1731 research outputs found
Sort by
Mining human Behaviors: automated behavioral Analysis from small to big Data
No full textThis research thesis aims to address complex problems in Human Behavior Understanding from a computational standpoint: to develop novel methods for enabling machines to capture not only what their sensors are perceiving but
also how and why the situation they are presented with is evolving in a certain manner.
Touching several fields, from Computer Vision to Social Psychology through Natural Language Processing and Data Mining, we will move from more to less constrained scenarios, describing models for automated behavioral analysis in different contexts: from the individual perspective, e.g. a user interacting with technology, to the group perspective, e.g. a brainstorming session; from
living labs, e.g. hundreds of people transparently tracked in their everyday life through smart-phone sensors, to the World Wide Web
Dry sliding and contact fatigue behavior of different high density Ni-Cu-Mo PM Steels
No full textThe usage of PM steels in industry, claims for continuous efforts in research and development. For this purpose, the world biggest part manufacturer, GKN Sinter Metals and the Department of Industrial Engineering of University of Trento started a cooperation to investigate the wear and contact fatigue properties of PM steels.
Wear is one of the most important factors that limit the life of engineering components in different ways. In this research two very common phenomena, dry sliding and contact fatigue behavior of PM parts have been investigated.
In the first part of the work dry sliding behavior of Fe-Mo-Cu-C and Fe-Mo-Cu-Ni-C steels were investigated. This study was done to investigate the influence of Ni on dry sliding behavior of PM steels. Ni-rich constituent is expected to decrease wear resistance, due to a localized plastic deformation which reduces the load bearing capacity of the surface. Once the effect of the presence of Ni has been assessed, in the second part of the work dry sliding wear resistance of two Ni-alloyed sintered steels, differing for Ni content, was investigated to highlight any effect due to the amount of this alloying element. Based on experimental results, a procedure for the design of dry sliding wear resistant parts was proposed as well.
In the second part of the work the contact fatigue behavior of three different steels was investigated. The aim was the proposal of a conservative approach to predict the contact fatigue behavior of different sintered and heat treated steels, based on the nucleation of the fatigue crack. The model is developed by calculating the maximum local stress from the theory of the elastic contact and the matrix yield strength from the microhardness of the steels, under different mean Hertzian pressures. Effect of shot peening on contact fatigue behavior was also investigated. Two different types of shot peening, ceramic or steel+ceramic, were considered. Stress and strength calculations were made based on the local approach on contact fatigue, which was validated in the second part of the work
Silk fibroin-based injectable hydrogels for brain tissue engineering applications
No full textStroke and traumatic brain injury are among the leading causes of death in the world. Until now, there are no effective treatments available. Current pharmaceutical treatments have limited benefits to repair the damaged tissue. Brain tissue engineering is a promising strategy to help brain regeneration after the damage induced by stroke or traumatic brain injury. In this thesis, our work focused on designing and evaluating appropriate silk fibroin-based hydrogels combined with stem cells therapy for brain tissue regeneration. The work initially started from looking for appropriate silk fibroin-based hydrogel substrates which can support the viability and neural differentiation of pluripotent cells. Mouse embryonic stem cells (mESC) were used as a model. Different processing procedures of silk fibroin-based hydrogel substrates were prepared by chemical genipin crosslinking and physical sonication crosslinking. The viability and neural differentiation of pluripotent cells on these hydrogel substrates were evaluated, using tissue culture plates (TCP) as control. Different crosslinking processes were found to modulate the neural differentiation of pluripotent cells. Chemical genipin crosslinked hydrogel substrates could inhibit the neural differentiation of mESC compared to control TCP, while the physical sonication crosslinked hydrogel substrates could support the neural differentiation as TCP. According to the results obtained in the first stage, the physically sonication-crosslinked 3D silk fibroin hydrogel was produced to encapsulate human neural stem cells (hNSC). In order to improve the hNSC attachment and neuronal differentiation, the isoleucine-lysine-valine-alanine-valine (IKVAV) peptide derived from laminin was covalently conjugated to the silk fibroin. The viability and neural differentiation of hNSC were evaluated in the unmodified and IKVAV peptide modified silk fibroin hydrogels. We found that the IKVAV peptide modified silk fibroin hydrogel could increase the viability, proliferation and neuronal differentiation of hNSC. Furthermore, the angiogenesis potential of sonication-induced 3D silk fibroin unmodified and modified with IKVAV and a scramble peptide VVIAK (as control) were evaluated in a human outgrowth endothelial cells (OEC) mono-culture system and a co-culture system in which OEC were cultured with human bone marrow mesenchymal stem cells (BM-MSC). Both the silk fibroin unmodified and modified with IKVAV peptide could not induce angiogenesis in the mono-culture system under the VEGF condition. However, in the co-culture system, we found that unmodified, IKVAV-modified and VVIAK-modified silk fibroin hydrogels all could support angiogenesis. Furthermore, there were no significant differences among unmodified, IKVAV modified and VVIAK modified silk fibroin hydrogels influencing on angiogenesis structure and gene expression related to angiogenesis. The thesis will introduce the detailed work in three different chapters (from chapter 3 to chapter 5) respectively
Modulation of Drug Release from Polymeric Carriers and Systems
Get PDFControlled drug delivery systems, which are intended to deliver drugs at predetermined rates for predefined periods of time, have been used to overcome the shortcomings of conventional drug formulations. Injectable drug-loaded matrices and controlled release technology offer numerous advantages compared to conventional dosage. However, one of the greatest challenges in applying this system to the clinical phase is the relatively large initial burst release. To reduce this effect of large initial burst release, a new drug delivery system was fabricated in this thesis. Both alginate and gelatin are biocompatible nature polymers and have been largely used as biomaterials for long time. By cross-linking alginate solution carrying drug-loaded uncross-linked gelatin microbeads, the initial burst release was reduced significantly, compared with drug directly releasing from gelatin or cross-linked alginate matrix. Firstly, a series of gelatin microbeads were prepared in a water-in-oil (W/O) emulsion by a traditional emulsification method. The effects of the concentration of gelatin solution, volumetric ratio of water-to-oil phase, stirring speed and emulsifying time on the particle size and dispersity of gelatin microbeads were studied.
Secondly, drug-loaded gelatin microbeads were encapsulated into the cross-linked sodium alginate macro-beads. The release behavior of drug-loaded gelatin microbeads encapsulated within cross-linked alginate gel was characterized both at room temperature and 37°C and compared with the release from gelatin microbeads and cross-linked alginate gel alone. This system represents a promise for the development of novel and versatile injectable drug delivery systems. Thirdly, a dual-drug delivery system was fabricated by encapsulating drug-loaded gelatin microbeads into the mixture of cross-linkable sodium alginate and another drug. The effects of preparation methods of drug-loaded gelatin microbeads and ratio between gelatin microbeads and alginate on drug release behaviors of both drugs were studied. This system shows a significant potential in dual drug delivery field due to the synergistic effect between gelatin and alginate.
Additionally, combination of multi-drugs in one system has been revealed as a promising application in the drug delivery systems. Therefore, in the fifth chapter of this thesis, an idea for the multi-drug delivery system was simply demonstrated. Rods of gelatin gels loaded with different drugs were separated by the agarose gel in a polycarbonate tube. The effects of the length of agarose gel and temperature on the drug release profiles were investigated. The results suggested the feasibility to employ this idea to the practical applications
Heavy-tailed Phenomena and Tail Index Inference
No full textThis thesis focuses on the analysis of heavy-tailed distributions, which are widely applied to model phenomena in many disciplines. The definition of heavy tails based on the theory of regular variation highlights the importance of the tail index, which indicates the existence of moments and characterises the rate at which the tail decays. Two new approaches to make inference for the tail index are proposed. The first approach employs a regression technique and constructs an estimator of the tail index. It exploits the fact that the behaviour of the characteristic function near the origin reflects the behaviour of the distribution function at infinity. The main advantage of this approach is that it utilises all observations to constitute each point in the regression, not just extreme values. Moreover, the approach does not rely on prior information on the starting point of the tail behaviour of the underlying distribution and shows excellent performance in a wide range of cases: Pareto distributions, heavy-tailed distributions with a non-constant slowly varying factor, and composite distributions with heavy tails. The second approach is motivated by the asymptotic properties of a special moment statistic, the so-called partition function. This statistic considers blocks of data and is generally used in the context of multifractality. Due to the interplay between the weak law of large numbers and the generalised central limit theorem, the asymptotic behaviour of the partition function is strongly affected by the existence of moments even for weakly dependent samples. Via a quantity, the scaling function, a graphical method to identify the existence of heavy tails is proposed. Moreover, the plot of the scaling function allows one to make inference for the underlying distribution: with infinite variance, finite variance with tail index larger than two, or all moments finite. Furthermore, since the tail index is reflected at the breakpoint of the plot of the scaling function, this gives the possibility to estimate the tail index. Both these two approaches use the entire distribution, not just the tail, to analyse the tail behaviour. This sheds a new light on the analysis of heavy-tailed distributions. At the end of this thesis, these two approaches are used to detect power laws in empirical data sets from a variety of fields and contribute to the debate on whether city sizes are better approximated by a power law or a log-normal distribution
Computational techniques for nonlinear codes and Boolean functions
Get PDFWe present some upper bounds on the size of nonlinear codes and their restriction to systematic codes and linear codes. These bounds, which are an improvement of a bound by Zinoviev, Litsyn and Laihonen, are independent of other classical known theoretical bounds. Among these, we mention the Griesmer bound for linear codes, of which we provide a partial generalization for the systematic case. Our experiments show that in some cases (the majority of cases for some q) our bounds provide the best value, compared to all other closed-formula upper-bounds. We also present an algebraic method for computing the minimum weight of any nonlinear code. We show that for some particular code, using a non-standard representation of the code, our method is faster than brute force. An application of this method allows to compute the nonlinearity of a Boolean function, improving existing algebraic methods and reaching the same complexity of algorithms based on the fast Fourier transform
Bio-physical controls on tidal network geomorphology
No full textLooking over a tidal wetland, the tidal network characterised by its intricate system of bifurcating, blind-ended tidal courses clearly stands out from the overall landscape. This tidal landform exerts a fundamental control on the morphology and ecology within the tidal environment. With today’s recognition of the ecological, economical and societal values provided by tidal wetlands, which has been notably reflected in the development of restoration management strategies across Europe and USA, there is a need to fully understand the nature and development of tidal networks as well as their relationships with associated landforms and biotic components (e.g. vegetation), to eventually guarantee the success of current and future restoration practices. Accordingly, this research aims to bring further insights into the bio-physical controls on the geomorphology of tidal networks. To this end, a combination of remote sensing, modelling and field activities was employed. A geo-spatial analysis was performed at Queen Mary, University of London (UK), to address the variability of tidal network patterns. A series of network scale morphometric variables was extracted using airborne LiDAR data among selected tidal networks across the UK depicting different planview morphologies, and supplemented with the collection of corresponding marsh scale environmental variables from published sources. Multivariate statistics were then performed to characterise the variability of tidal network patterns and identify the inherent environmental controls. The analysis has revealed that every network type can be characterised based upon measures of network size and complexity, with each network pattern depicting proper morphometric aspects. Particularly, the stream Strahler order and the median depth of the network main channel have the highest discriminating weight on the patterns investigated. High correlation between the latter variable and network main channel width has revealed that linear, linear-dendritic and dendritic networks followed a transitional gradient in their aspect ratio approximated by a power law and thus are seen to depict similar erosional processes. To the contrary, meandering networks clearly depart from this relationship, and show particular segregation in their aspect ratios with respect to dendritic networks. Globally, differentiation on network morphometric properties has been linked to environmental conditions specific to the marsh physiographic setting within which a tidal network develops. Conceptually, tidal networks seem to adapt to marsh environmental conditions by adopting suitable morphologies to drain their tidal basin effectively.An eco-geomorphic modelling framework was developed at University of Trento (Italy), to address tidal network morphological development. In line with current theories as well as modelling advances and challenges in the field of tidal network ontogeny, emphasis was thus placed on the investigation of tidal channel formation and evolution in progressive marsh accretional context. Under these environmental conditions, tidal network development can be ascribed to the combination of two channel-forming processes: channel initiation results from bottom incisions in regions where topographic depressions occur; channel elaboration results from differential deposition, contributing to the deepening of the tidal channels relative to the adjacent marsh platform. Further evolutionary stages including channel reduction proceed from the horizontal progradation of the marsh platform which may lead eventually to channel infilling. Moreover, both qualitative and quantitative results allude to an acceleration of the morphological development of the synthetic tidal networks with increasing sediment supply. These different observations thus emphasise the prevalence of depositional processes in shaping tidal channels. In a second stage, the investigation was extended to the role of the initial tidal flat morphology as an inherent control on tidal network development, by considering different scenarios of topographic perturbations, which has revealed its legacy on tidal network morphological features. Modelling experiments have also acknowledged salt marsh macrophytes as a potential control on network evolution depending on their biomass distribution within the tidal frame. However, tidal channel morphodynamcis appears to be sensitive to the way biomass growth is mathematically parameterised in the model. In view of the current challenges in transcribing mathematically such a dynamic process and the relevance of bio-physical interactions in driving salt marsh and tidal network evolution, a field survey was conducted in a temperate salt marsh in the Netherlands, as part of the mobility to UNESCO-IHE (Netherlands) in partnership with University of Antwerp (Belgium), to assess vegetation distribution and productivity in the tidal frame. Particularly, emphasis was placed on extending investigations on the possible presence of relationships involving vegetation properties in different climatic and ecological conditions from those characterising these previously documented relationships. Regression analysis has revealed that biomass growth can be expressed as a linear function of marsh relative elevation, providing therefore direct empirical validation for corresponding assumptions reported in the literature and used in the present modelling framework; surprisingly, that increase did not correlate with an increase in species richness and diversity. Analysis of likely associations between vegetation morphometrics and total standing biomass yielded only a single linear relationship linking the latter variable to stem height. In truth, these observations may bear reconsiderations on the global validity of the assumptions used in the formulation of some eco-geomorphic processes which are applied in the study and prediction of wetland resiliency facing climate change
Psychological and neural mechanisms of stay/leave decision making
No full textThis thesis describes three behavioral and one fMRI study in which the underlying psychological and neural mechanisms of stay/leave decision making in both a nonsocial and social context were investigated and compared. In Study 1A and 1B participants played a social or nonsocial version of a novel economic game in which participants were dependent on a social partner or nonsocial resource to increase their monetary outcomes. Both actual and expected reward probability of social partners and nonsocial resources positively affected the time participants decided to stay with a specific social partner or nonsocial resource. The effect of expected reward probability on staying times was much more pronounced in the nonsocial context. In Study 2, Reinforcement Learning models were fitted to participant's choices in a social and nonsocial version of the 4-armed bandit task in which prior beliefs about partners or resources were manipulated. The effects of Study 1B were replicated. Furthermore, it was found that participants weight expectancy-consistent and inconsistent new observations differentially when learning about both social partners and nonsocial resources. Participants had generally higher learning rates in the social than nonsocial context but this was not the case when prior beliefs were not induced. In Study 3, participants played a social and nonsocial version of the 4-armed bandit task while undergoing fMRI. Increased activation in caudate nucleus was associated with stay decisions in a social rather than nonsocial context; and the dorsal anterior cingulate cortex showed differential activation as a function of stay and leave decisions and this pattern was reversed in the nonsocial and social context. Interpretation of these results are discussed in light of a fundamental need to belong that motivates people to stay with social partners specifically and resist against breaking social connections
Complete Arcs and Caps in Galois Spaces
Get PDFGalois spaces, that is affine and projective spaces of dimension N ≥ 2 defined over a finite (Galois) field F_q, are well known to be rich of nice geometric, combinatorial
and group theoretic properties that have also found wide and relevant applications in several branches of Combinatorics, as well as in more practical areas, notably Coding Theory and Cryptography. The systematic study of Galois spaces was initiated in the late 1950’s by the pioneering work of B. Segre [59]. The trilogy [34, 36, 42] covers the general theory of Galois spaces including the study of objects which are linked to linear codes. Typical such objects are plane arcs and their generalizations - especially caps and arcs in higher dimensions - whose code theoretic counterparts are distinguished types of error-correcting and covering linear codes. Their investigation has received a great stimulus from Coding Theory, especially in the last decades; see the survey papers [40, 41]. An important issue in this context is to ask for explicit constructions of small complete arcs and small complete caps. A cap in a Galois space is a set of points no three of which are collinear. A cap is complete if its secants (lines through two points of the set) cover the whole space. An arc in a Galois space of dimension N is a set of points no N+1 of which lying on the same hyperplane. In analogy with caps, an arc which is maximal with respect to set-theoretical inclusion is said to be complete.
Also, arcs coincide with caps in Galois planes. From these geometrical objects, there arise linear codes which turn out to have very good covering properties, provided
that the size of the set is small with respect to the dimension N and the order q of the ambient space.
For the size t(AG(N,q)) of the smallest complete caps in a Galois affine space AG(N,q) of dimension N over F_q, the trivial lower bound is √2q^{N−1/2}. General constructions of complete caps whose size is close to this lower bound are only known for q even and N odd, see [16, 19, 29, 52]. When N is even, complete caps of size of the same order of magnitude as cq^{N/2}, with c a constant independent of q, are known to exist for both the odd and the even order case, see [16, 18, 28, 29, 31] (see also Section 2.2 and the references therein). Whereas, few constructions of small complete arcs in Galois spaces of dimension N>2 are known. In [65, 66, 67], small complete arcs having many points in common with the normal rational curve are investigated (see Section 4.2.3 for comparisons with our results). In this thesis, new infinite families of complete arcs and caps in higher dimensional spaces are constructed from algebraic curves defined over a finite field. In most
cases, no smallest complete caps/arcs were previously known in the literature. Although caps and arcs are rather combinatorial objects, constructions and proofs sometimes heavily rely on concepts and results from Algebraic Geometry in positive characteristic
Quaternionic slice regular functions on domains without real points
No full textIn this thesis I've explored the theory of quaternionic slice regular functions. More precisely I've studied some rigidity properties, some differential issues and an application in
complex differential geometry.
This application concerns the constructions and classifications of orthogonal complex structures on open domains of the four dimensional euclidean space