abstract: Stock market news and investing tips are popular topics in Twitter. In this dissertation, first I utilize a 5-year financial news corpus comprising over 50,000 articles collected from the NASDAQ website matching the 30 stock symbols in Dow Jones Index (DJI) to train a directional stock price prediction system based on news content. Next, I proceed to show that information in articles indicated by breaking Tweet volumes leads to a statistically significant boost in the hourly directional prediction accuracies for the DJI stock prices mentioned in these articles. Secondly, I show that using document-level sentiment extraction does not yield a statistically significant boost in the directional predictive accuracies in the presence of other 1-gram keyword features. Thirdly I test the performance of the system on several time-frames and identify the 4 hour time-frame for both the price charts and for Tweet breakout detection as the best time-frame combination. Finally, I develop a set of price momentum based trade exit rules to cut losing trades early and to allow the winning trades run longer. I show that the Tweet volume breakout based trading system with the price momentum based exit rules not only improves the winning accuracy and the return on investment, but it also lowers the maximum drawdown and achieves the highest overall return over maximum drawdown. / Dissertation/Thesis / Doctoral Dissertation Computer Science 2016

No description available.

This thesis focuses on the re-generation of sound from a spike based coding system. Three different types of spike based coding system have been analyzed. Two of them are biologically inspired spike based coding systems i.e. the spikes are generated in a similar way to how our auditory nerves generate spikes. They have been called AN (Auditory Nerve) spikes and AN Onset (Amplitude Modulated Onset) spikes. Sounds have been re-generated from spikes generated by both of those spike coding technique. A related event based coding technique has been developed by Koickal and the sounds have been re-generated from spikes generated by Koickal's spike coding technique and the results are compared. Our brain does not reconstruct sound from the spikes received from auditory nerves, it interprets it. But by reconstructing sounds from these spike coding techniques, we will be able to identify which spike based technique is better and more efficient for coding different types of sounds. Many issues and challenges arise in reconstructing sound from spikes and they are discussed. The AN spike technique generates the most spikes of the techniques tested, followed by Koickal's technique (54.4% lower) and the AN Onset technique (85.6% lower). Both subjective and objective types of testing have been carried out to assess the quality of reconstructed sounds from these three spike coding techniques. Four types of sounds have been used in the subjective test: string, percussion, male voice and female voice. In the objective test, these four types and many other types of sounds have been included. From the results, it has been established that AN spikes generates the best quality of decoded sounds but it produces many more spikes than the others. AN Onset spikes generates better quality of decoded sounds than Koickal's technique for most of sounds except choir type of sounds and noises, however AN Onset spikes produces 68.5% fewer spikes than Koickal's spikes. This provides evidences that AN Onset spikes can outperform Koickal's spikes for most of the sound types.

Learning is central to infusing intelligence to any biologically inspired system. This study introduces a novel Cross-Correlated Delay Shift (CCDS) learning method for spiking neurons with the ability to learn and reproduce arbitrary spike patterns in a supervised fashion with applicability tospatiotemporalinformation encoded at the precise timing of spikes. By integrating the cross-correlated term,axonaland synapse delays, the CCDS rule is proven to be both biologically plausible and computationally efficient. The proposed learning algorithm is evaluated in terms of reliability, adaptive learning performance, generality to different neuron models, learning in the presence of noise, effects of its learning parameters and classification performance. The results indicate that the proposed CCDS learning rule greatly improves classification accuracy when compared to the standards reached with the Spike Pattern Association Neuron (SPAN) learning rule and the Tempotron learning rule. Network structureis the crucial partforany application domain of Artificial Spiking Neural Network (ASNN). Thus, temporal learning rules in multilayer spiking neural networks are investigated. As extensions of single-layer learning rules, the multilayer CCDS (MutCCDS) is also developed. Correlated neurons are connected through fine-tuned weights and delays. In contrast to the multilayer Remote Supervised Method (MutReSuMe) and multilayertempotronrule (MutTmptr), the newly developed MutCCDS shows better generalization ability and faster convergence. The proposed multilayer rules provide an efficient and biologically plausible mechanism, describing how delays and synapses in the multilayer networks are adjusted to facilitate learning. Interictalspikes (IS) aremorphologicallydefined brief events observed in electroencephalography (EEG) records from patients with epilepsy. The detection of IS remains an essential task for 3D source localization as well as in developing algorithms for seizure prediction and guided therapy. In this work, we present a new IS detection method using the Wavelet Encoding Device (WED) method together with CCDS learning rule and a specially designed Spiking Neural Network (SNN) structure. The results confirm the ability of such SNN to achieve good performance for automatically detecting such events from multichannel EEG records.

In the stochastic modeling of neurons, the first passage time problem arises as a natural object of study when considering the inter spike interval distribution. In this report, we study some aspects of this problem as it arises in the context of neuroscience. In the first chapter we describe the basic neurophysiology required to model the neuron. In the second, we study the Poisson model, Stein’s model, and some diffusion models, calculating or indicating methods to compute the density of the first passage time random variable or its moments. In the third and fourth chapters, we study the Fokker-Planck equation, and use it to compute the first passage time in the discrete and continuous time random walk cases. In the final chapter, we study sequences of neurons and the change in the density of the waiting time distributions, and hence in the inter spike intervals, as the output spike train from one neuron is considered as the input in the subsequent neuron.

Les neurones sensoriels réagissent à des stimuli externes en émettant des séquences de potentiels d’action (“spikes”). Ces spikes transmettent collectivement de l’information sur le stimulus en formant des motifs spatio-temporels qui constituent le code neural. On observe expérimentalement que ces motifs se produisent de façon irrégulière, mais avec une structure qui peut être mise en évidence par l’utilisation de descriptions probabilistes et de méthodes statistiques. Cependant, la caractérisation statistique des données expérimentales présente plusieurs contraintes majeures: en dehors de celles qui sont inhérentes aux statistiques empiriques comme la taille de l’échantillonnage, ‘le’ modèle statistique sous-jacent est inconnu. Dans cette thèse, nous abordons le problème d’un point de vue complémentaire à l’approche expérimentale. Nous nous intéressons à des modèles neuro-mimétiques permettant d’étudier la statistique collective des potentiels d’action et la façon dont elle dépend de l’architecture et l’histoire du réseau ainsi que du stimulus. Nous considérons tout d’abord un modèle de type Intègre-et-Tire à conductance incluant synapses électriques et chimiques. Nous montrons que la statistique des potentiels d’action est caractérisée par une distribution non stationnaire et de mémoire infinie, compatible avec les probabilités conditionnelles (left interval-specification), qui est non-nulle et continue, donc une distribution de Gibbs. Nous présentons ensuite une méthode qui permet d’unifier les modèles dits d’entropie maximale spatio-temporelle (dont la mesure invariante est une distribution de Gibbs dans le sens de Bowen) et les modèles neuro-mimétiques, en fou / Sensory neurons respond to external stimulus using sequences of action potentials (“spikes”). They convey collectively to the brain information about the stimulus using spatio-temporal patterns of spikes (spike trains), that constitute a “neural code”. Since spikes patterns occur irregularly (yet highly structured) both within and over repeated trials, it is reasonable to characterize them using statistical methods and probabilistic descriptions. However, the statistical characterization of experimental data presents several major constraints: apart from those inherent to empirical statistics like finite size sampling, ‘the’ underlying statistical model is unknown. In this thesis we adopt a complementary approach to experiments. We consider neuromimetic models allowing the study of collective spike trains statistics and how it depends on network architecture and history, as well as on the stimulus. First, we consider a conductance-based Integrate-and-Fire model with chemical and electric synapses. We show that the spike train statistics is characterized by non-stationary, infinite memory, distribution consistent with conditional probabilities (Left interval specifications), which is continuous and non null, thus a Gibbs distribution. Then, we present a novel method that allows us to unify spatio-temporal Maximum Entropy models (whose invariant measure are Gibbs distributions in the Bowen sense) and neuro-mimetic models, providing a solid ground towards biophysical explanation of spatio-temporal correlations observed in experimental data. Finally, using these tools, we discuss the stimulus response of retinal ganglion cells, and the possible generalization of the co

Epileptic seizures are characterised by an onset of abnormal brain activity that evolves in space and time, which ultimately returns to normal background activity. For different types of seizures, the abnormal activity can be vastly different both in duration, electrographic morphology and spatial extent. Mechanistic understanding of the different seizure dynamics (spatially, as well as temporally) is crucial for the advancement and improvement of clinical treatment. To gain a deeper mechanistic insight into different seizure dynamics, mathematical models of brain processes were developed in this thesis. These models are used to explain electrographic seizure dynamics in their temporal, as well as their spatio-temporal evolution. Our studies show that the temporal evolution of seizure dynamics can be understood in terms of prototypic waveforms, which in turn can be represented in terms of three neural population processes. Such a minimal framework lends itself to a detailed phase space analysis, which elucidates seizure waveforms and seizure transitions as topological properties of the phase space. Based on the phase space considerations we show how during spike-wave seizures, single-pulse stimuli can have more complex effects than previously thought. In terms of the spatio-temporal dynamics of seizures, mechanisms for focal seizure onset and propagation are investigated in a model cortical sheet of coupled, discretised columns. The coupling followed nearest-neighbour, as well as realistic mesoscopic cortical connectivities. Different possible causes (e.g. spatial heterogeneities) of seizure generation, as well as different seizure spreading patterns (via different networks) have been investigated. We conclude that focal seizure onset can be due to global (e.g. whole-brain level) causes, global conditions & local triggers, and local (e.g. cortical column level) causes. Clinically relevant predictions from this work include the suggestion of a specific stimulation protocol in spike-wave seizures that incorporates phase space information; and the suggestion of using microscopic cortical incisions to disrupt the integrity of abnormal cortical tissue in order to prevent focal seizure onset. In conclusion, multi-scale computational modelling of seizure dynamics is proposed as an important tool to link theoretical understanding, experimental results, and patient-specific clinical data.

Dans cette thèse, nous étudions les approches calculatoires dédiées des réseaux de neurones profonds et plus particulièrement des réseaux de neurones convolutionnels (CNN). En effet, l'efficacité des réseaux de neurones convolutionnels en font des structures calculatoires intéressantes dans de nombreuses applications. Nous étudions les différentes possibilités d'implémentation de ce type de réseaux pour en déduire leur complexité calculatoire. Nous montrons que la complexité calculatoire de ce type de structure peut rapidement devenir incompatible avec les ressources de l'embarqué. Pour résoudre cette problématique, nous avons fait une exploration des différents modèles de neurones et architectures susceptibles de minimiser les ressources nécessaires à l'application. Dans un premier temps, notre approche a consisté à explorer les possibles gains par changement de modèle de neurones. Nous montrons que les modèles dits impulsionnels permettent en théorie de réduire la complexité calculatoire tout en offrant des propriétés dynamiques intéressantes, mais nécessitent de repenser entièrement l'architecture matériel de calcul. Nous avons alors proposé notre approche impulsionnelle du calcul des réseaux de neurones convolutionnels avec une architecture associée. Nous avons mis en place une chaîne logicielle et de simulation matérielle dans le but d'explorer les différents paradigmes de calcul et implémentation matérielle et évaluer leur adéquation avec les environnements embarqués. Cette chaîne nous permet de valider les aspects calculatoires mais aussi d'évaluer la pertinence de nos choix architecturaux. Notre approche théorique a été validée par notre chaîne et notre architecture a fait l'objet d'une simulation en FDSOI 28 nm. Ainsi nous avons montré que cette approche est relativement efficace avec des propriétés intéressantes un terme de passage à l'échelle, de précision dynamique et de performance calculatoire. Au final, l'implémentation des réseaux de neurones convolutionnels en utilisant des modèles impulsionnels semble être prometteuse pour améliorer l'efficacité des réseaux. De plus, cela permet d'envisager des améliorations par l'ajout d'un apprentissage non supervisé type STDP, l'amélioration du codage impulsionnel ou encore l'intégration efficace de mémoire de type RRAM. / In this thesis, we study the dedicated computational approaches of deep neural networks and more particularly the convolutional neural networks (CNN).We highlight the convolutional neural networks efficiency make them interesting choice for many applications. We study the different implementation possibilities of this type of networks in order to deduce their computational complexity. We show that the computational complexity of this type of structure can quickly become incompatible with embedded resources. To address this issue, we explored differents models of neurons and architectures that could minimize the resources required for the application. In a first step, our approach consisted in exploring the possible gains by changing the model of neurons. We show that the so-called spiking models theoretically reduce the computational complexity while offering interesting dynamic properties but require a complete rethinking of the hardware architecture. We then proposed our spiking approach to the computation of convolutional neural networks with an associated architecture. We have set up a software and hardware simulation chain in order to explore the different paradigms of computation and hardware implementation and evaluate their suitability with embedded environments. This chain allows us to validate the computational aspects but also to evaluate the relevance of our architectural choices. Our theoretical approach has been validated by our chain and our architecture has been simulated in 28 nm FDSOI. Thus we have shown that this approach is relatively efficient with interesting properties of scaling, dynamic precision and computational performance. In the end, the implementation of convolutional neural networks using spiking models seems to be promising for improving the networks efficiency. Moreover, it allows improvements by the addition of a non-supervised learning type STDP, the improvement of the spike coding or the efficient integration of RRAM memory.

Speech processing in the human cochlea introduces travelling waves on the basilar membrane. These travelling waves have largely been ignored in most processing strategies. This study implements a hydrodynamical model in a speech processing strategy in order to investigate the neural spike train patterns for a travelling wave processing strategy. In cochlear implants a trade-off remains between the simulation rate and the number of electrode channels. This trade-off was investigated in the proposed travelling wave strategy. Taking into consideration existing current spread and electrical stimulation models, predicted neural spike train responses have shown that stimulating fewer channels (six and four) at stimulation rates of 2 400 pps and 3 600 pps gives better approximations of predicted normal hearing responses for input frequencies of 200 Hz, 600 Hz and 1 kHz, compared to stimulating more channels at lower channel stimulation rates. The predicted neural spike train patterns suggest that these resulting neural patterns might contain both spatial and temporal information that could be extracted by the auditory system. For a frequency of 4 kHz the predicted neural patterns for a channel-number stimulation-rate configuration of 2 - 7 200 pps suggested that although there is no travelling wave delay information, the predicted neural patterns still contain temporal information. The predicted ISI histograms show peaks at the input tone period and multiples thereof, with clusters of spikes evident around the tone period in the predicted spatio-temporal neural spike train patterns. Similar peaks at the tone period were observed for calculated ISI histograms for predicted normal hearing neural patterns and measured neural responses. The predicted spatio-temporal neural patterns for the input frequency of 200 Hz show the travelling wave delay with clusters of spikes at the tone period. This travelling wave delay can also be seen from predicted normal hearing neural responses. The current spread, however, shows a significant distortion effect around the characteristic frequency place where the travelling wave delay increases rapidly. Spacing electrodes more closely results in an increase in this distortion, with the nerve fibre threshold decreasing in adjacent populations of nerve fibres, increasing the probability of firing. The current spread showed a more limited distortion effect on travelling wave delays when electrodes were spaced across the cochlea, at an electrode spacing of 6.08 mm. ISI histogram results also showed increased peaks around the tone period and multiples thereof. These predicted neural spike train patterns suggest that travelling waves in processing strategies, although mostly ignored, might provide the auditory system with both the spatial and temporal information needed for better pitch perception. / Dissertation (MEng)--University of Pretoria, 2012. / Electrical, Electronic and Computer Engineering / MEng / Unrestricted

Die vorliegende Arbeit beschäftigt sich mit der Analyse des Volleyballangriffsschlages von der Position vier unter Wettkampfbedingungen. Der Angriffsschlag von dieser Position ist oft die spielentscheidende Einflußgröße, da die meisten Punkte von dieser Position aus erzielt werden. Es handelt sich um einen komplexen Bewegungsablauf und es gibt eine Vielzahl an Untersuchungen, die sich mit der bewegungsanalytischen Untersuchung des Angriffsschlages befassen. Eine Herausforderung der generellen Problemstellung liegt darin, den Bewegungsablauf adäquat zu parametrisieren. Verschiedene Studien beschäftigten sich beispielsweise ausschließllich mit der Armbewegung während der Schlagphase oder mit der Beinbewegung während der Absprungphase. Die Dissertation ist darauf ausgerichtet eine breite Datenbasis für die Analyse von Volleyballangriffsschlägen zu schaffen. Der innovative Charakter der Arbeit liegt dabei in drei wesentlichen Punkten: - Definition leistungsrelevanter Parameter - Analyse von Angriffsschlägen unter Wettkampfbedingungen . große Stichprobe Insbesondere die Analyse von Angriffsschlägen während internationaler Wettkämpfe stellt dabei eine Herausforderung dar und hebt diese Arbeit von anderen Studien in der Literatur ab. Einerseits bietet dieser Ansatz die Möglichkeit "reale" Bewegungsabläufe im Spiel zu betrachten, andererseits verringert sich dadurch die Standardisierbarkeit der Umgebungsbedingungen. Die methodische Innovation liegt darin, zu untersuchen, welche Bewegungsabläufe unter echten Wettkampfbedingungen ausgeführt werden. Die wissenschaftliche Innovation liegt in der Identifikation und Definition leistungsrelevanter Parameter, die den Bewegungsablauf quantifizieren können. Damit kann ein Einblick gewonnen werden, was unter Spielbedingungen einen erfolgreichen Angriffsschlag ausmacht.