New kernel functions and learning methods for text and data mining

Recent advances in machine learning methods enable increasingly the automatic construction of various types of computer assisted methods that have been difficult or laborious to program by human experts. The tasks for which this kind of tools are needed arise in many areas, here especially in the fields of bioinformatics and natural language processing. The machine learning methods may not work satisfactorily if they are not appropriately tailored to the task in question. However, their learning performance can often be improved by taking advantage of deeper insight of the application domain or the learning problem at hand. This thesis considers developing kernel-based learning algorithms incorporating this kind of prior knowledge of the task in question in an advantageous way. Moreover, computationally efficient algorithms for training the learning machines for specific tasks are presented. In the context of kernel-based learning methods, the incorporation of prior knowledge is often done by designing appropriate kernel functions. Another well-known way is to develop cost functions that fit to the task under consideration. For disambiguation tasks in natural language, we develop kernel functions that take account of the positional information and the mutual similarities of words. It is shown that the use of this information significantly improves the disambiguation performance of the learning machine. Further, we design a new cost function that is better suitable for the task of information retrieval and for more general ranking problems than the cost functions designed for regression and classification. We also consider other applications of the kernel-based learning algorithms such as text categorization, and pattern recognition in differential display. We develop computationally efficient algorithms for training the considered learning machines with the proposed kernel functions. We also design a fast cross-validation algorithm for regularized least-squares type of learning algorithm. Further, an efficient version of the regularized least-squares algorithm that can be used together with the new cost function for preference learning and ranking tasks is proposed. In summary, we demonstrate that the incorporation of prior knowledge is possible and beneficial, and novel advanced kernels and cost functions can be used in algorithms efficiently.

Development of phase retrieval algorithm and techniques for optical spectrum analysis

Coherent anti-Stokes Raman scattering is the powerful method of laser spectroscopy in which significant successes are achieved. However, the non-linear nature of CARS complicates the analysis of the received spectra. The objective of this Thesis is to develop a new phase retrieval algorithm for CARS. It utilizes the maximum entropy method and the new wavelet approach for spectroscopic background correction of a phase function. The method was developed to be easily automated and used on a large number of spectra of different substances.. The algorithm was successfully tested on experimental data.

Investigation and data adaptive analysis of CARS spectra

In this diploma work advantages of coherent anti-Stokes Raman scattering spectrometry (CARS) and various methods of the quantitative analysis of substance structure with its help are considered. The basic methods and concepts of the adaptive analysis are adduced. On the basis of these methods the algorithm of automatic measurement of a scattering strip size of a target component in CARS spectrum is developed. The algorithm uses known full spectrum of target substance and compares it with a CARS spectrum. The form of a differential spectrum is used as a feedback to control the accuracy of matching. To exclude the influence of a background in CARS spectra the differential spectrum is analysed by means of its second derivative. The algorithm is checked up on the simulated simple spectra and on the spectra of organic compounds received experimentally.

Efficient total variation algorithm for fetal MRI reconstruction

Fetal MRI reconstruction aims at finding a high-resolution image given a small set of low-resolution images. It is usually modeled as an inverse problem where the regularization term plays a central role in the reconstruction quality. Literature has considered several regularization terms s.a. Dirichlet/Laplacian energy [1], Total Variation (TV)based energies [2,3] and more recently non-local means [4]. Although TV energies are quite attractive because of their ability in edge preservation, standard explicit steepest gradient techniques have been applied to optimize fetal-based TV energies. The main contribution of this work lies in the introduction of a well-posed TV algorithm from the point of view of convex optimization. Specifically, our proposed TV optimization algorithm for fetal reconstruction is optimal w.r.t. the asymptotic and iterative convergence speeds O(1/n(2)) and O(1/root epsilon), while existing techniques are in O(1/n) and O(1/epsilon). We apply our algorithm to (1) clinical newborn data, considered as ground truth, and (2) clinical fetal acquisitions. Our algorithm compares favorably with the literature in terms of speed and accuracy.

Pose-based SLAM with probabilistic scan matching algorithm using a mechanical scanned imaging sonar

This paper proposes a pose-based algorithm to solve the full SLAM problem for an autonomous underwater vehicle (AUV), navigating in an unknown and possibly unstructured environment. The technique incorporate probabilistic scan matching with range scans gathered from a mechanical scanning imaging sonar (MSIS) and the robot dead-reckoning displacements estimated from a Doppler velocity log (DVL) and a motion reference unit (MRU). The proposed method utilizes two extended Kalman filters (EKF). The first, estimates the local path travelled by the robot while grabbing the scan as well as its uncertainty and provides position estimates for correcting the distortions that the vehicle motion produces in the acoustic images. The second is an augment state EKF that estimates and keeps the registered scans poses. The raw data from the sensors are processed and fused in-line. No priory structural information or initial pose are considered. The algorithm has been tested on an AUV guided along a 600 m path within a marina environment, showing the viability of the proposed approach

Niche Differentiation and Fine-Scale Projections for Argentine Ants Based on Remotely Sensed Data

Modeling ecological niches of species is a promising approach for predicting the geographic potential of invasive species in new environments. Argentine ants (Linepithema humile) rank among the most successful invasive species: native to South America, they have invaded broad areas worldwide. Despite their widespread success, little is known about what makes an area susceptible - or not - to invasion. Here, we use a genetic algorithm approach to ecological niche modeling based on high-resolution remote-sensing data to examine the roles of niche similarity and difference in predicting invasions by this species. Our comparisons support a picture of general conservatism of the species' ecological characteristics, in spite of distinct geographic and community contexts

Analysis of Satellite Data for Forest Segmentation

In this work we study the classification of forest types using mathematics based image analysis on satellite data. We are interested in improving classification of forest segments when a combination of information from two or more different satellites is used. The experimental part is based on real satellite data originating from Canada. This thesis gives summary of the mathematics basics of the image analysis and supervised learning , methods that are used in the classification algorithm. Three data sets and four feature sets were investigated in this thesis. The considered feature sets were 1) histograms (quantiles) 2) variance 3) skewness and 4) kurtosis. Good overall performances were achieved when a combination of ASTERBAND and RADARSAT2 data sets was used.

Towards an Integrative Information Society: Studies on Individuality in Speech and Sign

The flow of information within modern information society has increased rapidly over the last decade. The major part of this information flow relies on the individual’s abilities to handle text or speech input. For the majority of us it presents no problems, but there are some individuals who would benefit from other means of conveying information, e.g. signed information flow. During the last decades the new results from various disciplines have all suggested towards the common background and processing for sign and speech and this was one of the key issues that I wanted to investigate further in this thesis. The basis of this thesis is firmly within speech research and that is why I wanted to design analogous test batteries for widely used speech perception tests for signers – to find out whether the results for signers would be the same as in speakers’ perception tests. One of the key findings within biology – and more precisely its effects on speech and communication research – is the mirror neuron system. That finding has enabled us to form new theories about evolution of communication, and it all seems to converge on the hypothesis that all communication has a common core within humans. In this thesis speech and sign are discussed as equal and analogical counterparts of communication and all research methods used in speech are modified for sign. Both speech and sign are thus investigated using similar test batteries. Furthermore, both production and perception of speech and sign are studied separately. An additional framework for studying production is given by gesture research using cry sounds. Results of cry sound research are then compared to results from children acquiring sign language. These results show that individuality manifests itself from very early on in human development. Articulation in adults, both in speech and sign, is studied from two perspectives: normal production and re-learning production when the apparatus has been changed. Normal production is studied both in speech and sign and the effects of changed articulation are studied with regards to speech. Both these studies are done by using carrier sentences. Furthermore, sign production is studied giving the informants possibility for spontaneous speech. The production data from the signing informants is also used as the basis for input in the sign synthesis stimuli used in sign perception test battery. Speech and sign perception were studied using the informants’ answers to questions using forced choice in identification and discrimination tasks. These answers were then compared across language modalities. Three different informant groups participated in the sign perception tests: native signers, sign language interpreters and Finnish adults with no knowledge of any signed language. This gave a chance to investigate which of the characteristics found in the results were due to the language per se and which were due to the changes in modality itself. As the analogous test batteries yielded similar results over different informant groups, some common threads of results could be observed. Starting from very early on in acquiring speech and sign the results were highly individual. However, the results were the same within one individual when the same test was repeated. This individuality of results represented along same patterns across different language modalities and - in some occasions - across language groups. As both modalities yield similar answers to analogous study questions, this has lead us to providing methods for basic input for sign language applications, i.e. signing avatars. This has also given us answers to questions on precision of the animation and intelligibility for the users – what are the parameters that govern intelligibility of synthesised speech or sign and how precise must the animation or synthetic speech be in order for it to be intelligible. The results also give additional support to the well-known fact that intelligibility in fact is not the same as naturalness. In some cases, as shown within the sign perception test battery design, naturalness decreases intelligibility. This also has to be taken into consideration when designing applications. All in all, results from each of the test batteries, be they for signers or speakers, yield strikingly similar patterns, which would indicate yet further support for the common core for all human communication. Thus, we can modify and deepen the phonetic framework models for human communication based on the knowledge obtained from the results of the test batteries within this thesis.

Solving Leontief Input-Output Model with Trapezoidal Fuzzy Numbers and Gauss-Seidel Algorithm

In this work a fuzzy linear system is used to solve Leontief input-output model with fuzzy entries. For solving this model, we assume that the consumption matrix from di erent sectors of the economy and demand are known. These assumptions heavily depend on the information obtained from the industries. Hence uncertainties are involved in this information. The aim of this work is to model these uncertainties and to address them by fuzzy entries such as fuzzy numbers and LR-type fuzzy numbers (triangular and trapezoidal). Fuzzy linear system has been developed using fuzzy data and it is solved using Gauss-Seidel algorithm. Numerical examples show the e ciency of this algorithm. The famous example from Prof. Leontief, where he solved the production levels for U.S. economy in 1958, is also further analyzed.

Mental Well-being at Work. A Sign of a Healthy Organisation and a Necessary Precondition for Organisational Development

One of the aims of the study was to clarify the reliability and validity of the Job Diagnostic Survey (JDS) and the Eigenzustand (EZ) method as measures of the objective characteristics of work and short-term mental work load in the Finnish data. The reliability and validity were examined taking into consideration the theoretical backgrounds of the methods and the reliability of the measurements. The methods were used for finding out the preconditions for organisational development based on self-improvement and clarifying the impacts of working environment (organisational functioning and job characteristics) on a worker’s mental state and health. The influences were examined on a general level - regardless of individual personal or specific contextual factors. One aim was also to clarify how cognitions and emotions are intertwined and how they influence a person’s perception of the working environment. The data consisted of 15 blue-collar organisations in the public sector. The organisations were divided in target and comparison groups depending on the research frames. The data was collected by questionnaires by post. The exploratory and confirmatory factor analyses (Lisrel) were used as the main statistical methods in examining the structures of the methods and impacts between the variables. It was shown that it is possible for organisations to develop their working conditions themselves on specific preconditions. The advance of the development processes could be shown by the amount of the development activity as well as by the changes of the mental well-being (ability to act) and sick absenteeism of the personnel. It was found that the JDS and the EZ methods were reliable and valid measures in the Finnish data. It was shown that, in addition to the objective working environment (organisational functioning and job characteristics), also such a personal factor as selfesteem influences a person’s perception of mental work load. However, the influence did not seem to be direct. The importance of job satisfaction as a general indicator of perceived working conditions was emphasised. Emotional and cognitive factors were found to be functionally intertwined constituting a common factor. Organisational functioning and the characteristics of work had connections with a person’s health measured by sick absenteeism.

Use of data mining and spectral profiles to differentiate condition after harvest of coffee plants

This study aimed at identifying different conditions of coffee plants after harvesting period, using data mining and spectral behavior profiles from Hyperion/EO1 sensor. The Hyperion image, with spatial resolution of 30 m, was acquired in August 28th, 2008, at the end of the coffee harvest season in the studied area. For pre-processing imaging, atmospheric and signal/noise effect corrections were carried out using Flaash and MNF (Minimum Noise Fraction Transform) algorithms, respectively. Spectral behavior profiles (38) of different coffee varieties were generated from 150 Hyperion bands. The spectral behavior profiles were analyzed by Expectation-Maximization (EM) algorithm considering 2; 3; 4 and 5 clusters. T-test with 5% of significance was used to verify the similarity among the wavelength cluster means. The results demonstrated that it is possible to separate five different clusters, which were comprised by different coffee crop conditions making possible to improve future intervention actions.

Using data mining to identify factors that influence the degree of leg injuries in broilers

Locomotor problems prevent the bird to move freely, jeopardizing the welfare and productivity, besides generating injuries on the legs of chickens. The objective of this study was to evaluate the influence of age, use of vitamin D, the asymmetry of limbs and gait score, the degree of leg injuries in broilers, using data mining. The analysis was performed on a data set obtained from a field experiment in which it was used two groups of birds with 30 birds each, a control group and one treated with vitamin D. It was evaluated the gait score, the asymmetry between the right and left toes, and the degree of leg injuries. The Weka ® software was used in data mining. In particular, C4.5 algorithm (also known as J48 in Weka environment) was used for the generation of a decision tree. The results showed that age is the factor that most influences the degree of leg injuries and that the data from assessments of gait score were not reliable to estimate leg weakness in broilers.

Software for calculation of reservoir active capacity with the sequent-peak algorithm

It is presented a software developed with Delphi programming language to compute the reservoir's annual regulated active storage, based on the sequent-peak algorithm. Mathematical models used for that purpose generally require extended hydrological series. Usually, the analysis of those series is performed with spreadsheets or graphical representations. Based on that, it was developed a software for calculation of reservoir active capacity. An example calculation is shown by 30-years (from 1977 to 2009) monthly mean flow historical data, from Corrente River, located at São Francisco River Basin, Brazil. As an additional tool, an interface was developed to manage water resources, helping to manipulate data and to point out information that it would be of interest to the user. Moreover, with that interface irrigation districts where water consumption is higher can be analyzed as a function of specific seasonal water demands situations. From a practical application, it is possible to conclude that the program provides the calculation originally proposed. It was designed to keep information organized and retrievable at any time, and to show simulation on seasonal water demands throughout the year, contributing with the elements of study concerning reservoir projects. This program, with its functionality, is an important tool for decision making in the water resources management.

Data mining techniques for identification of spectrally homogeneous areas using NDVI temporal profiles of soybean crop

The aim of this study was to group temporal profiles of 10-day composites NDVI product by similarity, which was obtained by the SPOT Vegetation sensor, for municipalities with high soybean production in the state of Paraná, Brazil, in the 2005/2006 cropping season. Data mining is a valuable tool that allows extracting knowledge from a database, identifying valid, new, potentially useful and understandable patterns. Therefore, it was used the methods for clusters generation by means of the algorithms K-Means, MAXVER and DBSCAN, implemented in the WEKA software package. Clusters were created based on the average temporal profiles of NDVI of the 277 municipalities with high soybean production in the state and the best results were found with the K-Means algorithm, grouping the municipalities into six clusters, considering the period from the beginning of October until the end of March, which is equivalent to the crop vegetative cycle. Half of the generated clusters presented spectro-temporal pattern, a characteristic of soybeans and were mostly under the soybean belt in the state of Paraná, which shows good results that were obtained with the proposed methodology as for identification of homogeneous areas. These results will be useful for the creation of regional soybean "masks" to estimate the planted area for this crop.

Vocalization data mining for estimating swine stress conditions

This study aimed to identify differences in swine vocalization pattern according to animal gender and different stress conditions. A total of 150 barrow males and 150 females (Dalland® genetic strain), aged 100 days, were used in the experiment. Pigs were exposed to different stressful situations: thirst (no access to water), hunger (no access to food), and thermal stress (THI exceeding 74). For the control treatment, animals were kept under a comfort situation (animals with full access to food and water, with environmental THI lower than 70). Acoustic signals were recorded every 30 minutes, totaling six samples for each stress situation. Afterwards, the audios were analyzed by Praat® 5.1.19 software, generating a sound spectrum. For determination of stress conditions, data were processed by WEKA® 3.5 software, using the decision tree algorithm C4.5, known as J48 in the software environment, considering cross-validation with samples of 10% (10-fold cross-validation). According to the Decision Tree, the acoustic most important attribute for the classification of stress conditions was sound Intensity (root node). It was not possible to identify, using the tested attributes, the animal gender by vocal register. A decision tree was generated for recognition of situations of swine hunger, thirst, and heat stress from records of sound intensity, Pitch frequency, and Formant 1.