Estudos comportamentais têm sido conduzidos, há séculos, por cientistas e filósofos, abordando assuntos tais como trajetórias de estrelas e planetas, organizações da sociedade, evolução dos seres vivos, comportamento e linguagem humana. Com o advento da computação, grandes quantidades de informação tornaram-se disponíveis, as quais geram novos desafios a fim de explorar e compreender variações comportamentais de interação com esses sistemas. Motivado por esses desafios e pela disponibilidade de informações, esta dissertação de mestrado propõe uma metodologia com objetivo de classificar, detectar e identificar padrões de comportamento. A fim de validar essa metodologia, modelou-se conhecimentos embutidos em informações relativas a interações de usuários durante a grafia digital de assinaturas (tais informações foram obtidas de uma base de dados do campeonato SVC2004 -- First International Signature Verification Competition). Os modelos de conhecimento gerados foram, posteriormente, empregados em experimentos visando o reconhecimento de assinaturas. Resultados obtidos foram comparados a outras abordagens propostas na literatura / Throughout the centuries, behavioral studies have been conducted by scientists and philosophers, approaching subjects such as stars and planet trajectories, social organizations, living beings, human behavior and language. With the advent of computer science, large amounts of information have been made available, which brings out new challenges in the interactive behavior context. Such challenges have motivated this master thesis which proposes a methodology to classify, detect and identify behavioral patterns. A digital signature verification database, obtained from the First International Signature Verification Competition (SVC2004), was used to validate the proposed methodology. Knowledge models were obtained and, afterwards, employed in signature verification experiments. Results were compared to other approaches from the literature

Estudos comportamentais têm sido conduzidos, há séculos, por cientistas e filósofos, abordando assuntos tais como trajetórias de estrelas e planetas, organizações da sociedade, evolução dos seres vivos, comportamento e linguagem humana. Com o advento da computação, grandes quantidades de informação tornaram-se disponíveis, as quais geram novos desafios a fim de explorar e compreender variações comportamentais de interação com esses sistemas. Motivado por esses desafios e pela disponibilidade de informações, esta dissertação de mestrado propõe uma metodologia com objetivo de classificar, detectar e identificar padrões de comportamento. A fim de validar essa metodologia, modelou-se conhecimentos embutidos em informações relativas a interações de usuários durante a grafia digital de assinaturas (tais informações foram obtidas de uma base de dados do campeonato SVC2004 -- First International Signature Verification Competition). Os modelos de conhecimento gerados foram, posteriormente, empregados em experimentos visando o reconhecimento de assinaturas. Resultados obtidos foram comparados a outras abordagens propostas na literatura / Throughout the centuries, behavioral studies have been conducted by scientists and philosophers, approaching subjects such as stars and planet trajectories, social organizations, living beings, human behavior and language. With the advent of computer science, large amounts of information have been made available, which brings out new challenges in the interactive behavior context. Such challenges have motivated this master thesis which proposes a methodology to classify, detect and identify behavioral patterns. A digital signature verification database, obtained from the First International Signature Verification Competition (SVC2004), was used to validate the proposed methodology. Knowledge models were obtained and, afterwards, employed in signature verification experiments. Results were compared to other approaches from the literature

This study intended to validate if clustering could be used to simplify models generated with process mining. The intention was also to see if these clusters could suggest anything about user efficiency. To that end a new metric where devised, average mean duration deviation. This metric aimed to show if a trace was more or less efficient than a comparative trace. Since the intent was to find traces with similar characteristics the clustering was done with characteristic features instead of time efficiency features. The aim was to find a correlation between efficiency after the fact. A correlation with efficiency could not be found.

In the realm of autonomous vehicles, ensuring safety through advanced anomaly detection is crucial. This thesis integrates Reservoir Computing with temporal-aware data analysis to enhance driver behavior assessment and traffic flow prediction. Our approach combines Reservoir Computing with autoencoder-based feature extraction to analyze driving metrics from vehicle sensors, capturing complex temporal patterns efficiently. Additionally, we extend our analysis to forecast traffic flow dynamics within road networks using the same framework. We evaluate our model using the PEMS-BAY and METRA-LA datasets, encompassing diverse traffic scenarios, along with a GPS dataset of 10,000 taxis, providing real-world driving dynamics. Through a support vector machine (SVM) algorithm, we categorize drivers based on their performance, offering insights for tailored anomaly detection strategies. This research advances anomaly detection for autonomous vehicles, promoting safer driving experiences and the evolution of vehicle safety technologies. By integrating Reservoir Computing with temporal-aware data analysis, this thesis contributes to both driver behavior assessment and traffic flow prediction, addressing critical aspects of autonomous vehicle systems. / Master of Science / Our cities are constantly growing, and traffic congestion is a major challenge. This project explores how innovative technology can help us predict traffic patterns and develop smarter management strategies. Inspired by the rigorous safety systems being developed for self-driving cars, we'll delve into the world of machine learning. By combining advanced techniques for identifying unusual traffic patterns with tools that analyze data over time, we'll gain a deeper understanding of traffic flow and driver behavior. We'll utilize data collected by car sensors, such as speed and turning patterns, to not only predict traffic jams but also see how drivers react in different situations. However, our project has a broader scope than just traffic flow. We aim to leverage this framework to understand driver behavior in general, with a particular focus on its implications for self-driving vehicles. Through meticulous data analysis and sophisticated algorithms, we can categorize drivers based on their performance. This valuable information can be used to develop improved methods for detecting risky situations, ultimately leading to safer roads and smoother traffic flow for everyone. To ensure the effectiveness of our approach, we'll rigorously test it using real-world data from GPS data from taxi fleets and nationally recognized traffic datasets. By harnessing the power of machine learning and tools that can adapt to changing data patterns, this project has the potential to revolutionize traffic management in cities. This paves the way for a future with safer roads, less congestion, and a more positive experience for everyone who lives in and travels through our bustling urban centers.

Background. Adolescent antisocial behaviors are versatile in terms of their onset, severity, pervasiveness, continuity, and developmental outcomes. A substantial body of literature on developmental pathway of antisocial behaviors indicates that meaningful subtypes exist within these heterogeneous antisocial behaviors, rendering important implications to their etiology, causal mechanism and intervention. This study tests a taxonomy of antisocial behavior by examining whether different offending groups can be distinguished by their different group features including background risks and external correlates. First, two broad offending groups, i.e., the early-onset group and the adolescent-onset group were identified in a clinical sample of 118 adjudicated male adolescents based on age of onset of symptoms of Conduct Disorder. Further, two distinct subtypes, i.e. antisocial behavior associated with symptoms of Attention Deficit Hyperactivity Disorder (ADHD) and antisocial behavior associated with callous-unemotional traits ii (CD traits), a defining feature of psychopathy, were hypothesized to coexist within the broad early-onset offending group, based on two lines of recent studies indicating ADHD and CD traits as important correlates of antisocial behaviors. These two subgroups were identified within the sample in this current study. / Conclusion. Different offending groups could be discerned by their distinctive associated group risks and deficits, giving evidence to different developmental pathways to antisocial behaviors. Implications to understanding and intervention of antisocial behaviors were discussed. / Method. Data were collected from 118 adjudicated male adolescents from a centralized probation facility in Hong Kong and 63 non-delinquent male control subjects from mainstream secondary schools, all aged between 12 and 17. Group comparisons and multinominal logistic regression were performed to test whether these offending groups could be distinguished by different background risks and deficits including variables pertaining to cognitive processes, family, parenting, and deviant peers, etc. / Results. The early-onset offending group could be differentiated from the adolescent-onset offending group by their association with adolescent adjustment iii difficulties, more background risks, ADHD diagnosis, and callous unemotional traits. The two early-onset subgroups, early-onset ADHD and early-onset CU traits group, shared similarities of having severer delinquency and poorer adolescent adjustment, but demonstrated differences in terms of disinhibitory processes. / Law, Yuen Wah Sonya. / Adviser: Patrick Wing-leung Leung. / Source: Dissertation Abstracts International, Volume: 73-04, Section: B, page: . / Thesis (Ph.D.)--Chinese University of Hong Kong, 2011. / Includes bibliographical references (leaves 265-289). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. [Ann Arbor, MI] : ProQuest Information and Learning, [201-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstract also in Chinese; appendix in Chinese.