For hierarchical clustering, dendrograms provide convenient and powerful visualization. Although many visualization methods have been suggested for partitional clustering, their usefulness deteriorates quickly with increasing dimensionality of the data and/or they fail to represent structure between and within clusters simultaneously. In this paper we extend (dissimilarity) matrix shading with several reordering steps based on seriation. Both methods, matrix shading and seriation, have been well-known for a long time. However, only recent algorithmic improvements allow to use seriation for larger problems. Furthermore, seriation is used in a novel stepwise process (within each cluster and between clusters) which leads to a visualization technique that is independent of the dimensionality of the data. A big advantage is that it presents the structure between clusters and the micro-structure within clusters in one concise plot. This not only allows for judging cluster quality but also makes mis-specification of the number of clusters apparent. We give a detailed discussion of the construction of dissimilarity plots and demonstrate their usefulness with several examples. / Series: Research Report Series / Department of Statistics and Mathematics

One goal of persistent homology is to recover meaningful information from point-cloud data by examining long-lived topological features of filtered simplicial complexes built over the point-cloud. Motivated by real-world applications, the classic setting for this approach has been on finite metric spaces where many suitable complexes can be defined, and a natural filtration exists via sublevel sets of the metric. We consider the extension of persistent homology to dissimilarity networks equipped with a relaxed metric that does not assume symmetry nor the triangle inequality, by computing persistent homology on the directed clique complex defined over weighted directed graphs induced from a dissimilarity network and filtered by an adapted Rips filtration. We characterize digraph maps that induce maps on homology, describe a procedure to lift any digraph map to one that does induce maps on homology, and present a homotopy classification that provides a condition for two such digraph maps to induce the same map at the homology level. We also prove functoriality of directed clique homology and describe filtrations of digraphs induced by digraph maps. We then prove stability of persistent directed clique homology by showing that the persistence modules of a digraph and that of an admissible perturbation are interleaved. These admissible perturbations include perturbing dissimilarity measures in the network that either preserve the digraph structure or collapse series of arrows. We also explore similar constructions for maps between digraphs that allow reversal of arrows and show that while such maps, in general, produce unstable persistence barcodes, one can recover stability by inducing a reverse filtration and truncating at an appropriate threshold. Finally, we present an application of persistent directed clique homology to trace patterns and shapes embedded in migration and remittance networks.

Ο νεοπλατωνικός φιλόσοφος Πρόκλος, στο α΄ μέρος της πραγματείας του Εις τον Παρμενίδη του Πλάτωνος Β΄, διατυπώνει μία εισαγωγική σκέψη για την πλατωνική θεωρία των Ιδεών, την οποία παρουσιάζει αναλυτικά στην συνέχεια του έργου του, δηλώνοντας ρητά την αδιαμφισβήτητη αντικειμενική ύπαρξη των Ιδεών, ως αρχετύπων των όσων λαμβάνουν χώρα στον κόσμο της αισθητής εμπειρίας. Τονίζει , ότι τόσο οι Ιδέες, όσο και τα αισθητά οφείλουν την ύπαρξή τους στο Έν, την ανώτατη υπερβατική Αρχή, η οποία αποτελεί την απόλυτη ενότητα και από την οποία απορρέει το πλήθος. Επομένως, η ενότητα προηγείται του πλήθους και η εν λόγω προτεραιότητα παρουσιάζεται με μία σειρά από επιχειρήματα. Έπειτα αναζητείται η καταστατική τάξη των Ιδεών, οι οποίες βρίσκονται τοποθετημένες στον δικό τους «τόπο», συμπλέκονται χωρίς να υφίστανται αλλοιώσεις, διατηρώντας την μοναδικότητα, την ταυτότητα καθώς και την ενότητά τους. Συγκεκριμένα, παρακολουθούμε ποια είναι η ουσία και οι δυνάμεις της Ομοιότητας και της Ανομοιότητας αντιστοίχως, ποια είναι η διάταξή τους ανάμεσα στις Ιδέες ποια είναι η λειτουργική παρουσία τους και εν τέλει ποια είναι η μεταξύ τους σχέση. Πρόκειται για ένα ζήτημα που ανήκει στον συστηματικό κλάδο της μεταφυσικής οντολογίας αλλά και συγχρόνως εντάσσεται και στην ιστορία της φιλοσοφίας, καθότι παρουσιάζει ένα στάδιο της εξέλιξης μιας θεωρίας που αναπτύχθηκε τον 4ο αιώνα π. Χ. / --

Made available in DSpace on 2014-06-12T16:01:14Z (GMT). No. of bitstreams: 2 arquivo7274_1.pdf: 1733810 bytes, checksum: 2d9eb7a4489382e5afbf1790810474a0 (MD5) license.txt: 1748 bytes, checksum: 8a4605be74aa9ea9d79846c1fba20a33 (MD5) Previous issue date: 2005 / A análise de dados simbólicos (Symbolic Data Analysis) é um novo domínio na área de descoberta automática de conhecimento que visa desenvolver métodos para dados descritos por variáveis que podem assumir como valor conjuntos de categorias, intervalos ou distribuições de probabilidade. Essas novas variáveis permitem levar em conta a variabilidade e/ou a incerteza presente nos dados. O tratamento de dados simbólicos através de técnicas estatísticas e de aprendizagem de máquina necessita da introdução de medidas de distância capazes de manipular tal tipo de dado. Com esse objetivo, diversas funções de dissimilaridade têm sido propostas na literatura. Entretanto, nenhum estudo comparativo acerca do desempenho de tais funções em problemas que envolvem simultaneamente dados simbólicos booleanos e modais foi realizado. A principal contribuição dessa dissertação é realizar uma análise comparativa e uma avaliação empírica sobre funções de dissimilaridade para dados simbólicos, uma vez que esse tipo de estudo, apesar de muito relevante, é quase inexistente na literatura. Além disso, este trabalho também introduz novas funções de dissimilaridade que podem ser usadas no agrupamento dinâmico de dados simbólicos. Os algoritmos de agrupamento dinâmico consistem em obter, simultaneamente, uma partição em um número fixo de classes e a identificação de um representante para cada classe, minimizando localmente um critério que mede a adequação entre as classes e os seus representantes. Para validar esse estudo, foram realizados experimentos com bases de dados de referência na literatura e dois conjuntos de dados artificiais de intervalos com diferentes graus de dificuldade de classificação, objetivando a comparação das funções avaliadas. A precisão dos resultados foi mensurada por um índice externo de agrupamento aplicado na validação cruzada não supervisionada, para as bases de dados reais, e também no quadro de uma experiência Monte Carlo, para as bases de dados artificiais. Com os resultados alcançados é possível verificar a adequação das diversas funções de dissimilaridade aos diferentes tipos de dados simbólicos (multivalorado, multivalorado ordinal, intervalar, e modal de mesmo suporte e de suportes diferentes), bem como identificar as melhores configurações de funções. Testes estatísticos validam as conclusões

Clustering text documents is a fundamental task in modern data analysis, requiring approaches which perform well both in terms of solution quality and computational efficiency. Spherical k-means clustering is one approach to address both issues, employing cosine dissimilarities to perform prototype-based partitioning of term weight representations of the documents. This paper presents the theory underlying the standard spherical k-means problem and suitable extensions, and introduces the R extension package skmeans which provides a computational environment for spherical k-means clustering featuring several solvers: a fixed-point and genetic algorithm, and interfaces to two external solvers (CLUTO and Gmeans). Performance of these solvers is investigated by means of a large scale benchmark experiment. (authors' abstract)

Of the 20 standard amino acids, cysteines are the only amino acids that have a reactive sulphur atom, thus enabling two cysteines to form strong covalent bonds known as disulfide bonds. Even though almost all proteins have cysteines, not all of them have disulfide bonds. Disulfide bonds provide structural stability to proteins and hence are an important constraint in determining the structure of a protein. As a result, disulfide bonds are used to study various protein properties, one of them being protein folding. Protein structure prediction is the problem of predicting the three-dimensional structure of a protein from its one-dimensional amino acid sequence. Ab initio methods are a group of methods that attempt to solve this problem from first principles, using only basic physico-chemical properties of proteins. These methods use structure libraries of short amino acid fragments in the process of predicting the structure of a protein. The protein structures from which these structure libraries are created are not classified in any other way apart from being non-redundant. In this thesis, we investigate the structural dissimilarities of short amino acid fragments when occurring in proteins with disulfide bonds and when occurring in those proteins without disulfide bonds. We are interested in this because, as mentioned earlier, the protein structures from which the structure libraries of ab initio methods are created, are not classified in any form. This means that any significant structural difference in amino acids and short fragments when occurring in proteins with and without disulfide bonds would remain unnoticed as these structure libraries have both fragments from proteins with disulfide bonds and without disulfide bonds together. Our investigation of structural dissimilarities of amino acids and short fragments is done in four phases. In phase one, by statistically analysing the phi and psi backbone dihedral angle distributions we show that these fragments have significantly different structures in terms of dihedral angles when occurring in proteins with and without disulfide bonds. In phase two, using directional statistics we investigate how structurally different are the 20 different amino acids and the short fragments when occurring in proteins with and without disulfide bonds. In phase three of our work, we investigate the differences in secondary structure preference of the 20 amino acids in proteins with and without disulfide bonds. In phase four, we further investigate and show that there are significant differences within the same secondary structure region of amino acids when they occur in proteins with and without disulfide bonds. Finally, we present the design and implementation details of a dihedral angle and secondary structure database of short amino acid fragments (DASSD) that is publicly available. Thus, in this thesis we show previously unknown significant structure differences in terms of backbone dihedral angles and secondary structures in amino acids and short fragments when they occur in proteins with and without disulfide bonds.

No description available.

The aim of this thesis is to defend a reading of Descartes' theory of sensory perception in which, against a widespread interpretation, the mind is not a passive receiver of inputs from the environment, but an active decoder of neural information that contributes to the representational content of ideas. I call this the 'mental activity thesis' and, in the overall picture, I identify it as one of the philosophical implications of the seventeenth-century scientific revolution. Within Descartes' dualism, to offer a theory of sensory perception amounts to describing the interplay between the natural world, the brain, and the mind. Given his mechanistic, micro-corpuscular conception of matter, Descartes developed detailed physiological descriptions of the interaction between external objects and the brain. He envisaged it as an isomorphic relation in which the characteristics of objects are transmitted through the nerves to the brain as patterns of geometrically reduced properties. This process is often read as culminating with the mind being passively affected by a corporeal isomorph. Descartes' doctrine becomes elusive in its mental phase, but the passivity reading, so I contend, remains inadequate. I argue for the mental activity thesis through four claims. First, I subscribe the known view that Descartes is concerned about a version of the mind-body problem that is not equivalent to the problem of substance interaction. It is rather a problem of dissimilarity between mental representations and mechanistic explanations. The question is how the qualitative character of sensory experiences can arise from the quantitative notions of physical science. As a way of emphasising the weight that the problem of dissimilarity has for Descartes' philosophical decisions, I show that it motivates a metaphysically interesting distinction between types of causes for the case of brain-mind interaction. Second, I defend the position that, despite not holding a perfectly unambiguous doctrine, Descartes' introduction of natural signs is the closest that he got to formulating a full-fledged theory of sensory perception. The appeal to natural signs has been normally deemed as metaphorical in the literature. I argue that, on the contrary, it is possible to reconstruct a causal story for brain-mind interaction along the lines of a semantic model based on Descartes' identification of neural events with natural signs. A causal-semantic model emerges as a charitable, plausible reading that reveals the mind as an active interpreter. Third, in light of the mental activity thesis, I read Descartes' late appeal to the innateness of all ideas (notably in the Comments on a Certain Broadsheet) as a strategy to account for a type of representational content needed for sensory ideas that, while produced by the mind, is different from that of his paradigmatic innate ideas. I assist Descartes in exploring how the category of innateness captures mental activity within a causal-semantic theory. Fourth, in the course of this argumentation, and for further support, I address the role of the distinction between primary and secondary qualities in Descartes' theory. I tackle a pervasive objection stemming from his alleged association of the perception of primary qualities with the intellect. By reassessing Descartes' views on mental activity, this interpretation aims at a lucid description of sensory perception that goes beyond the rigid rationalism that is often credited to him.

Segregationen i Sverige har ökat de senaste åren vilket skapat ett allt större samhällsproblem. Det är därför viktigt att kommunerna i ett tidigt stadie aktivt börjar arbeta mot segregation och för social hållbarhet. Med denna studie och dess metodutveckling vill vi främja arbetet för mindre svenska tätorter genom att ta fram och testa en metod som kommuner kan använda vid en kartläggning av segregation. För att kunna hantera segregationens mångsidiga problemställning har flera metodsteg använts. En litteraturgenomgång resulterade i sex variabler som alla har betydande påverkan på segregationen. Dessa var fastighetsbestånd, förvärvsfrekvens, inrikes- och utrikesfödda, trångboddhet, utbildning och ålder. En enkätundersökning genomfördes sedan med 14 samhällsplanerare för att rangordna de variabler som litteraturgenomgången påvisade. Rangordningen användes sedan i metoden rank sum weights, som sedermera ledde fram till en multikriterieanalys (MKA) baserad på en linjär additiv metod. Målet med studien är att kunna presentera en kartläggning för segregationen i Bollnäs tätort utifrån den metod vi har utvecklat. Resultatet från studien visar att Bollnäs tätort har fyra områden som klassas som “mest segregerade” och tre områden som klassas som “minst segregerade”. Dessa resultat har testats med tre validitets- och reliabilitetsmetoder vilka var Index of Dissimilarity, intervju, samt känslighetsanalys. Resultatet från de tre testerna påvisar tydliga tendenser och snarlika resultat från vår metod, vilket indikerar att vår metod är robust och trovärdig. Behovet av ordentliga kartläggningar likt vår studie framkommer bland annat i intervjun där en planerare från Bollnäs kommun fick peka ut de fem områden personen ansåg vara mest segregerade. För dessa fem områden pekade planeraren ut ett område som enligt vår studie var minst segregerat, vilket påvisar svårigheten kring segregationsfrågan. Vår förhoppning med denna studie är att skapa en metod som främjar kommuners arbete med social hållbarhet. / Segregation in Sweden has been rising in recent years, which has lead to an increase in social problems. It is therefore important that municipalities, in an early stage, actively begin to work against segregation and towards social sustainability. With this study and its methodology we want to encourage smaller Swedish cities by developing and testing a method that municipalities can use in mapping of segregation. To handle segregation's complex problems, several of method steps have been applied. A literature review resulted in six variables, all of which have a significant impact on segregation. The variables were: housing tenure, employment rates, domestic/foreign-born, overcrowding, education and age. A survey was then conducted with 14 experts in the urban planning sector. The planners were asked to rank the variables that the literature review highlighted. The rankings were then used to create rank sum weights, which later led to a multicriteria decision analysis (MCDA) based on a linear additive method. The aim of the study is to present a mapping of the segregation in Bollnäs urban area based on the method we have developed. The result of the study shows that out of 15 areas in Bollnäs urban area, four were classified as "most segregated" and three areas classified as "least segregated". These results have been tested using three validity and reliability methods: The Index of Dissimilarity, interview and sensitivity analysis. The results of the three tests demonstrate clear trends and similar results compared with our method, which indicates that our method is robust and trustworthy. The necessity for proper mapping of segregation is demonstrated by the interview where a planner from Bollnäs municipality was given the chance to pinpoint five areas the person considered to be most segregated. For these five areas, the planner pointed out one area that according to our study was least segregated, which proves the difficulty of identifying segregation. Our hope with this study is to create a method that support municipalities work with social sustainability.

No description available.