With a large population of people who suffer from aphasia, it is imperative that an effective form of therapy is utilized. The purpose of this study was to investigate the effectiveness of augmentative and alternative communication (AAC) programs in improving the communication needs and lives of people affected by aphasia. Individuals (n = 20) suffering from aphasia for 3 months or more completed therapy sessions with speech and language pathologists and the AAC program. Pre- and post-intervention evaluations were administered, consisting of communication satisfaction and success questionnaires, the Western Aphasia Battery (WAB), and probing tests (prompt and response) with an EEG component. Only preliminary data analyses were completed, on three individuals, due to setbacks. All individuals improved on their post-WAB score and average scores on the questionnaires generally increased, but none were statistically significant. However, clinicians noted a clinical significance in improvements, which suggests that AACs are beneficial in aiding and improving people’s communicative functions and daily life. Resting state EEG data of one subject exhibited high mean power spectral densities (PSD) for delta and theta bands in the lobes before and after therapy, supporting previous literature. Mean PSDs of the left frontal lobe demonstrated a statistically significant decrease from pre- to post-, which in the case of the delta and theta bands may indicate possible recovery. More research is necessary to substantiate these conclusions and to explore the use of EEG in mapping brain lesions and tracking the brain’s rehabilitation, as well as the benefits of AACs.

Advances in social networks and multimedia technologies greatly facilitate the recording and sharing of video data on violent social and/or political events via In- ternet. These video data are a rich source of information in terms of identifying the individuals responsible for damaging public and private property through vio- lent behavior. Any abnormal, violent individual behavior could trigger a cascade of undesirable events, such as vandalism and damage to stores and public facilities. When such incidents occur, investigators usually need to analyze thousands of hours of videos recorded using handheld devices in order to identify suspects. The exhaus- tive manual investigation of these video data is highly time and resource-consuming. Automated detection techniques of abnormal events and actions based on computer vision would o↵er a more e cient solution to this problem. The first contribution described in this thesis consists of a novel method for fire detection in riot videos acquired with handheld cameras and smart-phones. This is a typical example of computer vision in the wild, where we have no control over the data acquisition process, and where the quality of the video data varies considerably. The proposed spatial model is based on the Mixtures of Gaussians model and exploits color adjacency in the visible spectrum of incandescence. The experimental results demonstrate that using this spatial model in concert with motion cues leads to highly accurate results for fire detection in noisy, complex scenes of rioting crowds. The second contribution consists in a method for detecting abnormal, violent actions that are performed by individual subjects and witnessed by passive crowds. The problem of abnormal individual behavior, such as a fight, witnessed by passive bystanders gathered into a crowd has not been studied before. We show that the presence of a passive, standing crowd is an important indicator that an abnormal action might occur. Thus, detecting the standing crowd improves the performance of detecting the abnormal action. The proposed method performs crowd detection first, followed by the detection of abnormal motion events. Our main theoretical contribution consists in linking crowd detection to abnormal, violent actions, as well as in defining novel sets of features that characterize static crowds and abnormal individual actions in both spatial and spatio-temporal domains. Experimental results are computed on a custom dataset, the Vancouver Riot Dataset, that we generated using amateur video footage acquired with handheld devices and uploaded on public social network sites. Our approach achieves good precision and recall values, which validates our system’s reliability of localizing the crowds and the abnormal actions. To summarize, this thesis focuses on the detection of two types of abnormal events occurring in violent street movements. The data are gathered by passive participants to these movements using handheld devices. Although our data sets are drawn from one single social movement (the Vancouver 2011 Stanley cup riot) we are confident that our approaches would generalize well and would be helpful to forensic activities performed in the context of other similar violent occasions. / Graduate

When offenders commit crime there is the potential that they may leave behind trace amounts of their DNA, even when there has been no apparent body fluid spill. During the examination of crime scenes, scene investigators try to identify areas that may be sampled to locate these traces. Specialist techniques are then required within the laboratory to enable such small amounts to be analysed to obtain a profile. These techniques are referred to as Low Template DNA analysis (LTDNA), of which Low Copy Number DNA (LCN DNA) is one instance. In 2008, following the Omagh Bombing trial, and comments made by Judge Weir, the UK Forensic Regulator commissioned a review of the science of LTDNA analysis. The subsequent report made specific mention of the fact that there was no available information on the success rate of the use of such DNA techniques and that there seemed to be confusion over what constituted a success. The report went on to state that there was no information on where such trace amounts of DNA were likely to be found, or what factors could influence the likelihood of obtaining a trace DNA profile (Caddy, 2008). This research considered the outcomes of LCN DNA analysis from 3,552 samples to try to establish where trace amounts of DNA could be found, whether some areas sampled were more successful in generating profiles than others, and the likelihood of the profiles obtained being of use to a criminal investigation. Analysis of results identified areas that were more successful in generating profiles of use to an investigation and highlighted significant differences in results across a variety of items from which samples were taken. DNA samples taken from items associated with communication such as mobile phones were much more likely to produce a profile useful to a criminal investigation than those taken from fixed surfaces within premises. The results obtained showed that obtaining a DNA profile did not necessarily correlate with the profile being of use to a criminal investigation. This was due to the fact that a large number of these profiles were anticipated eliminations from legitimate sources. Items that produced high numbers of profiles but were anticipated eliminations, and therefore of no value to an investigation, came from items associated with skin samples and clothing. The research went further to identify key factors that affected the profiling rates. Factors that had a positive influence on the ability to obtain a profile included: any area that had been in close proximity to saliva (direct contact was not required); samples that had been recovered from the inside of premises or vehicles and therefore protected from the elements; those that were dry; items that were of a porous nature; and those that had a rough texture. No differences were found between the actual surface materials (plastic, glass, wood, metal), as all showed a propensity to generate profiles. Other factors that were considered but proved to have no effect on the profiling rates included seasonal differences and whether the area targeted for sampling was clearly defined. Items that had had high contact with a victim, were recovered from outside or had been wet, all proved to be less useful to an nvestigation. A further finding of the research was that swabs that had been recovered and stored frozen appeared to deteriorate in their ability to profile. This was particularly notable if they were submitted later than 5 months after recovery. Items stored in dry conditions did not deteriorate in this way. Overall the research can be used to provide investigators with the knowledge of what areas of crime scenes are most likely to yield trace DNA material, the key factors that can affect the likelihood of obtaining a profile, and those areas that are more likely to produce profiles useful to criminal investigations.

As a proactive, information-based policing approach, problem-oriented policing emphasizes the use of crime analysis techniques in the analysis of the underlying causes of the problems that police deal with. In particular, analysis applications can be powerful tools for criminal investigation, such as crime reconstruction, profiling, IAFIS, VICAP, and CODIS. The SARA Model represents a problem-solving strategy of problemoriented policing. It aims to address the underlying causes of the problems and create substantial solutions. However, implementing problem-oriented policing requires a significant change in both the philosophy and structure of police agencies. Not only American policing but also the Turkish National Police should consider problem-oriented policing as an alternative approach for solving criminal activities.

Mémoire numérisé par la Direction des bibliothèques de l'Université de Montréal.

Issues in Automatic Video Biography Editing are similar to those in Video Scene Detection and Topic Detection and Tracking (TDT). The techniques of Video Scene Detection and TDT can be applied to interviews to reduce the time necessary to edit a video biography. The system has attacked the problems of extraction of video text, story segmentation, and correlation. This thesis project was divided into three parts: extraction, scene detection, and correlation. The project successfully detected scene breaks in series television episodes and displayed scenes that had similar content.

For my thesis, I recorded the progression of my show, The Constant Wife. I discuss the conception of my original scene design as well as the changes that occurred while working with the director and other designers. Throughout the thesis, I address the problems that arose during the pre-production, what I did to rectify those situations, and what I might have done differently if given the opportunity. I have divided the thesis into the following sections – Process of the Pre-Production, Execution of the Design and Evaluation of the Design. The final section of my thesis details what I learned from the events that transpired and how I can apply those lessons to future designs. Several appendixes are included that give various pieces of information that I used to develop my design as well as the majority of my research, including a larger version of the images in this paper, my drafting, and photos of the finished set. This document was created in Microsoft Word MAC.

This thesis is a narrative describing the set design for the Richmond premiere of Doubt: A Parable by John Patrick Shanley. The project was a creation of a set to be used by the Barksdale Theatre in Richmond, Virginia. The performance was held in February 15, 2008 through March 22, 2008 at the Barksdale Theatre in Richmond, Virginia. The production was a cooperation between the Virginia Commonwealth University Theatre Department and the Barksdale Theatre Inc.

This text is a partial record of the conception and development of Reflections of Red in a Mirror of Desire that opened on February 20th 2006 for a three day run ending on February 22nd. The majority of the text is focused on the events and research that provided the concept for the show. Some of the research includes influential choreographers and companies such as Pina Bausch, Graziela Daniele, Moses Pendleton, Pilobolus, Momix, Julie Tayrnor, and Merce Cunningham. Included in the analysis of this event are the experiences and lessons that came about during the process of directing Reflections of Red in a Mirror of desire. The result of this evaluation is the creation of an approach to choreographing and directing sexual or intimate scenes for the stage. Further reflections on directing, choreographing, collaboration, creative process, and aesthetics serve as the culmination of lessons inherent in both the creation of the production and the author's three years of study at the Virginia Commonwealth University Theatre Pedagogy Program with an emphasis in Movement and Choreography.

Psychological profiling of an offender Psychological profiling is a method of investigation developed in the 70's in the USA. This method is founded on research of behaviour of unknown offender with purpose to make conclusions about his personality. The purpose of my thesis is to find out how psychological profile can help during investigation. I would like to prove that psychological profiling belongs to investigation, that it can help during investigation and that it can bring a new view on case. The source of knowledge about the offender's behaviour and personality is the crime scene or text analyse. The thesis is composed of three chapters, each of them dealing with different aspects of psychological profiling. Chapter One is introductory and is subdivided into two parts. Part one describes history of psychological profiling. Part Two deals with theorethical background of psychological profiling. Chapter Two focuses on practical part of psychological profiling and consists of seven parts. This chapter explains the process of making a psychological profile, differences between mass-murder, serial-killer ad spree killer, characteristics of expert who creates a psychological profile, limits of psychological profile and differences between crimes committed by intimate people and those by unknown...