No description available.

The amount of data generated is on rise due to increased demand for fields like IoT, smart monitoring applications, etc. Data generated through such systems have many distinct characteristics like continuous data generation, evolutionary, multi-source nature, and heterogeneity. In addition, the real-world data generated in these fields is largely unlabelled. Clustering is an unsupervised learning technique used to group, analyze and interpret unlabelled data. Conventional clustering algorithms are not suitable for dealing with data having previously mentioned characteristics due to memory and computational constraints, their inability to handle concept drift, distributed location of data. Therefore novel clustering approaches capable of analyzing and interpreting evolving and/or multi-source streaming data are needed.  The thesis is focused on building evolutionary clustering algorithms for data that evolves over time. We have initially proposed an evolutionary clustering approach, entitled Split-Merge Clustering (Paper I), capable of continuously updating the generated clustering solution in the presence of new data. Through the progression of the work, new challenges have been studied and addressed. Namely, the Split-Merge Clustering algorithm has been enhanced in Paper II with new capabilities to deal with the challenges of multi-view data applications. A multi-view or multi-source data presents the studied phenomenon/system from different perspectives (views), and can reveal interesting knowledge that is not visible when only one view is considered and analyzed. This has motivated us to continue in this direction by designing two other novel multi-view data stream clustering algorithms. The algorithm proposed in Paper III improves the performance and interpretability of the algorithm proposed in Paper II. Paper IV introduces a minimum spanning tree based multi-view clustering algorithm capable of transferring knowledge between consecutive data chunks, and it is also enriched with a post-clustering pattern-labeling procedure.  The proposed and studied evolutionary clustering algorithms are evaluated on various data sets. The obtained results have demonstrated the robustness of the algorithms for modeling, analyzing, and mining evolving data streams. They are able to adequately adapt single and multi-view clustering models by continuously integrating newly arriving data.

L'objectif de ce mémoire est de comprendre le fonctionnement hydro-sédimentaire des embouchures tidales dominées par la marée et la construction d'une flèche sableuse à crochets perpendiculaires. Deux sites voisins ont été choisis : les embouchures de l'Orne et de la Dives sur la côte du Calvados (Normandie). L'utilisation de données d'agitation et topographiques, acquises selon plusieurs échelles spatio-temporelles, permet d'étudier la morphodynamique de ces deux embouchures.L'embouchure de l'Orne possède une évolution soumise à une forte pression anthropique causée par de nombreuses infrastructures portuaires. Elle est comparée à celle de l'embouchure de la Dives, où l'impact anthropique est plus limité. Il apparaît que ces deux embouchures partagent les mêmes mécanismes de transports sédimentaires, avec une composante transversale importante. Ce transport vers la côte est assuré par la migration de barres de swash sur les parties aval-dérive des deltas de jusant. Ces barres de swash possèdent des volumes et des vitesses de migration similaires, comprises entre 3.5 et 8.0 m/mois.Contrairement à la Dives, les côtes en amont et en aval-dérive de l'embouchure de l'Orne sont en accrétion. Si en amont, cette accrétion suit les nombreuses phases de construction du port de Caen-Ouistreham, en aval se développe depuis un peu plus d'un siècle une flèche sableuse à la morphologie atypique. En effet, cette flèche possède plusieurs crochons emboités, dont les extrémités - les crochets - sont perpendiculaires au corps à la flèche. Il est démontré que l'attachement des barres de swash à la côte est responsable de la formation d'un nouveau crochon. La diffraction des vagues au niveau de l'extrémité des barres de swash et des crochons, ainsi que les directions locales de l'agitation, expliquent la formation des crochets perpendiculaires à l'extrémité des crochons. / The aim of this thesis is to understand the mechanisms of sediment bypassing at tidal inlets in a macrotidal environment and the formation of a sandspit with perpendicular hooks. Two closed areas are studied, the Orne and Dives inlets on the coast of Normandy (France). Wave and topographic data, at several space-time scales, are used to study the inlet morphodynamics.The Orne inlet is affected by strong human impacts due to several port facilities whereas the Dives inlet knows a limited anthropogenic impact. The study of the two inlets allows to deduce the impact of the human activities in the sediment transport patterns. The Orne and Dives inlets share the same mechanisms of sediment transport. Both show a major cross-shore component of the transport, underlines by the swash bars migration over the ebb-tidal deltas. All swash bars have similar volumes and migration rates, between 3.5 and 8.0 m/month.The Orne inlet shows deposition on both sides. The accretion on the updrift coast follows the building stages of the harbour of Caen-Ouistreham. On the downdrift coast, the deposition creates the Merville spit, which has several interlocked ridges. A perpendicular hook is present on the distal part of each ridge. This work shows that the swash bar attachment at the shoreline shapes a new ridge. Wave diffraction around the distal part of the swash bars and ridges, and the local wave directions, explain the formation of perpendicular hooks.

abstract: Semiconductor devices often face reliability issues due to their operational con- ditions causing performance degradation over time. One of the root causes of such degradation is due to point defect dynamics and time dependent changes in their chemical nature. Previously developed Unified Solver was successful in explaining the copper (Cu) metastability issues in cadmium telluride (CdTe) solar cells. The point defect formalism employed there could not be extended to chlorine or arsenic due to numerical instabilities with the dopant chemical reactions. To overcome these shortcomings, an advanced version of the Unified Solver called PVRD-FASP tool was developed. This dissertation presents details about PVRD-FASP tool, the theoretical framework for point defect chemical formalism, challenges faced with numerical al- gorithms, improvements for the user interface, application and/or validation of the tool with carefully chosen simulations, and open source availability of the tool for the scientific community. Treating point defects and charge carriers on an equal footing in the new formalism allows to incorporate chemical reaction rate term as generation-recombination(G-R) term in continuity equation. Due to the stiff differential equations involved, a reaction solver based on forward Euler method with Newton step is proposed in this work. The Jacobian required for Newton step is analytically calculated in an elegant way improving speed, stability and accuracy of the tool. A novel non-linear correction scheme is proposed and implemented to resolve charge conservation issue. The proposed formalism is validated in 0-D with time evolution of free carriers simulation and with doping limits of Cu in CdTe simulation. Excellent agreement of light JV curves calculated with PVRD-FASP and Silvaco Atlas tool for a 1-D CdTe solar cell validates reaction formalism and tool accuracy. A closer match with the Cu SIMS profiles of Cu activated CdTe samples at four different anneal recipes to the simulation results show practical applicability. A 1D simulation of full stack CdTe device with Cu activation at 350C 3min anneal recipe and light JV curve simulation demonstrates the tool capabilities in performing process and device simulations. CdTe device simulation for understanding differences between traps and recombination centers in grain boundaries demonstrate 2D capabilities. / Dissertation/Thesis / Doctoral Dissertation Electrical Engineering 2019

A data warehouse (DW) is a transformation of many sources of transactional data integrated into a single collection that is non-volatile and time-variant that can provide decision support to managerial roles within an organization. For this application, the database server needs to process multiple users’ queries by joining various datasets and loading the result in main memory to begin calculations. In current systems, this process is reactionary to users’ input and can be undesirably slow. In previous studies, it was shown that a personalization scheme of a single user’s query patterns and loading the smaller subset into main memory the query response time significantly shortened the query response time. The LPCDA framework developed in this research handles multiple users’ query demands, and the query patterns are subject to change (so-called concept drift) and noise. To this end, the LPCDA framework detects changes in user behaviour and dynamically adapts the personalized smart cube definition for the group of users. Numerous data mart (DM)s, as components of the DW, are subject to intense aggregations to assist analytics at the request of automated systems and human users’ queries. Subsequently, there is a growing need to properly manage the supply of data into main memory that is in closest proximity to the CPU that computes the query in order to reduce the response time from the moment a query arrives at the DW server. As a result, this thesis proposes an end-to-end adaptive learning ensemble for resource allocation of cuboids within a a DM to achieve a relevant and timely constructed smart cube before the time in need, as a way of adopting the just-in-time inventory management strategy applied in other real-world scenarios. The algorithms comprising the ensemble involve predictive methodologies from Bayesian statistics, data mining, and machine learning, that reflect the changes in the data-generating process using a number of change detection algorithms. Therefore, given different operational constraints and data-specific considerations, the ensemble can, to an effective degree, determine the cuboids in the lattice of a DM to pre-construct into a smart cube ahead of users submitting their queries, thereby benefiting from a quicker response than static schema views or no action at all.

This study examines the post-earnings announcement drift (PEAD) anomaly on the Swedish stock market. By constructing a corporate governance index based on share structure, board independence and board gender diversity, we test how the quality of firms’ corporate governance affects the drift – a link which is previously unexplored. We find no evidence of PEAD for firms with good corporate governance, while firms with bad corporate governance do experience a drift. Furthermore, a PEAD trading strategy based on bad governance firms yields significantly larger abnormal returns compared to the corresponding trading strategy for good governance firms. Our results are robust to controlling for the risk factors of the Fama-French 3-factor model. The findings support that investors tend to underreact to extreme earnings surprises reported by bad governance firms due to a higher degree of information uncertainty, while the stock price reactions are more complete for good governance firms.

A Wireless Sensor Network (WSN) is a collection of tiny devices called sensor nodes which are deployed in an area to be monitored. Each node has one or more sensors with which they can measure the characteristics of their surroundings. In a typical WSN, the data gathered by each node is sent wirelessly through the network from one node to the next towards a central base station. Each node typically has a very limited energy supply. Therefore, in order for WSNs to have acceptable lifetimes, energy efficiency is a design goal that is of utmost importance and must be kept in mind at all levels of a WSN system. The main consumer of energy on a node is the wireless transceiver and therefore, the communications that occur between nodes should be carefully controlled so as not to waste energy. The Medium Access Control (MAC) protocol is directly in charge of managing the transceiver of a node. It determines when the transceiver is on/off and synchronizes the data exchanges among neighbouring nodes so as to prevent collisions etc., enabling useful communications to occur. The MAC protocol thus has a big impact on the overall energy efficiency of a node. Many WSN MAC protocols have been proposed in the literature but it was found that most were not optimized for the group of WSNs displaying very low volumes of traffic in the network. In low traffic WSNs, a major problem faced in the communications process is clock drift, which causes nodes to become unsynchronized. The MAC protocol must overcome this and other problems while expending as little energy as possible. Many useful WSN applications show low traffic characteristics and thus a new MAC protocol was developed which is aimed at this category of WSNs. The new protocol, Dynamic Preamble Sampling MAC (DPS-MAC) builds on the family of preamble sampling protocols which were found to be most suitable for low traffic WSNs. In contrast to the most energy efficient existing preamble sampling protocols, DPS-MAC does not cater for the worst case clock drift that can occur between two nodes. Rather, it dynamically learns the actual clock drift experienced between any two nodes and then adjusts its operation accordingly. By simulation it was shown that DPS-MAC requires less protocol overhead during the communication process and thus performs more energy efficiently than its predecessors under various network operating conditions. Furthermore, DPS-MAC is less prone to become overloaded or unstable in conditions of high traffic load and high contention levels respectively. These improvements cause the use of DPS-MAC to lead to longer node and network lifetimes, thus making low traffic WSNs more feasible. / Dissertation (MEng)--University of Pretoria, 2008. / Electrical, Electronic and Computer Engineering / MEng / Unrestricted

Implementeringen av tablets ämnade för inspektioner inom drift och underhåll av vägar har studerats och utvärderats med målet att fördjupa kunskaperna inom drift- och underhållsbranschen och skapa effektivare arbetsmetoder som gynnar alla byggprocessens parter.  Examensarbetet är utfört hos Peab på deras anläggningskontor i Sundsvall och implementeringen har ägt rum på deras driftområde Sundsvall som består av cirka 122 mil statlig väg där Trafikverket är beställare. Genom litteraturstudier, tester av tableten och deltagande observationer så har tabletens användbarhet och utvecklingspotential analyserats och slutsatser har dragits för att ge Peab feedback på deras innovation med ett antal förbättringsförslag. Utvärderingen blev ej fullständig då Peab inte hann införa tabletsen inom vägdriften innan de 10 veckorna för examensarbetet var avslutade. Att använda tabletsen inom vägdriften kommer att effektivisera inspektionerna med dels digital dokumenthantering och det viktiga införande av GPS positionering  i de fordon som utför inspektionerna. Detta kommer bli mycket smidigare än den tidigare arbetsmetodiken där i stort sett alla processer innefattande mycket onödigt pappersarbete. Peab rekommenderas fortsätta att se utvecklingsmöjligheter med tabletsen inom andra områden av vägdriften samt fullfölja utvecklingen av tableten för att få en användbar produkt. / The implementation of tablets within road maintenance has been studied and evaluated with the goal of deepening knowledge in the road maintenance industry and create new methods of working that benefits every part of the construction process. The degree project has been conducted at Peab on their site office in Sundsvall where the operation area of the implementation has been. It contains about 1220 kilometers of state roads where Trafikverket are the client. Through literature studies, tests and participant observations of the tablets usability and development has been analyzed and conclusions has been made to give Peab feedback and suggestions for improvement. The evaluation wasn’t completed because of the fact that Peab didn’t finish the development of tablet within my ten weeks of work there. Use of tablets in road maintenance will increase efficiency of the inspections with the handling of documents digitally, and the important implementation of GPS positioning in the vehicles who conducts the inspections. It will be a lot more flexible than the previous methods where almost every part of the process included a lot of unnecessary paperwork. Peab are recommended to keep looking for possibilities for development of tablets within other areas of road maintenance to get a complete and productive product.

Microfinance is a relatively young and somewhat ambiguous concept. The phenomenon has, however, proven to contribute to making the lives better for many poor people, thus the interest for the industry has grown substantially. The increased attention has stimulated the movement towards more financially sustainable organizations. Along with this transformation, concerns regarding how it affects the poor have been raised. This study aims to map the key characteristics of financially sustainable microfinance institutions (MFIs) and what features that separates them from their non-sustainable counterparts. By analyzing data from 1109 MFIs, some significant differences between sustainable and non-sustainable organizations have been found. The study shows that for-profit MFIs are self-sufficient to a greater extent than the non-sufficient ones, which might be caused by the pressure to deliver value to shareholders. Furthermore, there are indications that self-sufficient MFIs are more efficient, which can be assumed to be caused by technological advantages, or different lending methods. The findings on outreach are somewhat contradictory; sustainable MFIs are reaching more clients on average, which discards a mission drift. On the other hand, self-sufficient MFIs have larger average loan sizes and less female borrowers, two indications that a mission drift actual exists. Self-sufficient MFIs have also proven to have lower loan loss rates and lower yields on loan portfolio. Positive findings, as they indicate that the MFIs have sound loan portfolios and that they have managed to become self-sustainable not by exploiting the poor, but by reducing costs and increasing efficiency. Financial sustainability can therefore be assumed to be achieved without forsaking the poor, if the social aims of the organizations are consistent with the financial objectives.

Title: Ionospheric drift measurements Author: Daniel Kouba Department: Department of Surface and Plasma Science Supervisor: RNDr. Petra Koucká Knížová, Ph.D. Institute of Atmospheric Physics, Academy of Sciences of the Czech Republic Abstract: This thesis deals with ionospheric drift measurements using Digisonde DPS-4. The results are divided into theoretical and practical part. It is shown that for drift data processing it is necessary to choose correct reflection points corresponding to drift movements first. The selection is made in three steps: restriction of Doppler frequency shift, height range selection, choice of the maximum zenith angle. After- wards it is possible to credibly estimate the vector of drift velocity. The experimental results of drift measurements are based on data from the Pru- honice station. To study the common behavior of ionospheric drifts during quiet conditions, the statistical characteristics of drift velocity components are investiga- ted in layers E and F during geomagnetic quiet year 2006. The maximal daily amplitude of horizontal velocity demonstrates the influence of geomagnetic activity. It is proved to be considerable in the F layer meanwhile it was not observed in the E layer. The effect of strong geomagnetic disturbances is shown at two examples of extreme storms. The...