We demonstrate spectral learning can be combined with a random forest classifier to produce a hybrid recommender system capable of incorporating meta information. Spectral learning is supervised learning in which data is in the form of one or more networks. Responses are predicted from features obtained from the eigenvector decomposition of matrix representations of the networks. Spectral learning is based on the highest weight eigenvectors of natural Markov chain representations. A random forest is an ensemble technique for supervised learning whose internal predictive model can be interpreted as a nearest neighbor network. A hybrid recommender can be constructed by first deriving a network model from a recommender's similarity matrix then applying spectral learning techniques to produce a new network model. The response learned by the new version of the recommender can be meta information. This leads to a system capable of incorporating meta data into recommendations.

Extremely polymorphic genes of major histocompatibility complex (MHC) play a significant role in the function of immune system by recognizing heterogeneous particles, mainly pathogenic origin. Previous research on various vertebrate species indicates that MHC influences individual body odour and mate choice preferences. Many individuals tend to prefer MHC dissimilar partner so that warrants them an offspring resistant against wider spectrum of infections. Research on MHC-related mate preferences in humans, however, is inconclusive to date. Several studies indicate that women not taking hormonal contraceptives prefer the smell of MHC dissimilar partners while other studies have not come to this conclusion. This can be caused by the absence of potentially influencing factors like the menstrual cycle phase. The aim of this study was to test MHC-similarity mate choice preferences in odour, facial and vocal modalities. In particular, we focused on a potential effect of hormonal contraception. Furtermore, we tested preferential shifts across the menstrual cycle by comparing women's preferences in the follicular and the luteal phase in pill and non-pill users. A group of 52 women in different phases of their menstrual cycle rated odour samples, photos and vocal recordings taken from 51 men. All...

Implicit feedback is usually data that comes from users’ clicks, search queries and text highlights. It exists in abun- dance, but it is riddled with much noise and requires advanced algorithms to properly make good use of it. Several findings suggest that factors such as click-through data and reading time could be used to create user behaviour models in order to predict the users’ information need. This Master’s thesis aims to use click-through data and search queries together with heuristics to create a model that prioritises metadata-fields of the documents in order to predict the information need of a user. Simply put, implicit feedback will be used to improve the precision of a search engine. The Master’s thesis was carried out at Findwise AB - a search engine consultancy firm. Documents from the benchmark dataset INEX were indexed into a search engine. Two different heuristics were proposed that increment the priority of different metadata-fields based on the users’ search queries and clicks. It was assumed that the heuristics would be able to change the listing order of the search results. Evaluations were carried out for the two heuristics and the unmodified search engine was used as the baseline for the experiment. The evaluations were based on simulating a user that searches queries and clicks on documents. The queries and documents, with manually tagged relevance, used in the evaluation came from a data set given by INEX. It was expected that listing order would change in a way that was favourable for the user; the top-ranking results would be documents that truly were in the interest of the user. The evaluations revealed that the behaviour of the heuristics and the baseline have erratic behaviours and metrics never converged to any specific mean-relevance. A statistical test revealed that there is no difference in accuracy between the heuristics and the baseline. These results mean that the proposed heuristics do not improve the precision of the search engine and several factors, such as the indexing of too redundant metadata, could have been responsible for this outcome. / Implicit feedback är oftast data som kommer från användarnas klick, sökfrågor och textmarkeringar. Denna data finns i överflöd, men har för mycket brus och kräver avancerade algoritmer för att man ska kunna dra nytta av den. Flera rön föreslår att faktorer som klickdata och läsningstid kan användas för att skapa beteendemodeller för att förutse användarens informationsbehov. Detta examensarbete ämnar att använda klickdata och sökfrågor tillsammans med heuristiker för att skapa en modell som prioriterar metadata-fält i dokument så att användarens informationsbehov kan förutses. Alltså ska implicit feedback användas för att förbättra en sökmotors precision. Examensarbetet utfördes hos Findwise AB - en konsultfirma som specialiserar sig på söklösningar. Dokument från utvärderingsdatamängden INEX indexerades i en sökmotor. Två olika heuristiker skapades för att ändra prioriteten av metadata-fälten utifrån användarnas sök- och klickdata. Det antogs att heuristikerna skulle kunna förändra ordningen av sökresultaten. Evalueringar utfördes för båda heuristiker och den omodifierade sökmotorn användes som måttstock för experimentet. Evalueringarna gick ut på att simulera en användare som söker på frågor och klickar på dokument. Dessa frågor och dokument, med manuellt taggad relevansdata, kom från en datamängd som tillhandahölls av INEX. Evalueringarna visade att beteendet av heuristikerna och måttstocket är slumpmässiga och oberäkneliga. Ingen av heuristikerna konvergerar mot någon specifik medelrelevans. Ett statistiskt test visar att det inte är någon signifikant skillnad på uppmätt träffsäkerhet mellan heuristikerna och måttstocket. Dessa resultat innebär att heuristikerna inte förbättrar sökmotorns precision. Detta utfall kan bero på flera faktorer som t.ex. indexering av överflödig meta-data.

Radiologists often have to look through many different patients and examinations in quick succession, and to aid in the workflow the different types of images should be presented for the radiologist in the same manner and order between each new examination. Thus decreasing the time needed for the radiologist to either find the correct image or rearrange the images to their liking. A step in thisprocess requires a comparison between two images to be made and produce a score between 0-1 describing how similar the images are. A similar algorithm already exists at Sectra, but that algorithm only uses the metadata from the images without considering the actual pixel data. The aim of this thesis were to explore different methods of doing the same comparison as the previous algorithm but only using the pixel data. Considering only 3D volumes from CT examinations of the abdomen and thorax region, this thesis explores the possibility of using SSIM, SIFT and SIFT together with a histogram comparison using the Bhattacharyya distance for this task. It was deemed very important that the ranking produced when ordering the images in terms of similarity to one reference image followed a specific order. This order was determined by consulting personnel at Sectra that works closely with the clinical side of radiology. SSIM were able to differentiate between different plane orientations since they usually had large resolution differences in each led, but it could not be made tofollow the desired ranking and was thus disregarded as a reliable option for this problem. The method using SIFT followed the desired ranking better, but struggled a lot with differentiating between the different contrast phases. A histogram component were also added to this method, which increased the accuracy and improved the ranking. Although, further development is still needed for thismethod to be a reliable option that could be used in a clinical setting.

No description available.

The reliable operation of Integrally Geared Centrifugal Compressors (IGCCs), used in the coal-fired power generation industry of South Africa, is essential for economic, environmental and safety considerations. However, due to the unavailability of individual stage performance curves, the ability of a compressor owner to identify underperforming stages to maintain these compressors proactively remains limited. This study addresses the stage performance prediction of an IGCC when only the compressor’s overall performance characteristic, in conjunction with the impeller diameters and tip speeds, are known. The study is limited to IGCCs used in the coal-fired power generation industry of South Africa. Based on the limited inputs, two performance modelling methods were considered for this application, namely stage stacking and 1-dimensional modelling. However, stage stacking requires known operating points on each stage performance curve from which the rest of the curve can be extrapolated while 1-dimensional models require detailed stage design information to model stage performance. This study developed a revised stage stacking procedure which in contrast to the traditional stage stacking procedure, does not require a known operating point on each stage’s performance curve, for it assesses the relative stage performance at the compressor’s surge flow rate. The relative maximum pressure ratio of each stage is acquired through the application of similarity principles while a simplified 1-dimensional impeller analysis model is used to assess relative impeller head coefficients. The modelling process was developed based on performance and design data for IGCCs obtained from a compressor manufacturer. Performance data of four IGCCs, consisting of 13 stages, were obtained, including the design data for ten impellers. Hence, the IGCCs satisfy the requirements of geometric and aerodynamic similarity, unveiling a linear relationship between the stage impeller tip speed and maximum pressure ratio. A simplified 1-dimensional performance model was used to assess relative impeller head coefficients. A verification procedure ensured the integrity of the findings of the 1-dimensional model was maintained by comparing the model results to findings obtained using commercial compressor performance modelling software. A sensitivity analysis was conducted on the 1-dimensional performance model to ascertain which input parameters could be scaled as a function of the impeller tip diameter. For the four IGCCs for which data were obtained, the stage-discharge pressure and isentropic efficiency curves were calculated using the developed model. The maximum variation between the measured and calculated pressure and isentropic efficiency curves equaled 8.20% and 10.84%, respectively. The prediction accuracy of the developed modelling procedure is similar to map-based models found in literature and is considered adequate for identifying an underperforming stage. Thus, the developed model could serve as a valuable conditioning monitoring tool for site-based compressor owners. / Dissertation (MEng)--University of Pretoria, 2019. / Mechanical and Aeronautical Engineering / MEng / Unrestricted

4 Abstract The Relation Between Facial Similarity of Long-Term Couples and Overall and Sexual Satisfaction in the Relationship Results suggest that people tend to choose a partner based on the theory of positive assortative mating (Penton-Voak et al., 1999b). According to this theory, individuals prefer partners with physical and mental characteristics similar to theirs. Preference for facial similarities among partners was confirmed in some studies (Hinsz, 1989; Bereczkei et al., 2004). Other studies suggest that facial resemblance evokes a feeling of familiarity and confidence, but excludes mutual attraction (De Bruine, 2004). The aim of this study was to determine whether facial similarity of long-term partners is related to the overall and sexual satisfaction in the relationship. During the final session of cohabiting relationship research (Klapilová et al., 2006 - 2009) both members of 50 long-term couples completed questionnaires assessing their overall satisfaction (Dyadic Adjustment Scale), sexual satisfaction (Hurlbert Index of Sexual Compatibility), and facial photographs of partners were taken. Then 80 participants judged the similarity of standardized facial photographs of long-term partners. Then, using a regression analysis and LMM analysis the relation between rated couples similarity and...

Title: Source Parameters of Microearthquakes and their Uncertainty Author: Jan Michálek Department: Department of Geophysics Supervisor: Doc. RNDr. Tomáš Fischer, PhD., Institute of Hydrogeology, Engineering Geology and Applied Geophysics, Charles University in Prague Abstract: The aim of this thesis is to analyze the spectral methods used for the determination of the earth- quake source parameters like the seismic moment M0 and the corner frequency fc and to apply these meth- ods to seismic data from the West Bohemian region. Considering some assumptions about the source the other important parameters like the source radius r or the stress drop in the source ∆σ can be evaluated. Determination of the parameters is performed in the spectral domain by comparing a simple Brune's source model (non-causal slip on a circular rupture; spectral slope ω−2) with the displacement of the P or S wave. The methods were applied to 56 selected earthquakes of the West Bohemian swarms from 2000 and 2008 by the absolute and relative approach in several modifications. The absolute method allows to determine not only the source parameters but also the quality factor Q (attenuation), which significantly affects the deter- mination of fc. Therefore, the absolute method was applied also as the joint inversion when Q is stabilized....

The focus was to evaluate the Inertial Dissipation Method (IDM) over land duringunstable conditions. This was done by comparing the friction velocity, u*, from theeddy-correlation method (ECM) with u* from IDM. The result can be used to see ifIDM can rely on its assumptions, since the surface-layer theory is more fulfilled overland, where we for example do not have wave influence. The measurements weretaken from the flat agricultural site Marsta, 8 km north of Uppsala, Sweden. The result shows that IDM works well over land (relative standard deviation of about 10 %). For weakly unstable stratification, it is enough to use an assumption of neutralconditions in the IDM calculations. If it is more unstable, one should include theinfluence of stability and also include an imbalance term. The imbalance term isintroduced implicitly by varying the effective Kolmogorov’s constant with stability.The effective Kolmogorov’s constant used here, varied from 0.50 up to above 0.80. To calculate u* using IDM, a first estimation of u* was calculated from aparameterised drag coefficient CD. Also, to imitate the measuring setup on a movingplatform on the sea, the stability parameter, z/L, was calculated using a bulkestimatedheat flux. The large scatter showed that it is important that theparameterisations of CD and the heat flux are good. One can conclude that the IDM as a method to determine turbulent fluxes over landworks satisfactory. The larger scatter over sea is probably an effect of sea waveinfluence, even though the sea surface is considered more homogeneous and theconditions more stationary. / Syftet var att utvärdera inertial-dissipationsmetoden (IDM) över land under instabilaförhållanden. Detta gjordes genom att jämföra friktionshastigheten, u* , från eddycorrelation-metoden (ECM) med u* från IDM. Resultaten kan användas till att sehuruvida man kan använda de antaganden som IDM vilar på. Ytskiktsteorin ärtroligen mer uppfylld över land, eftersom det t.ex. inte finns något våginflytande.Mätningarna är gjorda i Marsta, som ligger i ett flackt jordbrukslandskap, 8 km norrom Uppsala. Resultatet visar att IDM fungerar tillfredställande över land (relativ standardavvikelseca 10 %). För svagt instabila förhållanden, räcker det med att anta neutral skiktning iIDM-beräkningarna. Om det är mer instabilt, bör man ta hänsyn till stabiliteten ochäven inkludera en obalansterm. Obalanstermen introduceras implicit genom att varieraden effektiva Kolmogorovkonstanten med stabiliteten. Den effektivaKolmogorovkonstanten som användes här, varierade från 0.5 till över 0.8. För att beräkna u* med hjälp av IDM, beräknades en första uppskattning av u* genom att använda en parametriserad ”drag-coefficient” CD. För att efterlikna demedel man har att tillgå på skepp och bojar i rörelse ute på havet, beräknadesstabilitetsparametern, z/L, med hjälp av ett bulkbestämt värmeflöde. Den storaspridningen av data visade att det är viktigt att parameteriseringen av CD samt värmeflödet är bra. Man kan dra slutsatsen att dissipationsmetoden fungerar tillfredsställande, som metodatt bestämma turbulenta flöden över land. Den större spridningen som man fått överhav är förmodligen en effekt av t.ex. vågor, trots att havsytan anses vara merhomogen och förhållandena mer stationära.

Productivity for digital circuit design is being outpaced currently by the rate at which<br />silicon is growing such as FPGAs. Complex designs take a large amount of engineering<br />hours to complete. Reuse of existing design can potentially decrease this cost and increase<br />design productivity. However, existing digital hardware designs are not being effectively<br />reused by the hardware community due to the inability of designers to have knowledge of<br />all the attributes of designs that can be reused. In addition, designers will have to accustom<br />themselves to designs in the hardware library. By having a back-end system that looks for<br />similar circuits, there is little to no effort for the designer to reuse the design. This thesis<br />provides an overview and comparison of different methods for characterizing and comparing<br />digital circuits in order to suggest candidate circuits that engineers can reuse. Several of<br />these methods are implemented, modified, and compared to show the feasibility of utilizing<br />this work for increasing overall productivity.<br /> / Master of Science