New definitions of margin for multi-class Boosting algorithms

La familia de algoritmos de Boosting son un tipo de técnicas de clasificación y regresión que han demostrado ser muy eficaces en problemas de Visión Computacional. Tal es el caso de los problemas de detección, de seguimiento o bien de reconocimiento de caras, personas, objetos deformables y acciones. El primer y más popular algoritmo de Boosting, AdaBoost, fue concebido para problemas binarios. Desde entonces, muchas han sido las propuestas que han aparecido con objeto de trasladarlo a otros dominios más generales: multiclase, multilabel, con costes, etc. Nuestro interés se centra en extender AdaBoost al terreno de la clasificación multiclase, considerándolo como un primer paso para posteriores ampliaciones. En la presente tesis proponemos dos algoritmos de Boosting para problemas multiclase basados en nuevas derivaciones del concepto margen. El primero de ellos, PIBoost, está concebido para abordar el problema descomponiéndolo en subproblemas binarios. Por un lado, usamos una codificación vectorial para representar etiquetas y, por otro, utilizamos la función de pérdida exponencial multiclase para evaluar las respuestas. Esta codificación produce un conjunto de valores margen que conllevan un rango de penalizaciones en caso de fallo y recompensas en caso de acierto. La optimización iterativa del modelo genera un proceso de Boosting asimétrico cuyos costes dependen del número de etiquetas separadas por cada clasificador débil. De este modo nuestro algoritmo de Boosting tiene en cuenta el desbalanceo debido a las clases a la hora de construir el clasificador. El resultado es un método bien fundamentado que extiende de manera canónica al AdaBoost original. El segundo algoritmo propuesto, BAdaCost, está concebido para problemas multiclase dotados de una matriz de costes. Motivados por los escasos trabajos dedicados a generalizar AdaBoost al terreno multiclase con costes, hemos propuesto un nuevo concepto de margen que, a su vez, permite derivar una función de pérdida adecuada para evaluar costes. Consideramos nuestro algoritmo como la extensión más canónica de AdaBoost para este tipo de problemas, ya que generaliza a los algoritmos SAMME, Cost-Sensitive AdaBoost y PIBoost. Por otro lado, sugerimos un simple procedimiento para calcular matrices de coste adecuadas para mejorar el rendimiento de Boosting a la hora de abordar problemas estándar y problemas con datos desbalanceados. Una serie de experimentos nos sirven para demostrar la efectividad de ambos métodos frente a otros conocidos algoritmos de Boosting multiclase en sus respectivas áreas. En dichos experimentos se usan bases de datos de referencia en el área de Machine Learning, en primer lugar para minimizar errores y en segundo lugar para minimizar costes. Además, hemos podido aplicar BAdaCost con éxito a un proceso de segmentación, un caso particular de problema con datos desbalanceados. Concluimos justificando el horizonte de futuro que encierra el marco de trabajo que presentamos, tanto por su aplicabilidad como por su flexibilidad teórica. Abstract The family of Boosting algorithms represents a type of classification and regression approach that has shown to be very effective in Computer Vision problems. Such is the case of detection, tracking and recognition of faces, people, deformable objects and actions. The first and most popular algorithm, AdaBoost, was introduced in the context of binary classification. Since then, many works have been proposed to extend it to the more general multi-class, multi-label, costsensitive, etc... domains. Our interest is centered in extending AdaBoost to two problems in the multi-class field, considering it a first step for upcoming generalizations. In this dissertation we propose two Boosting algorithms for multi-class classification based on new generalizations of the concept of margin. The first of them, PIBoost, is conceived to tackle the multi-class problem by solving many binary sub-problems. We use a vectorial codification to represent class labels and a multi-class exponential loss function to evaluate classifier responses. This representation produces a set of margin values that provide a range of penalties for failures and rewards for successes. The stagewise optimization of this model introduces an asymmetric Boosting procedure whose costs depend on the number of classes separated by each weak-learner. In this way the Boosting procedure takes into account class imbalances when building the ensemble. The resulting algorithm is a well grounded method that canonically extends the original AdaBoost. The second algorithm proposed, BAdaCost, is conceived for multi-class problems endowed with a cost matrix. Motivated by the few cost-sensitive extensions of AdaBoost to the multi-class field, we propose a new margin that, in turn, yields a new loss function appropriate for evaluating costs. Since BAdaCost generalizes SAMME, Cost-Sensitive AdaBoost and PIBoost algorithms, we consider our algorithm as a canonical extension of AdaBoost to this kind of problems. We additionally suggest a simple procedure to compute cost matrices that improve the performance of Boosting in standard and unbalanced problems. A set of experiments is carried out to demonstrate the effectiveness of both methods against other relevant Boosting algorithms in their respective areas. In the experiments we resort to benchmark data sets used in the Machine Learning community, firstly for minimizing classification errors and secondly for minimizing costs. In addition, we successfully applied BAdaCost to a segmentation task, a particular problem in presence of imbalanced data. We conclude the thesis justifying the horizon of future improvements encompassed in our framework, due to its applicability and theoretical flexibility.

Self-Learning Embedded System for Object Identification in Intelligent Infrastructure Sensors

The emergence of new horizons in the field of travel assistant management leads to the development of cutting-edge systems focused on improving the existing ones. Moreover, new opportunities are being also presented since systems trend to be more reliable and autonomous. In this paper, a self-learning embedded system for object identification based on adaptive-cooperative dynamic approaches is presented for intelligent sensor’s infrastructures. The proposed system is able to detect and identify moving objects using a dynamic decision tree. Consequently, it combines machine learning algorithms and cooperative strategies in order to make the system more adaptive to changing environments. Therefore, the proposed system may be very useful for many applications like shadow tolls since several types of vehicles may be distinguished, parking optimization systems, improved traffic conditions systems, etc.

A methodology for improving computed individual regressions predictions.

This research proposes a methodology to improve computed individual prediction values provided by an existing regression model without having to change either its parameters or its architecture. In other words, we are interested in achieving more accurate results by adjusting the calculated regression prediction values, without modifying or rebuilding the original regression model. Our proposition is to adjust the regression prediction values using individual reliability estimates that indicate if a single regression prediction is likely to produce an error considered critical by the user of the regression. The proposed method was tested in three sets of experiments using three different types of data. The first set of experiments worked with synthetically produced data, the second with cross sectional data from the public data source UCI Machine Learning Repository and the third with time series data from ISO-NE (Independent System Operator in New England). The experiments with synthetic data were performed to verify how the method behaves in controlled situations. In this case, the outcomes of the experiments produced superior results with respect to predictions improvement for artificially produced cleaner datasets with progressive worsening with the addition of increased random elements. The experiments with real data extracted from UCI and ISO-NE were done to investigate the applicability of the methodology in the real world. The proposed method was able to improve regression prediction values by about 95% of the experiments with real data.

Evaluating knowledge-based approaches to the multilingual extension of a temporal expression normalizer

The extension to new languages is a well known bottleneck for rule-based systems. Considerable human effort, which typically consists in re-writing from scratch huge amounts of rules, is in fact required to transfer the knowledge available to the system from one language to a new one. Provided sufficient annotated data, machine learning algorithms allow to minimize the costs of such knowledge transfer but, up to date, proved to be ineffective for some specific tasks. Among these, the recognition and normalization of temporal expressions still remains out of their reach. Focusing on this task, and still adhering to the rule-based framework, this paper presents a bunch of experiments on the automatic porting to Italian of a system originally developed for Spanish. Different automatic rule translation strategies are evaluated and discussed, providing a comprehensive overview of the challenge.

Sentiment classification for early detection of health alerts in the chemical textile domain

In the chemical textile domain experts have to analyse chemical components and substances that might be harmful for their usage in clothing and textiles. Part of this analysis is performed searching opinions and reports people have expressed concerning these products in the Social Web. However, this type of information on the Internet is not as frequent for this domain as for others, so its detection and classification is difficult and time-consuming. Consequently, problems associated to the use of chemical substances in textiles may not be detected early enough, and could lead to health problems, such as allergies or burns. In this paper, we propose a framework able to detect, retrieve, and classify subjective sentences related to the chemical textile domain, that could be integrated into a wider health surveillance system. We also describe the creation of several datasets with opinions from this domain, the experiments performed using machine learning techniques and different lexical resources such as WordNet, and the evaluation focusing on the sentiment classification, and complaint detection (i.e., negativity). Despite the challenges involved in this domain, our approach obtains promising results with an F-score of 65% for polarity classification and 82% for complaint detection.

Computational Methods for Identifying Hypertrophic Cardiomyopathy using 12-lead ECG

Hypertrophic cardiomyopathy (HCM) is a cardiovascular disease where the heart muscle is partially thickened and blood flow is - potentially fatally - obstructed. It is one of the leading causes of sudden cardiac death in young people. Electrocardiography (ECG) and Echocardiography (Echo) are the standard tests for identifying HCM and other cardiac abnormalities. The American Heart Association has recommended using a pre-participation questionnaire for young athletes instead of ECG or Echo tests due to considerations of cost and time involved in interpreting the results of these tests by an expert cardiologist. Initially we set out to develop a classifier for automated prediction of young athletes’ heart conditions based on the answers to the questionnaire. Classification results and further in-depth analysis using computational and statistical methods indicated significant shortcomings of the questionnaire in predicting cardiac abnormalities. Automated methods for analyzing ECG signals can help reduce cost and save time in the pre-participation screening process by detecting HCM and other cardiac abnormalities. Therefore, the main goal of this dissertation work is to identify HCM through computational analysis of 12-lead ECG. ECG signals recorded on one or two leads have been analyzed in the past for classifying individual heartbeats into different types of arrhythmia as annotated primarily in the MIT-BIH database. In contrast, we classify complete sequences of 12-lead ECGs to assign patients into two groups: HCM vs. non-HCM. The challenges and issues we address include missing ECG waves in one or more leads and the dimensionality of a large feature-set. We address these by proposing imputation and feature-selection methods. We develop heartbeat-classifiers by employing Random Forests and Support Vector Machines, and propose a method to classify full 12-lead ECGs based on the proportion of heartbeats classified as HCM. The results from our experiments show that the classifiers developed using our methods perform well in identifying HCM. Thus the two contributions of this thesis are the utilization of computational and statistical methods for discovering shortcomings in a current screening procedure and the development of methods to identify HCM through computational analysis of 12-lead ECG signals.

Lightweight Internet Traffic Classification - A Subject Based Solution with Word Embeddings

Internet traffic classification is a relevant and mature research field, anyway of growing importance and with still open technical challenges, also due to the pervasive presence of Internet-connected devices into everyday life. We claim the need for innovative traffic classification solutions capable of being lightweight, of adopting a domain-based approach, of not only concentrating on application-level protocol categorization but also classifying Internet traffic by subject. To this purpose, this paper originally proposes a classification solution that leverages domain name information extracted from IPFIX summaries, DNS logs, and DHCP leases, with the possibility to be applied to any kind of traffic. Our proposed solution is based on an extension of Word2vec unsupervised learning techniques running on a specialized Apache Spark cluster. In particular, learning techniques are leveraged to generate word-embeddings from a mixed dataset composed by domain names and natural language corpuses in a lightweight way and with general applicability. The paper also reports lessons learnt from our implementation and deployment experience that demonstrates that our solution can process 5500 IPFIX summaries per second on an Apache Spark cluster with 1 slave instance in Amazon EC2 at a cost of $ 3860 year. Reported experimental results about Precision, Recall, F-Measure, Accuracy, and Cohen's Kappa show the feasibility and effectiveness of the proposal. The experiments prove that words contained in domain names do have a relation with the kind of traffic directed towards them, therefore using specifically trained word embeddings we are able to classify them in customizable categories. We also show that training word embeddings on larger natural language corpuses leads improvements in terms of precision up to 180%.

Graph-based Semi-Supervised Learning in Acoustic Modeling for Automatic Speech Recognition

Thesis (Ph.D.)--University of Washington, 2016-06

Learning Models of Gene Expression from Synthetic DNA Sequences

Thesis (Ph.D.)--University of Washington, 2016-06

System for foreign exchange trading using genetic algorithms and reinforcement learning

Foreign exchange trading has emerged recently as a significant activity in many countries. As with most forms of trading, the activity is influenced by many random parameters so that the creation of a system that effectively emulates the trading process will be very helpful. A major issue for traders in the deregulated Foreign Exchange Market is when to sell and when to buy a particular currency in order to maximize profit. This paper presents novel trading strategies based on the machine learning methods of genetic algorithms and reinforcement learning.

GANN: Genetic algorithm neural networks for the detection of conserved combinations of features in DNA

Background: The multitude of motif detection algorithms developed to date have largely focused on the detection of patterns in primary sequence. Since sequence-dependent DNA structure and flexibility may also play a role in protein-DNA interactions, the simultaneous exploration of sequence-and structure-based hypotheses about the composition of binding sites and the ordering of features in a regulatory region should be considered as well. The consideration of structural features requires the development of new detection tools that can deal with data types other than primary sequence. Results: GANN ( available at http://bioinformatics.org.au/gann) is a machine learning tool for the detection of conserved features in DNA. The software suite contains programs to extract different regions of genomic DNA from flat files and convert these sequences to indices that reflect sequence and structural composition or the presence of specific protein binding sites. The machine learning component allows the classification of different types of sequences based on subsamples of these indices, and can identify the best combinations of indices and machine learning architecture for sequence discrimination. Another key feature of GANN is the replicated splitting of data into training and test sets, and the implementation of negative controls. In validation experiments, GANN successfully merged important sequence and structural features to yield good predictive models for synthetic and real regulatory regions. Conclusion: GANN is a flexible tool that can search through large sets of sequence and structural feature combinations to identify those that best characterize a set of sequences.

The applicability of recurrent neural networks for biological sequence analysis

Selection of machine learning techniques requires a certain sensitivity to the requirements of the problem. In particular, the problem can be made more tractable by deliberately using algorithms that are biased toward solutions of the requisite kind. In this paper, we argue that recurrent neural networks have a natural bias toward a problem domain of which biological sequence analysis tasks are a subset. We use experiments with synthetic data to illustrate this bias. We then demonstrate that this bias can be exploitable using a data set of protein sequences containing several classes of subcellular localization targeting peptides. The results show that, compared with feed forward, recurrent neural networks will generally perform better on sequence analysis tasks. Furthermore, as the patterns within the sequence become more ambiguous, the choice of specific recurrent architecture becomes more critical.

An examination of learning processes during critical incident training: Implications for the development of adaptable trainees

Three experiments are reported that examined the process by which trainees learn decision-making skills during a critical incident training program. Formal theories of category learning were used to identify two processes that may be responsible for the acquisition of decision-making skills: rule learning and exemplar learning. Experiments I and 2 used the process dissociation procedure (L. L. Jacoby, 1998) to evaluate the contribution of these processes to performance. The results suggest that trainees used a mixture of rule and exemplar learning. Furthermore, these learning processes were influenced by different aspects of training structure and design. The goal of Experiment 3 was to develop training techniques that enable trainees to use a rule adaptively. Trainees were tested on cases that represented exceptions to the rule. Unexpectedly, the results suggest that providing general instruction regarding the kinds of conditions in which a decision rule does not apply caused them to fixate on the specific conditions mentioned and impaired their ability to identify other conditions in which the rule might not apply. The theoretical, methodological, and practical implications of the results are discussed.

An Incremental EM-based Learning Approach for On-Line Prediction of Hospital Resource Utilization

Objective: Inpatient length of stay (LOS) is an important measure of hospital activity, health care resource consumption, and patient acuity. This research work aims at developing an incremental expectation maximization (EM) based learning approach on mixture of experts (ME) system for on-line prediction of LOS. The use of a batchmode learning process in most existing artificial neural networks to predict LOS is unrealistic, as the data become available over time and their pattern change dynamically. In contrast, an on-line process is capable of providing an output whenever a new datum becomes available. This on-the-spot information is therefore more useful and practical for making decisions, especially when one deals with a tremendous amount of data. Methods and material: The proposed approach is illustrated using a real example of gastroenteritis LOS data. The data set was extracted from a retrospective cohort study on all infants born in 1995-1997 and their subsequent admissions for gastroenteritis. The total number of admissions in this data set was n = 692. Linked hospitalization records of the cohort were retrieved retrospectively to derive the outcome measure, patient demographics, and associated co-morbidities information. A comparative study of the incremental learning and the batch-mode learning algorithms is considered. The performances of the learning algorithms are compared based on the mean absolute difference (MAD) between the predictions and the actual LOS, and the proportion of predictions with MAD < 1 day (Prop(MAD < 1)). The significance of the comparison is assessed through a regression analysis. Results: The incremental learning algorithm provides better on-line prediction of LOS when the system has gained sufficient training from more examples (MAD = 1.77 days and Prop(MAD < 1) = 54.3%), compared to that using the batch-mode learning. The regression analysis indicates a significant decrease of MAD (p-value = 0.063) and a significant (p-value = 0.044) increase of Prop(MAD

Fitness assessment of document model

Document classification is a supervised machine learning process, where predefined category labels are assigned to documents based on the hypothesis derived from training set of labelled documents. Documents cannot be directly interpreted by a computer system unless they have been modelled as a collection of computable features. Rogati and Yang [M. Rogati and Y. Yang, Resource selection for domain-specific cross-lingual IR, in SIGIR 2004: Proceedings of the 27th annual international conference on Research and Development in Information Retrieval, ACM Press, Sheffied: United Kingdom, pp. 154-161.] pointed out that the effectiveness of document classification system may vary in different domains. This implies that the quality of document model contributes to the effectiveness of document classification. Conventionally, model evaluation is accomplished by comparing the effectiveness scores of classifiers on model candidates. However, this kind of evaluation methods may encounter either under-fitting or over-fitting problems, because the effectiveness scores are restricted by the learning capacities of classifiers. We propose a model fitness evaluation method to determine whether a model is sufficient to distinguish positive and negative instances while still competent to provide satisfactory effectiveness with a small feature subset. Our experiments demonstrated how the fitness of models are assessed. The results of our work contribute to the researches of feature selection, dimensionality reduction and document classification.