parts. The first part of the thesis studies the problem of predicting future locations of moving objects in Euclidean space. We introduce a novel prediction approach, termed the hybrid prediction model, which utilizes not only the current motion of an object, but also the object's trajectory patterns for prediction. We define, mine, and index the trajectory patterns with a novel access method for efficient query processing. We then propose two different query processing techniques along given query time, i.e., for near future and for distant future. The second part covers the prediction problem for moving objects in network space. We formulate a network mobility model that offers a concise representation of mobility statistics extracted from massive collections of historical objects trajectories. This model captures turning patterns of the objects at junctions, at the granularity of individual objects as well as globally. Based on the model, we develop three different algorithms for predicting the future path of a mobile user moving in a road network, named the PathPredictors. The third part of the thesis extends the prediction problem for a single object to that for multiple objects. We introduce a convoy query that retrieves all groups of objects, i.e., convoys, from the objects' historical trajectories, each convoy consists of objects that have traveled together for some time; thus they may also move together in the future. We then propose three efficient algorithms for the convoy discovery, called the CuTS family, that adopt line simplification methods for reducing the size of the trajectories, permitting efficient query processing. For each part, we demonstrate comprehensive experimental results of our proposals, which show significantly improved accuracies for moving object prediction compared with state-of-the-art methods, while also facilitating efficient query processing.

This dissertation challenges an unstudied area in moving objects database domains; predicting (long-term) future locations of moving objects. Moving object prediction enables us to provide a wide range of applications, such as traffic prediction, pre-detection of an aircraft collision, and reporting attractive gas prices for drivers along their routes ahead. Nevertheless, existing location prediction techniques are limited to support such applications since they are generally capable only of short-term predictions. In the real world, many objects exhibit typical movement patterns. This pattern information is able to serve as an important background to tackle the limitations of the existing prediction methods. We aims at offering foundations of pattern-aware prediction for moving objects, rendering more precise prediction results. Specifically, this thesis focuses on three parts. The first part of the thesis studies the problem of predicting future locations of moving objects in Euclidean space. We introduce a novel prediction approach, termed the hybrid prediction model, which utilizes not only the current motion of an object, but also the object's trajectory patterns for prediction. We define, mine, and index the trajectory patterns with a novel access method for efficient query processing. We then propose two different query processing techniques along given query time, i.e., for near future and for distant future. The second part covers the prediction problem for moving objects in network space. We formulate a network mobility model that offers a concise representation of mobility statistics extracted from massive collections of historical objects trajectories. This model captures turning patterns of the objects at junctions, at the granularity of individual objects as well as globally. Based on the model, we develop three different algorithms for predicting the future path of a mobile user moving in a road network, named the PathPredictors. The third part of the thesis extends the prediction problem for a single object to that for multiple objects. We introduce a convoy query that retrieves all groups of objects, i.e., convoys, from the objects' historical trajectories, each convoy consists of objects that have traveled together for some time; thus they may also move together in the future. We then propose three efficient algorithms for the convoy discovery, called the CuTS family, that adopt line simplification methods for reducing the size of the trajectories, permitting efficient query processing. For each part, we demonstrate comprehensive experimental results of our proposals, which show significantly improved accuracies for moving object prediction compared with state-of-the-art methods, while also facilitating efficient query processing.

Marine environmental resource managers and consultants require comprehensive, accurate and current data on the status of marine biodiversity in order to fully evaluate resource consent applications that involve development, impact or encroachment within the marine environment, and for identifying areas of coast appropriate for conservation. The role and efficacy of existing global, national and regional marine biodiversity databases in delivering these types of data are evaluated. Consultation with environmental consultants revealed that none regularly, if ever used any existing marine biodiversity database during their routine consulting activities. Moreover, no existing biodiversity database had appropriate data-mining tools, although each was determined to provide information of value to resource managers and environmental consultants operating at national and regional scales; none was deemed to provide the sort of information required to manage marine resources at a local scale. To achieve the objectives of this research programme, resource managers, data users and data compilers were consulted to determine their ideal data and database requirements. Existing biodiversity data sets that included New Zealand marine biodiversity then were searched or procured, and these data and that of a novel data set of species occurring at 296 intertidal and 25 fringe-saline (effectively freshwater) sites within and proximal to the Hauraki Gulf Marine Park (from Mimiwhangata Bay in the north to Tauranga Harbour in the south) were compared with species inventories from environmental consultants operating in this region. Biodiversity data from the 296 saline, comprising presence/absence data for 713 taxa recorded from the survey region were analysed in detail. Significantly different species assemblages were identified amongst these 296 sites, five intertidal habitats being recognised, each with characteristic species assemblages: marine-hard shores, marine-soft shores, brackish-hard shores, brackish-soft shores, and mangrove shores. Species richness and diversity were consistently higher in marine habitats, and greatest on hard substrata. Most sites host unique assemblages of species. A novel index of species richness is proposed, and although the spatial distribution of richness isn’t particularly revealing, as obvious patterns in the distribution of richness are not apparent, this index has value in that the richness of any shore can be compared and contrasted with that of others throughout the region. Augmenting this richness index is a novel index of species rarity. Based on the frequency of occurrence of taxa on shores throughout the survey region, very rare through to ubiquitous taxa are recognised to routinely occur on almost all shores, regardless of the total species richness. Moreover, very rare to uncommon taxa often comprise a disproportionately high percentage of the total species occurring on any given intertidal shore, in any habitat. Accordingly, alarm bells should ring for reviewers of resource consent applications wherein environmental consultants state that an area subject to development ‘hosts no rare, unique or otherwise remarkable species or ecology.’ Two applications of these novel biodiversity data are demonstrated: the relationship between species richness and regional council consented activities is described, with a negative correlation reported for the intensity of disturbance (using the number of consented activities as a proxy for disturbance) and species richness on marine hard shores; and an appraisal of four selection criteria for marine reserves (naturalness, representativeness, uniqueness and complementarities), wherein the intertidal fauna and flora of no existing or proposed marine reserve appears to be natural, unique or representative, and effort seems to have been spent duplicating certain assemblages of species in reserve networks. The former is intuitively obvious, but the latter is alarming, and the ramifications of it far reaching in terms of conservation of the marine environment. Protocols for conducting biodiversity surveys must be established and implemented to elevate the standards of environmental consultants, resulting value judgements on the composition of species, and the likely and actual effects of these developments on the marine environment to ensure that statements made in reports are based on current data rather than perceptions and client expectations. It is possible that resource consents have been issued based on spurious appraisals of the immediate and cumulative effect of discharge on the environment, or of the relative rarity (or appreciation of this) of species that occur within it. Prior to development of the novel Monalisa data set, no existing database or data set existed that provided the information routinely required by managers and consultants to make informed judgements that affect coastal development throughout the survey region. Recommendations for additional research to build on findings detailed herein are made.

Diss. (sammanfattning) Stockholm : Karolinska institutet, 2007. / Härtill 4 uppsatser.

Diss. (sammanfattning) Stockholm : Karolinska institutet, 2005. / Härtill 6 uppsatser.