In this thesis work, we propose a Map-Reduce variant of the Growing Hierarchical Self Organizing Map (GHSOM) called MR-GHSOM, which is capable of handling mixed attribute datasets of massive size. The Self Organizing Map (SOM) has proved to be a useful unsupervised data analysis algorithm. It projects a high dimensional data onto a lower dimensional grid of neurons. However, the SOM has some limitations owing to its static structure and the incapability to mirror the hierarchical relations in the data. The GHSOM overcomes these shortcomings of the SOM by providing a dynamic structure that adapts its shape according to the input data. It is capable of growing dynamically in terms of the size of the individual neuron layers to represent data at the desired granularity as well as in depth to model the hierarchical relations in the data. However, the training of the GHSOM requires multiple passes over an input dataset. This makes it difficult to use the GHSOM for massive datasets. In this thesis work, we propose a Map-Reduce variant of the GHSOM called MR-GHSOM, which is capable of processing massive datasets. The MR-GHSOM is implemented using the Apache Spark cluster computing engine and leverages the popular Map-Reduce programming model. This enables us to exploit the usefulness and dynamic capabilities of the GHSOM even for a large dataset. Moreover, the conventional GHSOM algorithm can handle datasets with numeric attributes only. This is owing to the fact that it relies heavily on the Euclidean space dissimilarity measures of the attribute vectors. The MR-GHSOM further extends the GHSOM to handle mixed attribute - numeric and categorical - datasets. It accomplishes this by adopting the distance hierarchy approach of managing mixed attribute datasets. The proposed MR-GHSOM is thus capable of handling massive datasets containing mixed attributes. To demonstrate the effectiveness of the MR-GHSOM in terms of clustering of mixed attribute datasets, we present the results produced by the MR-GHSOM on some popular datasets. We further train our MR-GHSOM on a Census dataset containing mixed attributes and provide an analysis of the results.

We use ERA-Interim reanalysis data of 2 meter temperature to perform a pattern analysis of the Arctic temperatures exploiting an artificial neural network called Self Organizing-Map (SOM). The SOM method is used as a cluster analysis tool where the number of clusters has to be specified by the user. The different sized SOMs are analyzed in terms of how the size changes the representation of specific features. The results confirm that the larger the SOM is chosen the larger will be the root mean square error (RMSE) for the given SOM, which is followed by the fact that a larger number of patterns can reproduce more specific features for the temperature. / Wir benutzten das künstliche neuronale Netzwerk Self Organizing-Map (SOM), um eine Musteranalyse von ERA-Interim Reanalysedaten durchzuführen. Es wurden SOMs mit verschiedener Musteranzahl verglichen. Die Ergebnisse zeigen, dass SOMs mit einer größeren Musteranzahl deutlich spezifischere Muster produzieren im Vergleich zu SOMs mit geringen Musteranzahlen. Dies zeigt sich unter anderem in der Betrachtung der mittleren quadratischen Abweichung (RMSE) der Muster zu den zugeordneten ERA Daten.

Abstract Phospholipids are the main components of cell membranes, lipoproteins and other membrane structures in living organisms. Properties of lipid molecules are important to the overall behaviour and interactions of membranes. Furthermore, characteristics of the biological membranes act as important regulators of membrane functions. Molecular dynamics (MD) simulations were applied in this thesis to study properties of biological membranes. A certain degree of acyl chain polyunsaturation is essential for the proper functioning of membranes, but earlier MD simulations had not addressed the effects of polyunsaturation. Therefore a solvated all-atom bilayer model consisting of diunsaturated 1-palmitoyl-2-linoleoyl-3-phosphatidylcholine (PLPC) molecules was simulated. The analysis of the simulation data was focused on the effects of double bonds on a membrane structure. Self-organising neural networks were applied to the analysis of the conformational data from the 1-ns simulation of PLPC membrane. Mapping of 1.44 million molecular conformations to a two-dimensional array of neurons revealed, without human intervention or requirement of a priori knowledge, the main conformational features. This method provides a powerful tool for gaining insight into the main molecular conformations of any simulated molecular assembly. Furthermore, an application of MD simulations in the comparative analysis of the effects of lipid hydrolysis products on the membrane structure was introduced. The hydrolysis products of the phospholipase A2 (PLA2) enzyme are known to have a role in a variety of physiological processes and the membrane itself acts as an important regulator of this enzyme. The simulations revealed differences in the bilayer properties between the original and hydrolysed phospholipid membranes. This study provides further evidence that MD simulations on biomembranes are able to provide information on the properties of biologically and biochemically important lipid systems at the molecular level.

This study addresses the susceptibility of atmospheric rivers (ARs) to the behavior of the North Atlantic Subtropical High (NASH). ARs are a major mechanism for meridional moisture transport often connected to heavy precipitation and mid-latitude troughs. The NASH, a semi-permanent anticyclone over the subtropical North Atlantic Ocean, has been shown to be significantly influential on precipitation variability over the southeastern United States. A self-organizing map (SOM) was trained on a 4 x 3 regular grid over 250 iterations using ERA5 derived 6-hourly 850 hPa Geopotential Heights ≥ 1535 gpm from 1979-2020. The 12 resulting "nodes" were analyzed with respect to ARs defined by objects of ERA5 derived integrated water vapor transport (IVT) > 500 m-1 s-1 with lengths > 2000 km. Composites of thresholded 850 hPa heights, AR-concurrent PRISM precipitation, AR spatial frequency distribution maps, and seasonal AR frequency histograms per node illustrate seasonal interactions between the NASH and ARs that demonstrate a tendency of more frequent ARs and higher mean AR-driven precipitation over the Mississippi embayment and Ohio River Valley in the summer months, believed to be representative of extreme moisture transport events, when the NASH exhibits increased intensity, spatial expansion, and southwestward migration. Conversely, AR frequency and AR-concurrent precipitation composites suggest wintertime events are mainly supported by dynamically-driven nor'easter and bomb type cyclones when the NASH is constricted, at higher latitudes, and further east. Findings suggest that extreme summertime water vapor transport events associated with an AR are enhanced by the warm season NASH due to its increased intensity and proximity to the eastern US that acts as a supplementary lifting mechanism amidst low dynamic influence. / Master of Science / This study aims to investigate the response of atmospheric rivers (ARs) to the behavior of the North Atlantic Subtropical High (NASH). ARs are a major vehicle for the poleward transport of moisture from the tropics and subtropics. ARs are often affiliated with heavy precipitation and mid-latitude cyclones and frontal boundaries. The NASH, a semi-permanent anticyclone over the subtropical North Atlantic Ocean, has been shown to be significantly influential on precipitation variability over the southeastern United States. A self-organizing map (SOM), a method of vector quantification, was trained on a 4 x 3 regular grid over 250 iterations using ERA5 derived 6-hourly 850 hPa Geopotential Heights ≥ 1535 meters from 1979-2020. The 12 resulting "nodes" were analyzed with respect to ARs defined by objects that result from masking the rate of transport of water vapor within a vertical column from 1000 hPa to 300 hPa of which that are greater than 2000 km long. Composites of thresholded 850 hPa heights, AR-concurrent precipitation, AR spatial frequency distribution maps, and seasonal AR frequency histograms per node illustrate seasonal interactions between the NASH and ARs that demonstrate a tendency of more frequent ARs and higher mean AR-driven precipitation over the Mississippi embayment and Ohio River Valley in the summer months, believed to be representative of severe precipitation events, when the NASH is stronger, larger, and further southwestward. Conversely, AR frequency and AR-concurrent precipitation composites suggest wintertime events are mainly supported by nor'easter and bomb type cyclones that occur when the Polar jet stream is strongest and when the NASH is constricted, at higher latitudes, and further east. Findings suggest that extreme summertime water vapor transport events associated with an AR are enhanced by the warm season NASH due to its increased intensity and proximity to the eastern US that acts as a supplementary lifting mechanism amidst low dynamic influence.

Work deal about self-organizing maps, especially about Kohonen self-organizing map. About creating of aplication, which realize creating and learning of self-organizing map. And about usage of self-organizing map for self-localization of robot.

Šiame darbe pateikti ir aprašyti biologinio ir dirbtinio neurono modeliai. Didžiausias dėmesys skiriamas vieno tipo neuroniniams tinklams – saviorganizuojantiems žemėlapiams (SOM). Darbe pateiktas jų apmokymas, taip pat pagrindinių sąvokų (epocha, kaimynystės eilė, unifikuotų atstumų matrica ir kt.), susijusių su SOM neuroniniais tinklais (žemėlapiais), apibrėžimai. Buvo nagrinėtos keturios saviorganizuojančių neuroninių tinklų sistemos: NeNet, SOM-Toolbox, DataBionic ESOM, Viscovery SOMine ir Matlab įrankiai „nntool“, „nctool“, kurie naudojami SOM tinklams sukurti ir apmokyti. Pateikiamos sistemų naudojimosi instrukcijos, norint gauti paprasčiausią SOM žemėlapį. Matlab aplinkoje sukurta ir darbe aprašyta naują vizualizavimo būdą turinti sistema „Somas“, pateiktas jos išskirtinumas ir naudojimosi instrukcija. Sistemoje „Somas“ realizuota kita mokymo funkcija nei kitose minėtose sistemose. Pagrindinis analizuotų sistemų tikslas yra suskirstyti duomenis į klasterius pagal jų panašumą ir pateikti juos SOM žemėlapyje. Sistemos viena nuo kitos skiriasi duomenų pateikimu, mokymo taisyklėmis, vizualizavimo galimybėmis, todėl čia aptariami sistemų panašumai ir skirtumai. Nagrinėti susidarę SOM žemėlapiai ir gautos kvantavimo bei topografinės paklaidos, analizuojant tris duomenų aibes: irisų, stiklo ir vyno. Kvantavimo ir topografinės paklaidos yra kiekybiniai vaizdo kokybės įverčiai. Padarytos išvados apie susidariusius klasterius tiriamuose duomenyse. Naudojant naują sistemą „Somas“... [toliau žr. visą tekstą] / In this master thesis, biologic and artificial neuron models have been described. The focus is selforganizing maps (SOM). The self-organizing maps are one of types of artificial neural networks. SOM training as well as the main concepts which need to explain SOM networks (epochs, neighbourhood size, u-matrix and etc.) have been described. Four systems of self-organizing maps: NeNet, SOMToolbox, DataBionic ESOM, Viscovery SOMine, and Matlab tools “nntool” and “nctool” have been analyzed. In the thesis, a system use guide has been presented to make a simple SOM map. A new system “Somas” that has a new visualisation way has been developed in Matlab. The system has been described, its oneness has been emphasized, and a use guide is presented. The main target of the SOM systems is data clustering and their graphical presentation on the self-organizing map. The SOM systems are different one from other in their interfaces, the data pre-processing, learning rules, visualization manners, etc. Similarities and differences of the systems have been highlighted here. The experiments have been carried out with three data sets: iris, glass and wine. The SOM maps, obtained by each system, have been described and some conclusions on the clusters have been drawn. The quantization and topographic errors have been analyzed to estimate the quality of the maps obtained. An investigation has been carried out in the new system “Somas” and system “NeNet” in order to look how quantization and... [to full text]

Saviorganizuojantys neuroniniai tinklai (SOM) yra susilaukę nemažai populiarumo mokslininkų tarpe klasterizuojant ar vizualizuojant daugiamačius duomenis. Šiame magistro diplominiame darbe smulkiai išnagrinėtas SOM algoritmas bei veikimo principai, pateiktos galimos parametrų reikšmės ir kaimynystės funkcijos. Taip pat nurodyti tinklo kokybės įvertinimo kriterijai ir duomenų vizualizavimo metodai taikant saviorganizuojantį neuroninį tinklą. SOM pagrindinis tikslas yra duomenų klasterizavimas, o ne vizualizavimas, todėl duomenų vaizdavimas SOM'u turi savų trūkumų – žemėlapyje negalima matyti atstumų tarp klasės neuronų ir kaip toli nutolusios viena nuo kitos klasės. Pateikta alternatyva – SOM modifikacija ViSOM. Darbe išnagrinėti ViSOM algoritmo esminiai skirtumai, aprašyti parametrų parinkimo ypatumai. Nagrinėti SOM ir ViSOM vizualizavimo skirtumus sukurta MATLAB sistemoje programa, realizuojanti abu algoritmus bei pateiktas programos galimybių ir scenarijų aprašas. Pasirinkus tris kaimynines funkcijas su šia programa atlikti tyrimai, rodantys, kad kvantavimo ir topografinės paklaidos netinkamos vertinant ViSOM vaizdo kokybę. Pasiūlyti trys nauji vertinimo kriterijai, bei su jais atlikti tyrimai, parodantys jų veiksmingumą. Taip pat darbe vizualiai parodytas ir aprašytas ViSOM žemėlapio kitimas priklausomai nuo rezoliucijos. / A self-organizing map is a type of artificial neural networks that has received substantial popularity among scientists in regards to clustering and visualization of multidimensional data. In this master theses, the learning algorithm and the main principals are examined in detail, the neighbourhood functions and values of various parameters are given. Some criteria of the network evaluation quality and the data visualization methods using the self-organizing maps are given as well. The main goal of the SOM is clustering of data, but not the visualization, so the visual data representation by the SOM has its drawbacks – it is impossible to see the distances between neurons, corresponding the vectors belong to a class, and how far from each other the classes are in a map. The alternative – SOM modification, called ViSOM, has been developed. The main differences of SOM and ViSOM are investigated, the peculiarity of parameter selection is also examined in this work. In order to study the differences of SOM and ViSOM visualization, a system in MATLAB has been developed, both algorithms have been implemented, and the feature and scenario list of the program is presented. Some experiments have been carried out by selecting three neighborhood functions. The experiments have showed that the quantization and topographic errors are not suitable for studying the visualization of ViSOM. Three new evaluation criteria are proposed. The investigation shows their effectiveness. In the work... [to full text]

No description available.

The self-organizing map (SOM) is an unsupervised learning algorithm which has been successfully applied to various applications. One of advantages of SOM is it maintains an incremental property to handle data on the fly. In the last several decades, there have been variants of SOM used in many application domains. In this dissertation, two new SOM algorithms are developed for image quantization and compression. The first algorithm is a sample-size adaptive SOM algorithm that can be used for color quantization of images to adapt to the variations of network parameters and training sample size. The sweep size of neighborhood function is modulated by the size of the training data. In addition, the minimax distortion principle which is modulated by training sample size is used to search the winning neuron. Based on the sample-size adaptive self-organizing map, we use the sampling ratio of training data, rather than the conventional weight change between adjacent sweeps, as a stop criterion. As a result, it can significantly speed up the learning process. Experimental results show that the proposed sample-size adaptive SOM achieves much better PSNR quality, and smaller PSNR variation under various combinations of network parameters and image size. The second algorithm is a novel classified SOM method for edge preserving quantization of images using an adaptive subcodebook and weighted learning rate. The subcodebook sizes of two classes are automatically adjusted in training iterations based on modified partial distortions that can be estimated incrementally. The proposed weighted learning rate updates the neuron efficiently no matter of how large the weighting factor is. Experimental results show that the proposed classified SOM method achieves better quality of reconstructed edge blocks and more spread out codebook and incurs a significantly less computational cost as compared to the competing methods.

The demand for imported fruit in China has increased dramatically since it first appeared in Chinese markets around 1993. Although imported fruit is much more expensive than domestically produced fruit and people’s income is much lower in China than in developed countries, imported fruit still attracts many willing buyers. Conventional concepts such as meeting basic needs or increasing consumer awareness of the importance of fresh fruit to a healthy lifestyle cannot adequately explain this phenomenon, as there is an abundance of fresh, cheap, local produce available in almost every Chinese market. There must be other factors influencing buying behaviour. To explore these factors and to examine the dynamics of the market for imported fruit, this research adopted a mixed qualitative-quantitative methodology guided by the paradigm of phenomenology. The research examined the characteristics of imported fruit itself, criteria imposed by Chinese buyers on these characteristics, the intended uses of imported fruit and their associated consumption values. To shed light on the possible influence of socio-economic factors, the research also compared buyers from two Chinese cities, Guangzhou and Urumqi. The former is one of the most developed cities in China and the latter is regarded as among the more backward and conservative cities in China. The research identified ten attributes that appeal to Chinese buyers. Six relate to the fruit’s physical attributes: that it has better appearance and packaging, lower chemical residues, better or different taste, and freshness. The remaining four relate to symbolic attributes associated with the fruit: that it represents achievement, wealth, personality and social status. Five intended uses of imported fruit were identified: for gifts, self-consumption, children, aged parents and patients. Four consumption values underlying these intentions were also identified: symbolism, concern for health, meeting basic needs and hedonism. However, the research revealed that no single combination of intended use and consumption value drives the demand for imported fruit in the Chinese market. Most frequently, it is a mix of hedonic and symbolic values behind a range of different intended uses that stimulates demand. Pursuing hedonic and symbolic values also leads to the visual quality of imported fruit generally being the most appealing attribute to Chinese buyers, a pattern common to both Guangzhou and Urumqi. These findings make a significant contribution to empirical knowledge about Chinese consumer behaviour. Results provide valuable insights into the interrelationships among product attributes, intended uses, consumption values and cultural values, and would give essential guidance to the development of strategies to market imported fruit in China. The research also examined limitations of current analytical approaches to the study of consumer behaviour. It demonstrated that approaches based on neural networks and fuzzy logic could be used independently or combined with conventional statistical methods to improve the explanation of consumer behaviour in this case. A comparison was carried out between the most popular form of neural networks (feedforward networks) and multivariate statistical methods in terms of their ability to predict behavioural intention through consumers’ attitudes towards products. Results demonstrated that neural networks were capable of capturing nonlinear aspects of complex relationships and producing better predictions than conventional statistical models. To explore consumer cognitive patterns, the research also compared K-means clustering with a Self-organizing Map (SOM) neural network in terms of the ability to cluster consumers on the basis of perceptions towards imported fruit attributes. Results indicate a superior outcome when K-means is used in conjunction with SOM in clustering analysis: using SOM to determine the ‘natural’ numbers of clusters and using K-means to do clustering. Finally, to quantitatively evaluate the impact of consumption values, this research develops a new approach that combines Means-end Chain theory with fuzzy logic theory. Given the global importance of the Chinese market, the successful application of neural networks and fuzzy logic in this study of the behaviour of Chinese consumers purchasing imported fruit could have wider ramifications. If the approach were proven in other applications, it could significantly improve the ability to understand the demand for consumer goods in China.