Machine learning techniques for implicit interaction using mobile sensors

Interactions in mobile devices normally happen in an explicit manner, which means that they are initiated by the users. Yet, users are typically unaware that they also interact implicitly with their devices. For instance, our hand pose changes naturally when we type text messages. Whilst the touchscreen captures finger touches, hand movements during this interaction however are unused. If this implicit hand movement is observed, it can be used as additional information to support or to enhance the users’ text entry experience. This thesis investigates how implicit sensing can be used to improve existing, standard interaction technique qualities. In particular, this thesis looks into enhancing front-of-device interaction through back-of-device and hand movement implicit sensing. We propose the investigation through machine learning techniques. We look into problems on how sensor data via implicit sensing can be used to predict a certain aspect of an interaction. For instance, one of the questions that this thesis attempts to answer is whether hand movement during a touch targeting task correlates with the touch position. This is a complex relationship to understand but can be best explained through machine learning. Using machine learning as a tool, such correlation can be measured, quantified, understood and used to make predictions on future touch position. Furthermore, this thesis also evaluates the predictive power of the sensor data. We show this through a number of studies. In Chapter 5 we show that probabilistic modelling of sensor inputs and recorded touch locations can be used to predict the general area of future touches on touchscreen. In Chapter 7, using SVM classifiers, we show that data from implicit sensing from general mobile interactions is user-specific. This can be used to identify users implicitly. In Chapter 6, we also show that touch interaction errors can be detected from sensor data. In our experiment, we show that there are sufficient distinguishable patterns between normal interaction signals and signals that are strongly correlated with interaction error. In all studies, we show that performance gain can be achieved by combining sensor inputs.

Hierarchical modelling of timber reference properties using probabilistic methods: Maximum Likelihood Method, Bayesian Methods and Probability Networks

In recent decades, an increased interest has been evidenced in the research on multi-scale hierarchical modelling in the field of mechanics, and also in the field of wood products and timber engineering. One of the main motivations for hierar-chical modelling is to understand how properties, composition and structure at lower scale levels may influence and be used to predict the material properties on a macroscopic and structural engineering scale. This chapter presents the applicability of statistic and probabilistic methods, such as the Maximum Likelihood method and Bayesian methods, in the representation of timber’s mechanical properties and its inference accounting to prior information obtained in different importance scales. These methods allow to analyse distinct timber’s reference properties, such as density, bending stiffness and strength, and hierarchically consider information obtained through different non, semi or destructive tests. The basis and fundaments of the methods are described and also recommendations and limitations are discussed. The methods may be used in several contexts, however require an expert’s knowledge to assess the correct statistic fitting and define the correlation arrangement between properties.

Combining probabilistic land-use change and tree population dynamics modelling to simulate responses in mountain forests

Altitudinal tree lines are mainly constrained by temperature, but can also be influenced by factors such as human activity, particularly in the European Alps, where centuries of agricultural use have affected the tree-line. Over the last decades this trend has been reversed due to changing agricultural practices and land-abandonment. We aimed to combine a statistical land-abandonment model with a forest dynamics model, to take into account the combined effects of climate and human land-use on the Alpine tree-line in Switzerland. Land-abandonment probability was expressed by a logistic regression function of degree-day sum, distance from forest edge, soil stoniness, slope, proportion of employees in the secondary and tertiary sectors, proportion of commuters and proportion of full-time farms. This was implemented in the TreeMig spatio-temporal forest model. Distance from forest edge and degree-day sum vary through feed-back from the dynamics part of TreeMig and climate change scenarios, while the other variables remain constant for each grid cell over time. The new model, TreeMig-LAb, was tested on theoretical landscapes, where the variables in the land-abandonment model were varied one by one. This confirmed the strong influence of distance from forest and slope on the abandonment probability. Degree-day sum has a more complex role, with opposite influences on land-abandonment and forest growth. TreeMig-LAb was also applied to a case study area in the Upper Engadine (Swiss Alps), along with a model where abandonment probability was a constant. Two scenarios were used: natural succession only (100% probability) and a probability of abandonment based on past transition proportions in that area (2.1% per decade). The former showed new forest growing in all but the highest-altitude locations. The latter was more realistic as to numbers of newly forested cells, but their location was random and the resulting landscape heterogeneous. Using the logistic regression model gave results consistent with observed patterns of land-abandonment: existing forests expanded and gaps closed, leading to an increasingly homogeneous landscape.

Probabilistic failure modelling of reinforced concrete structures subjected to chloride penetration

Structural durability is an important criterion that must be evaluated for every type of structure. Concerning reinforced concrete members, chloride diffusion process is widely used to evaluate durability, especially when these structures are constructed in aggressive atmospheres. The chloride ingress triggers the corrosion of reinforcements; therefore, by modelling this phenomenon, the corrosion process can be better evaluated as well as the structural durability. The corrosion begins when a threshold level of chloride concentration is reached at the steel bars of reinforcements. Despite the robustness of several models proposed in literature, deterministic approaches fail to predict accurately the corrosion time initiation due the inherent randomness observed in this process. In this regard, structural durability can be more realistically represented using probabilistic approaches. This paper addresses the analyses of probabilistic corrosion time initiation in reinforced concrete structures exposed to chloride penetration. The chloride penetration is modelled using the Fick's diffusion law. This law simulates the chloride diffusion process considering time-dependent effects. The probability of failure is calculated using Monte Carlo simulation and the first order reliability method, with a direct coupling approach. Some examples are considered in order to study these phenomena. Moreover, a simplified method is proposed to determine optimal values for concrete cover.

Probabilistic world modelling for distributed team planning

This paper describes an application of decoupled probabilistic world modeling to achieve team planning. The research is based on the principle that tbe action selection mechanism of a member in a robot team cm select am effective action if a global world model is available to all team members. In the real world, the sensors are imprecise, and are individual to each robot, hence providing each robot a partial and unique view about the environment. We address this problem by creating a probabilistic global view on each agent by combining the perceptual information from each robot. This probsbilistie view forms the basis for selecting actions to achieve the team goal in a dynamic environment. Experiments have been carried ont to investigate the effectiveness of this principle using custom-built robots for real world performance, in addition, to extensive simulation results. The results show an improvement in team effectiveness when using probabilistic world modeling based on perception sharing for team planning.

A Bayesian state space modelling approach to probabilistic quantitative precipitation forecasting

The generation of very short range forecasts of precipitation in the 0-6 h time window is traditionally referred to as nowcasting. Most existing nowcasting systems essentially extrapolate radar observations in some manner, however, very few systems account for the uncertainties involved. Thus deterministic forecast are produced, which have a limited use when decisions must be made, since they have no measure of confidence or spread of the forecast. This paper develops a Bayesian state space modelling framework for quantitative precipitation nowcasting which is probabilistic from conception. The model treats the observations (radar) as noisy realisations of the underlying true precipitation process, recognising that this process can never be completely known, and thus must be represented probabilistically. In the model presented here the dynamics of the precipitation are dominated by advection, so this is a probabilistic extrapolation forecast. The model is designed in such a way as to minimise the computational burden, while maintaining a full, joint representation of the probability density function of the precipitation process. The update and evolution equations avoid the need to sample, thus only one model needs be run as opposed to the more traditional ensemble route. It is shown that the model works well on both simulated and real data, but that further work is required before the model can be used operationally. © 2004 Elsevier B.V. All rights reserved.

Probabilistic Graphical Modelling for Software Product Lines: A Frameweork for Modeling and Reasoning under Uncertainty

This work provides a holistic investigation into the realm of feature modeling within software product lines. The work presented identifies limitations and challenges within the current feature modeling approaches. Those limitations include, but not limited to, the dearth of satisfactory cognitive presentation, inconveniency in scalable systems, inflexibility in adapting changes, nonexistence of predictability of models behavior, as well as the lack of probabilistic quantification of model’s implications and decision support for reasoning under uncertainty. The work in this thesis addresses these challenges by proposing a series of solutions. The first solution is the construction of a Bayesian Belief Feature Model, which is a novel modeling approach capable of quantifying the uncertainty measures in model parameters by a means of incorporating probabilistic modeling with a conventional modeling approach. The Bayesian Belief feature model presents a new enhanced feature modeling approach in terms of truth quantification and visual expressiveness. The second solution takes into consideration the unclear support for the reasoning under the uncertainty process, and the challenging constraint satisfaction problem in software product lines. This has been done through the development of a mathematical reasoner, which was designed to satisfy the model constraints by considering probability weight for all involved parameters and quantify the actual implications of the problem constraints. The developed Uncertain Constraint Satisfaction Problem approach has been tested and validated through a set of designated experiments. Profoundly stating, the main contributions of this thesis include the following: • Develop a framework for probabilistic graphical modeling to build the purported Bayesian belief feature model. • Extend the model to enhance visual expressiveness throughout the integration of colour degree variation; in which the colour varies with respect to the predefined probabilistic weights. • Enhance the constraints satisfaction problem by the uncertainty measuring of the parameters truth assumption. • Validate the developed approach against different experimental settings to determine its functionality and performance.

A reference business process for ecological niche modelling

Ecological niche modelling combines species occurrence points with environmental raster layers in order to obtain models for describing the probabilistic distribution of species. The process to generate an ecological niche model is complex. It requires dealing with a large amount of data, use of different software packages for data conversion, for model generation and for different types of processing and analyses, among other functionalities. A software platform that integrates all requirements under a single and seamless interface would be very helpful for users. Furthermore, since biodiversity modelling is constantly evolving, new requirements are constantly being added in terms of functions, algorithms and data formats. This evolution must be accompanied by any software intended to be used in this area. In this scenario, a Service-Oriented Architecture (SOA) is an appropriate choice for designing such systems. According to SOA best practices and methodologies, the design of a reference business process must be performed prior to the architecture definition. The purpose is to understand the complexities of the process (business process in this context refers to the ecological niche modelling problem) and to design an architecture able to offer a comprehensive solution, called a reference architecture, that can be further detailed when implementing specific systems. This paper presents a reference business process for ecological niche modelling, as part of a major work focused on the definition of a reference architecture based on SOA concepts that will be used to evolve the openModeller software package for species modelling. The basic steps that are performed while developing a model are described, highlighting important aspects, based on the knowledge of modelling experts. In order to illustrate the steps defined for the process, an experiment was developed, modelling the distribution of Ouratea spectabilis (Mart.) Engl. (Ochnaceae) using openModeller. As a consequence of the knowledge gained with this work, many desirable improvements on the modelling software packages have been identified and are presented. Also, a discussion on the potential for large-scale experimentation in ecological niche modelling is provided, highlighting opportunities for research. The results obtained are very important for those involved in the development of modelling tools and systems, for requirement analysis and to provide insight on new features and trends for this category of systems. They can also be very helpful for beginners in modelling research, who can use the process and the experiment example as a guide to this complex activity. (c) 2008 Elsevier B.V. All rights reserved.

Modelling of microstructure formation and evolution during solidification

A comprehensive probabilistic model for simulating microstructure formation and evolution during solidification has been developed, based on coupling a Finite Differential Method (FDM) for macroscopic modelling of heat diffusion to a modified Cellular Automaton (mCA) for microscopic modelling of nucleation, growth of microstructures and solute diffusion. The mCA model is similar to Nastac's model for handling solute redistribution in the liquid and solid phases, curvature and growth anisotropy, but differs in the treatment of nucleation and growth. The aim is to improve understanding of the relationship between the solidification conditions and microstructure formation and evolution. A numerical algorithm used for FDM and mCA was developed. At each coarse scale, temperatures at FDM nodes were calculated while nucleation-growth simulation was done at a finer scale, with the temperature at the cell locations being interpolated from those at the coarser volumes. This model takes account of thermal, curvature and solute diffusion effects. Therefore, it can not only simulate microstructures of alloys both on the scale of grain size (macroscopic level) and the dendrite tip length (mesoscopic level), but also investigate nucleation mechanisms and growth kinetics of alloys solidified with various solute concentrations and solidification morphologies. The calculated results are compared with values of grain sizes and solidification morphologies of microstructures obtained from a set of casting experiments of Al-Si alloys in graphite crucibles.

A probabilistic analysis of Clostridium perfringens growth during food service operations

The purpose of this study was threefold: first, the study was designed to illustrate the use of data and information collected in food safety surveys in a quantitative risk assessment. In this case, the focus was on the food service industry; however, similar data from other parts of the food chain could be similarly incorporated. The second objective was to quantitatively describe and better understand the role that the food service industry plays in the safety of food. The third objective was to illustrate the additional decision-making information that is available when uncertainty and variability are incorporated into the modelling of systems. (C) 2002 Elsevier Science B.V. All rights reserved.

Sound Texture Modelling

In the PhD thesis “Sound Texture Modeling” we deal with statistical modelling or textural sounds like water, wind, rain, etc. For synthesis and classification. Our initial model is based on a wavelet tree signal decomposition and the modeling of the resulting sequence by means of a parametric probabilistic model, that can be situated within the family of models trainable via expectation maximization (hidden Markov tree model ). Our model is able to capture key characteristics of the source textures (water, rain, fire, applause, crowd chatter ), and faithfully reproduces some of the sound classes. In terms of a more general taxonomy of natural events proposed by Graver, we worked on models for natural event classification and segmentation. While the event labels comprise physical interactions between materials that do not have textural propierties in their enterity, those segmentation models can help in identifying textural portions of an audio recording useful for analysis and resynthesis. Following our work on concatenative synthesis of musical instruments, we have developed a pattern-based synthesis system, that allows to sonically explore a database of units by means of their representation in a perceptual feature space. Concatenative syntyhesis with “molecules” built from sparse atomic representations also allows capture low-level correlations in perceptual audio features, while facilitating the manipulation of textural sounds based on their physical and perceptual properties. We have approached the problem of sound texture modelling for synthesis from different directions, namely a low-level signal-theoretic point of view through a wavelet transform, and a more high-level point of view driven by perceptual audio features in the concatenative synthesis setting. The developed framework provides unified approach to the high-quality resynthesis of natural texture sounds. Our research is embedded within the Metaverse 1 European project (2008-2011), where our models are contributting as low level building blocks within a semi-automated soundscape generation system.

Analysis, modelling and classification of geospatial data using machine learning

The research considers the problem of spatial data classification using machine learning algorithms: probabilistic neural networks (PNN) and support vector machines (SVM). As a benchmark model simple k-nearest neighbor algorithm is considered. PNN is a neural network reformulation of well known nonparametric principles of probability density modeling using kernel density estimator and Bayesian optimal or maximum a posteriori decision rules. PNN is well suited to problems where not only predictions but also quantification of accuracy and integration of prior information are necessary. An important property of PNN is that they can be easily used in decision support systems dealing with problems of automatic classification. Support vector machine is an implementation of the principles of statistical learning theory for the classification tasks. Recently they were successfully applied for different environmental topics: classification of soil types and hydro-geological units, optimization of monitoring networks, susceptibility mapping of natural hazards. In the present paper both simulated and real data case studies (low and high dimensional) are considered. The main attention is paid to the detection and learning of spatial patterns by the algorithms applied.

A probabilistic approach to niche-based community models for spatial forecasts of assemblage properties and their uncertainties

Aim Conservation strategies are in need of predictions that capture spatial community composition and structure. Currently, the methods used to generate these predictions generally focus on deterministic processes and omit important stochastic processes and other unexplained variation in model outputs. Here we test a novel approach of community models that accounts for this variation and determine how well it reproduces observed properties of alpine butterfly communities. Location The western Swiss Alps. Methods We propose a new approach to process probabilistic predictions derived from stacked species distribution models (S-SDMs) in order to predict and assess the uncertainty in the predictions of community properties. We test the utility of our novel approach against a traditional threshold-based approach. We used mountain butterfly communities spanning a large elevation gradient as a case study and evaluated the ability of our approach to model species richness and phylogenetic diversity of communities. Results S-SDMs reproduced the observed decrease in phylogenetic diversity and species richness with elevation, syndromes of environmental filtering. The prediction accuracy of community properties vary along environmental gradient: variability in predictions of species richness was higher at low elevation, while it was lower for phylogenetic diversity. Our approach allowed mapping the variability in species richness and phylogenetic diversity projections. Main conclusion Using our probabilistic approach to process species distribution models outputs to reconstruct communities furnishes an improved picture of the range of possible assemblage realisations under similar environmental conditions given stochastic processes and help inform manager of the uncertainty in the modelling results

Quantitative modelling of transcription factor binding sites

Abstract One of the most important issues in molecular biology is to understand regulatory mechanisms that control gene expression. Gene expression is often regulated by proteins, called transcription factors which bind to short (5 to 20 base pairs),degenerate segments of DNA. Experimental efforts towards understanding the sequence specificity of transcription factors is laborious and expensive, but can be substantially accelerated with the use of computational predictions. This thesis describes the use of algorithms and resources for transcriptionfactor binding site analysis in addressing quantitative modelling, where probabilitic models are built to represent binding properties of a transcription factor and can be used to find new functional binding sites in genomes. Initially, an open-access database(HTPSELEX) was created, holding high quality binding sequences for two eukaryotic families of transcription factors namely CTF/NF1 and LEFT/TCF. The binding sequences were elucidated using a recently described experimental procedure called HTP-SELEX, that allows generation of large number (> 1000) of binding sites using mass sequencing technology. For each HTP-SELEX experiments we also provide accurate primary experimental information about the protein material used, details of the wet lab protocol, an archive of sequencing trace files, and assembled clone sequences of binding sequences. The database also offers reasonably large SELEX libraries obtained with conventional low-throughput protocols.The database is available at http://wwwisrec.isb-sib.ch/htpselex/ and and ftp://ftp.isrec.isb-sib.ch/pub/databases/htpselex. The Expectation-Maximisation(EM) algorithm is one the frequently used methods to estimate probabilistic models to represent the sequence specificity of transcription factors. We present computer simulations in order to estimate the precision of EM estimated models as a function of data set parameters(like length of initial sequences, number of initial sequences, percentage of nonbinding sequences). We observed a remarkable robustness of the EM algorithm with regard to length of training sequences and the degree of contamination. The HTPSELEX database and the benchmarked results of the EM algorithm formed part of the foundation for the subsequent project, where a statistical framework called hidden Markov model has been developed to represent sequence specificity of the transcription factors CTF/NF1 and LEF1/TCF using the HTP-SELEX experiment data. The hidden Markov model framework is capable of both predicting and classifying CTF/NF1 and LEF1/TCF binding sites. A covariance analysis of the binding sites revealed non-independent base preferences at different nucleotide positions, providing insight into the binding mechanism. We next tested the LEF1/TCF model by computing binding scores for a set of LEF1/TCF binding sequences for which relative affinities were determined experimentally using non-linear regression. The predicted and experimentally determined binding affinities were in good correlation.

The probabilistic risk analysis of external hazards of an interim storage for spent nuclear fuel

Tässä diplomityössä tehtiin Olkiluodon ydinvoimalaitoksella sijaitsevan käytetyn ydinpolttoaineen allasvarastointiin perustuvan välivaraston todennäköisyysperustainen ulkoisten uhkien riskianalyysi. Todennäköisyysperustainen riskianalyysi (PRA) on yleisesti käytetty riskien tunnistus- ja lähestymistapa ydinvoimalaitoksella. Työn tarkoituksena oli laatia täysin uusi ulkoisten uhkien PRA-analyysi, koska Suomessa ei ole aiemmin tehty vastaavanlaisia tämän tutkimusalueen riskitarkasteluja. Riskitarkastelun motiivina ovat myös maailmalla tapahtuneiden luonnonkatastrofien vuoksi korostunut ulkoisten uhkien rooli käytetyn ydinpolttoaineen välivarastoinnin turvallisuudessa. PRA analyysin rakenne pohjautui tutkimuksen alussa luotuun metodologiaan. Analyysi perustuu mahdollisten ulkoisten uhkien tunnistamiseen pois lukien ihmisen aikaansaamat tahalliset vahingot. Tunnistettujen ulkoisten uhkien esiintymistaajuuksien ja vahingoittamispotentiaalin perusteella ulkoiset uhat joko karsittiin pois tutkimuksessa määriteltyjen karsintakriteerien avulla tai analysoitiin tarkemmin. Tutkimustulosten perusteella voitiin todeta, että tiedot hyvin harvoin tapahtuvista ulkoisista uhista ovat epätäydellisiä. Suurinta osaa näistä hyvin harvoin tapahtuvista ulkoisista uhista ei ole koskaan esiintynyt eikä todennäköisesti koskaan tule esiintymään Olkiluodon vaikutusalueella tai edes Suomessa. Esimerkiksi salaman iskujen ja öljyaltistuksen roolit ja vaikutukset erilaisten komponenttien käytettävyyteen ovat epävarmasti tunnettuja. Tutkimuksen tuloksia voidaan pitää kokonaisuudessaan merkittävinä, koska niiden perusteella voidaan osoittaa ne ulkoiset uhat, joiden vaikutuksia olisi syytä tutkia tarkemmin. Yksityiskohtaisempi tietoisuus hyvin harvoin esiintyvistä ulkoisista uhista tarkentaisi alkutapahtumataajuuksien estimaatteja.