LDA-TM : A two-step approach to twitter topic data clustering

The Twitter System is the biggest social network in the world, and everyday millions of tweets are posted and talked about, expressing various views and opinions. A large variety of research activities have been conducted to study how the opinions can be clustered and analyzed, so that some tendencies can be uncovered. Due to the inherent weaknesses of the tweets - very short texts and very informal styles of writing - it is rather hard to make an investigation of tweet data analysis giving results with good performance and accuracy. In this paper, we intend to attack the problem from another aspect - using a two-layer structure to analyze the twitter data: LDA with topic map modelling. The experimental results demonstrate that this approach shows a progress in twitter data analysis. However, more experiments with this method are expected in order to ensure that the accurate analytic results can be maintained.

A New Clustering Methodology for the Analysis of Sorted or Categorized Stimuli

This paper introduces a new stochastic clustering methodology devised for the analysis of categorized or sorted data. The methodology reveals consumers' common category knowledge as well as individual differences in using this knowledge for classifying brands in a designated product class. A small study involving the categorization of 28 brands of U.S. automobiles is presented where the results of the proposed methodology are compared with those obtained from KMEANS clustering. Finally, directions for future research are discussed.

The Little Galaxies That Could Reionize the Universe

Thesis (Ph.D.)--University of Washington, 2016-08

De la métallicité et la formation stellaire des galaxies spirales barrées

Afin d’étudier l’historique de formation stellaire et d’enrichissement chimique des galaxies spirales barrées, j’ai simulé l’évolution de 27 galaxies spirales, barrées et non barrées, de diverses masses stellaires et fractions de gaz. Alors que les galaxies non barrées présentent une évolution lente et continue sur les deux milliards d’années que durent nos simulations, les galaxies barrées ont une évolution bien plus explosive, et ce particulièrement pour les galaxies les plus massives. Dans un premier temps, je montre que la présence de la barre entraine un flot important de gaz des régions périphériques vers le centre de la galaxie barrée, causant un sursaut de formation stellaire et une croissance importante de l’abondance chimique centrale, et que l’amplitude et la vitesse à laquelle ce sursaut arrive augmentent avec la masse de la galaxie. Cet épisode de sursaut stellaire entraine alors une diminution importante de la masse de gaz, entrainant à son tour une décroissance de la formation stellaire et une stagnation de l’enrichissement chimique pour le reste de l’évolution de la galaxie. Dans un deuxième temps, je montre qu’à cause de la dynamique en deux périodes très différentes des galaxies barrées, deux galaxies de masse très semblable peuvent avoir des taux de formation stellaire et des métallicités complètement différentes en fonction de leur stade évolutif, stade qu’on ne peut déterminer aisément. Cette difficulté est tout aussi importante lorsqu’on compare le coeur des galaxies barrées et non barrées entre elles, étant donné que des coeurs comparables sont situés dans les galaxies très différentes, et que des galaxies semblables ont des coeurs très différents.

Structure morphologique et populations stellaires d’un échantillon de galaxies spirales

Afin de caractériser la structure morphologique et les populations stellaires d’un échantillon de treize galaxies spirales, j’ai analysé des images WISE et GALEX, et j’ai construit des diagrammes magnitude-magnitude et couleur-magnitude pixel à pixel. Les diagrammes présentent des groupes de pixels qui correspondent spatialement aux composantes structurales des galaxies. Les diagrammes ainsi que les profils radiaux de brillance de surface indiquent que les variations de la densité surfacique de masse de la vieille population stellaire jouent un rôle important dans la différenciation des structures. On estime l’âge des jeunes complexes stellaires et l’extinction dans ces galaxies en les comparant à des modèles de populations stellaires simples nées de sursauts de formation stellaire instantanée. L’étude de ces propriétés est possible grâce à la combinaison des données ultraviolettes et infrarouge et à la grande sensibilité de la couleur ultraviolette à la variation de l’âge. On observe un gradient d’extinction dont la pente est liée à la présence d’une barre ou d’une activité nucléaire : en effet, l’extinction décroît avec la distance galactocentrique et la pente est plus petite pour les galaxies ayant une barre ou une activité nucléaire. On observe également un gradient d’âge où les régions externes sont moins évoluées que celles du centre sauf pour les galaxies de type tardif.

Search Based Clustering for Protecting Software with Diversified Updates

Reverse engineering is usually the stepping stone of a variety of at-tacks aiming at identifying sensitive information (keys, credentials, data, algo-rithms) or vulnerabilities and flaws for broader exploitation. Software applica-tions are usually deployed as identical binary code installed on millions of com-puters, enabling an adversary to develop a generic reverse-engineering strategy that, if working on one code instance, could be applied to crack all the other in-stances. A solution to mitigate this problem is represented by Software Diversity, which aims at creating several structurally different (but functionally equivalent) binary code versions out of the same source code, so that even if a successful attack can be elaborated for one version, it should not work on a diversified ver-sion. In this paper, we address the problem of maximizing software diversity from a search-based optimization point of view. The program to protect is subject to a catalogue of transformations to generate many candidate versions. The problem of selecting the subset of most diversified versions to be deployed is formulated as an optimisation problem, that we tackle with different search heuristics. We show the applicability of this approach on some popular Android apps.

Nuclei of Nearby Active Galaxies: NIR and Sub-mm Views and Finalizing the LINC-NIRVANA Fringe and Flexure Tracking System: Flexure and Temperature Behavior

The work presented in my thesis addresses the two cornerstones of modern astronomy: Observation and Instrumentation. Part I deals with the observation of two nearby active galaxies, the Seyfert 2 galaxy NGC 1433 and the Seyfert 1 galaxy NGC 1566, both at a distance of $\sim10$ Mpc, which are part of the Nuclei of Galaxies (NUGA) sample. It is well established that every galaxy harbors a super massive black hole (SMBH) at its center. Furthermore, there seems to be a fundamental correlation between the stellar bulge and SMBH masses. Simulations show that massive feedback, e.g., powerful outflows, in Quasi Stellar Objects (QSOs) has an impact on the mutual growth of bulge and SMBH. Nearby galaxies follow this relation but accrete mass at much lower rates. This gives rise to the following questions: Which mechanisms allow feeding of nearby Active Galactic Nuclei (AGN)? Is this feeding triggered by events, e.g., star formation, nuclear spirals, outflows, on $\sim500$ pc scales around the AGN? Does feedback on these scales play a role in quenching the feeding process? Does it have an effect on the star formation close to the nucleus? To answer these questions I have carried out observations with the Spectrograph for INtegral Field Observation in the Near Infrared (SINFONI) at the Very Large Telescope (VLT) situated on Cerro Paranal in Chile. I have reduced and analyzed the recorded data, which contain spatial and spectral information in the H-band ($1.45 \mic-1.85 \mic$) and K-band ($1.95 \mic-2.45 \mic$) on the central $10\arcsec\times10\arcsec$ of the observed galaxies. Additionally, Atacama Large Millimeter/Sub-millimeter Array (ALMA) data at $350$ GHz ($\sim0.87$ mm) as well as optical high resolution Hubble Space Telescope (HST) images are used for the analysis. For NGC 1433 I deduce from comparison of the distributions of gas, dust, and intensity of highly ionized emission lines that the galaxy center lies $\sim70$ pc north-northwest of the prior estimate. A velocity gradient is observed at the new center, which I interpret as a bipolar outflow, a circum nuclear disk, or a combination of both. At least one dust and gas arm leads from a $r\sim200$ pc ring towards the nucleus and might feed the SMBH. Two bright warm H$_2$ gas spots are detected that indicate hidden star formation or a spiral arm-arm interaction. From the stellar velocity dispersion (SVD) I estimate a SMBH mass of $\sim1.74\times10^7$ \msol. For NGC 1566 I observe a nuclear gas disk of $\sim150$ pc in radius with a spiral structure. I estimate the total mass of this disk to be $\sim5.4\times10^7$ \msol. What mechanisms excite the gas in the disk is not clear. Neither can the existence of outflows be proven nor is star formation detected over the whole disk. On one side of the spiral structure I detect a star forming region with an estimated star formation rate of $\sim2.6\times10^{-3}$ \msol\ yr$^{-1}$. From broad Br$\gamma$ emission and SVD I estimate a mean SMBH mass of $\sim5.3\times10^6$ \msol\ with an Eddington ratio of $\sim2\times10^{-3}$. Part II deals with the final tests of the Fringe and Flexure Tracker (FFTS) for LBT INterferometric Camera and the NIR/Visible Adaptive iNterferometer for Astronomy (LINC-NIRVANA) at the Large Binocular Telescope (LBT) in Arizona, USA, which I conducted. The FFTS is the subsystem that combines the two separate beams of the LBT and enables near-infrared interferometry with a significantly large field of view. The FFTS has a cryogenic system and an ambient temperature system which are separated by the baffle system. I redesigned this baffle to guarantee the functionality of the system after the final tests in the Cologne cryostat. The redesign did not affect any scientific performance of LINC-NIRVANA. I show in the final cooldown tests that the baffle fulfills the temperature requirement and stays $<110$ K whereas the moving stages in the ambient system stay $>273$ K, which was not given for the old baffle design. Additionally, I test the tilting flexure of the whole FFTS and show that accurate positioning of the detector and the tracking during observation can be guaranteed.

A fuzzy clustering approach to a demand response model

This paper proposes a novel demand response model using a fuzzy subtractive cluster approach. The model development provides support to domestic consumer decisions on controllable loads management, considering consumers' consumption needs and the appropriate load shape or rescheduling in order to achieve possible economic benefits. The model based on fuzzy subtractive clustering method considers clusters of domestic consumption covering an adequate consumption range. Analysis of different scenarios is presented considering available electric power and electric energy prices. Simulation results are presented and conclusions of the proposed demand response model are discussed. (C) 2016 Elsevier Ltd. All rights reserved.

Clustering through post inhibitory rebound in synaptically coupled neurons

Post inhibitory rebound is a nonlinear phenomenon present in a variety of nerve cells. Following a period of hyper-polarization this effect allows a neuron to fire a spike or packet of spikes before returning to rest. It is an important mechanism underlying central pattern generation for heartbeat, swimming and other motor patterns in many neuronal systems. In this paper we consider how networks of neurons, which do not intrinsically oscillate, may make use of inhibitory synaptic connections to generate large scale coherent rhythms in the form of cluster states. We distinguish between two cases i) where the rebound mechanism is due to anode break excitation and ii) where rebound is due to a slow T-type calcium current. In the former case we use a geometric analysis of a McKean type model to obtain expressions for the number of clusters in terms of the speed and strength of synaptic coupling. Results are found to be in good qualitative agreement with numerical simulations of the more detailed Hodgkin-Huxley model. In the second case we consider a particular firing rate model of a neuron with a slow calcium current that admits to an exact analysis. Once again existence regions for cluster states are explicitly calculated. Both mechanisms are shown to prefer globally synchronous states for slow synapses as long as the strength of coupling is sufficiently large. With a decrease in the duration of synaptic inhibition both systems are found to break into clusters. A major difference between the two mechanisms for cluster generation is that anode break excitation can support clusters with several groups, whilst slow T-type calcium currents predominantly give rise to clusters of just two (anti-synchronous) populations.

Hierarchical clustering with multiple-height branch-cut applied to short time-series gene expression data.

Rigid adherence to pre-specified thresholds and static graphical representations can lead to incorrect decisions on merging of clusters. As an alternative to existing automated or semi-automated methods, we developed a visual analytics approach for performing hierarchical clustering analysis of short time-series gene expression data. Dynamic sliders control parameters such as the similarity threshold at which clusters are merged and the level of relative intra-cluster distinctiveness, which can be used to identify "weak-edges" within clusters. An expert user can drill down to further explore the dendrogram and detect nested clusters and outliers. This is done by using the sliders and by pointing and clicking on the representation to cut the branches of the tree in multiple-heights. A prototype of this tool has been developed in collaboration with a small group of biologists for analysing their own datasets. Initial feedback on the tool has been positive.

Neuropsychiatric disease (NPD) clustering in families with Autism Spectrum Disorder (ASD): genetic, epigenetic and environmental issues

Autism Spectrum Disorder (ASD) is a neurodevelopmental disorder characterized by deficits in social communication/interaction and by unusual repetitive and restricted behaviors and interests. ASD often co-occurs in the same families with other neuropsychiatric diseases (NPD), such as intellectual disability, schizophrenia, epilepsy, depression and attention deficit hyperactivity disorder. Genetic factors have an important role in ASD etiology. Multiple copy number variants (CNVs) and single nucleotide variants (SNVs) in candidate genes have been associated with an increased risk to develop ASD. Nevertheless, recent heritability estimates and the high genotypic and phenotypic heterogeneity characteristic of ASD indicate a role of environmental and epigenetic factors, such as long noncoding RNA (lncRNA) and microRNA (miRNA), as modulators of genetic expression and further clinical presentation. Both miRNA and lncRNA are functional RNA molecules that are transcribed from DNA but not translated into proteins, instead they act as powerful regulators of gene expression. While miRNA are small noncoding RNAs with 22-25 nucleotides in length that act at the post-transcriptional level of gene expression, the lncRNA are bigger molecules (>200 nucleotides in length) that are capped, spliced, and polyadenylated, similar to messenger RNA. Although few lncRNA were well characterized until date, there is a great evidence that they are implicated in several levels of gene expression (transcription/post-transcription/post-translation, organization of protein complexes, cell– cell signaling as well as recombination) as shown in figure 1.

Examining the growth and evolution of magnetic fields in clusters of galaxies: a numerical perspective

Magnetic fields are ubiquitous in galaxy cluster atmospheres and have a variety of astrophysical and cosmological consequences. Magnetic fields can contribute to the pressure support of clusters, affect thermal conduction, and modify the evolution of bubbles driven by active galactic nuclei. However, we currently do not fully understand the origin and evolution of these fields throughout cosmic time. Furthermore, we do not have a general understanding of the relationship between magnetic field strength and topology and other cluster properties, such as mass and X-ray luminosity. We can now begin to answer some of these questions using large-scale cosmological magnetohydrodynamic (MHD) simulations of the formation of galaxy clusters including the seeding and growth of magnetic fields. Using large-scale cosmological simulations with the FLASH code combined with a simplified model of the acceleration of cosmic rays responsible for the generation of radio halos, we find that the galaxy cluster frequency distribution and expected number counts of radio halos from upcoming low-frequency sur- veys are strongly dependent on the strength of magnetic fields. Thus, a more complete understanding of the origin and evolution of magnetic fields is necessary to understand and constrain models of diffuse synchrotron emission from clusters. One favored model for generating magnetic fields is through the amplification of weak seed fields in active galactic nuclei (AGN) accretion disks and their subsequent injection into cluster atmospheres via AGN-driven jets and bubbles. However, current large-scale cosmological simulations cannot directly include the physical processes associated with the accretion and feedback processes of AGN or the seeding and merging of the associated SMBHs. Thus, we must include these effects as subgrid models. In order to carefully study the growth of magnetic fields in clusters via AGN-driven outflows, we present a systematic study of SMBH and AGN subgrid models. Using dark-matter only cosmological simulations, we find that many important quantities, such as the relationship between SMBH mass and galactic bulge velocity dispersion and the merger rate of black holes, are highly sensitive to the subgrid model assumptions of SMBHs. In addition, using MHD calculations of an isolated cluster, we find that magnetic field strengths, extent, topology, and relationship to other gas quantities such as temperature and density are also highly dependent on the chosen model of accretion and feedback. We use these systematic studies of SMBHs and AGN inform and constrain our choice of subgrid models, and we use those results to outline a fully cosmological MHD simulation to study the injection and growth of magnetic fields in clusters of galaxies. This simulation will be the first to study the birth and evolution of magnetic fields using a fully closed accretion-feedback cycle, with as few assumptions as possible and a clearer understanding of the effects of the various parameter choices.

Clustering of variables with a three-way approach for health sciences

© 2014 Cises This work is distributed with License Creative Commons Attribution-Non commercial-No derivatives 4.0 International (CC BY-BC-ND 4.0)