There has been a significant interest in the estimation of time and effort in fixing defects among both software practitioners and researchers over the past two decades. However, most of the focus has been on prediction of time and effort in resolving bugs, or other low level tasks, without much regard to predicting time needed to complete high-level requirements, a critical step in release planning. In this thesis, we describe a mixed-method empirical study on three large IBM projects in which we developed and evaluated a process of training a predictive model constituting a set of 29 features in nine categories in order to predict if whether or not a requirement will be completed within its planned iteration. We conducted feature engineering through iterative interviews with IBM software practitioners as well as analysis of large development and project management repositories of these three projects. Using machine learning techniques, we were able to make predictions on requirement completion time at four different stages of requirement lifetime. Using our industrial partner’s interest in high precision over recall, we then adopted a cost sensitive learning method and maximized precision of predictions (ranging from 0.8 to 0.97) while maintaining an acceptable recall. We also ranked the features based on their relative importance to the optimized predictive model. We show that although satisfying predictions can be made at early stages, even on the first day of requirement creation, performance of predictions improves over time by taking advantage of requirements’ progress data. Furthermore, feature importance ranking results show that although importance of features are highly dependent on project and prediction stage, there are certain features (e.g. requirement creator, time remained to the end of iteration, time since last requirement summary change and number of times requirement has been replanned for a new iteration) that emerge as important across most projects and stages, implying future worthwhile research directions for both researchers and practitioners. / Graduate

Magister Public Health - MPH / Background: TB is one of the most common opportunistic infections in the HIV infected population. In 2014, of the 9.6 million people estimated to have TB globally, 1.2 million were also infected with HIV. In the same year WHO reported 400 000 TB deaths in HIV infected people worldwide. TB Prevention strategies include ensuring HIV infected people take ART, TB infection control, treatment of TB cases and pharmacological prevention of primary TB infection or progression of latent TB into active TB. Isoniazid preventive therapy for a minimum of six months has been recommended to reduce the risk of TB in people living with HIV. Aim: The study’s aim was to determine the programmatic outcomes of isoniazid preventive therapy (IPT) and factors associated with treatment completion among people living with HIV aged 15 years and above at the AIDS Healthcare Foundation LaMvelase clinic in Manzini Swaziland, who were enrolled for IPT during the period March to December 2014. Methodology: This was a quantitative retrospective analytical cohort study that reviewed 3642 patient care records. IBM SPSS 20 was used for descriptive and statistical analysis of the data. Descriptive statistics were calculated and reported as frequencies and percentages. Bivariate statistics were carried out to test independent associations between socio-demographic and clinical characteristics, and IPT completion. Further multiple logistic regression analysis was done to determine the nature of association between the dependent and independent variables which had p < 0.1.

The study assesses the Lesotho government’s assistance programmes in addressing the financial implications of Education for All (EFA) at secondary school level. It does this by examining the objectives of the financial assistance programmes, their main problems and tensions in the implementation, and how they impact on the education of poor learners. This research was conducted using a qualitative methodology, which relied in part on the case study approach by focusing on two schools in the poor areas of Lesotho. The study used a multi-method data collection approach with interviews, questionnaires and document analysis. The respondents were selected on the basis of their relationship to the implementation of these government programmes. The school sample comprised two (2) principals, four (4) parents of learners targeted by such assistance programmes, and four (4) teachers working with learners who were beneficiaries of these programmes and who knew how these programmes assisted them (or not). There were two MOET officials, the bursary manager at central level, responsible for all government assistance programmes; and the district education bursary administrator responsible for the OVC programmes and entrusted with implementing the programmes in line with the regulations. The decision to get information from these respondents was because they were knowledgeable or experienced in different aspects of these government programmes, their implementation, and the impact on schools and learners. ii Using the rational and political analytical approaches, the findings reveal the limited conceptualization of the financial assistance programmes designed to widen the secondary education of poor learners as these did not address the deeper problems of Orphans and Vulnerable Children’s (OVC) problems in accessing and completing their education. The rational approach indicates poor system resources and capacity to effectively manage these programmes, as well as poor monitoring and accountability at all levels of the system. The political approach emphasises the poor implementation context of these assistance programmes due to the conflicting agendas of various programme implementers. The role of leadership is identified as crucial in interpreting and effectively mediating the operationalisation of these programmes on the ground so that the more deserving poor learners benefit. However, such leadership appears to be lacking at various levels of the implementation process. The study concludes that in order for these government assistance programmes to have a lasting and meaningful impact on OVC secondary schooling, there is a need to review their assumptions and scope to address more fully the OVC needs. Additional resource mobilization coupled with strong leadership, monitoring and evaluation are necessary for this to be realised.

Shadow removal is an important problem for both artists and algorithms. Previous methods handle some shadows well but, because they rely on the shadowed data, perform poorly in cases with severe degradation. Image-completion algorithms can completely replace severely degraded shadowed regions, and perform well with smaller-scale textures, but often fail to reproduce larger-scale macrostructure that may still be visible in the shadowed region. This paper provides a general framework that leverages degraded (e.g., shadowed) data to guide the image completion process by extending the objective function commonly used in current state-of-the-art image completion energy-minimization methods. This approach achieves realistic shadow removal even in cases of severe degradation and could be extended to other types of localized degradation.

In an attempt to demonstrate the interconnecting nature of the pillars of sport (Hacker, 2000) this thesis explores the relationship between physical, psychological and technical variables. Five National Collegiate Athletic Association (NCAA) Division I Women’s Soccer players were analyzed in four conference matches. The athlete's rate of perceived exertion (RPE) values were correlated to pass completion percentages (PC%) to investigate the potential effect that psychological satisfaction or dissatisfaction of technical performance has on perceived exertion ratings. PlayerLoadTM (PL) values, gathered through the utilization of a Catapult Minimax S4 GPS device (miniMax-10 Hz, Catapult Innovations, Melbourne, Australia), were used as a measure of actual workload in order to determine the accuracy of perceived exertion ratings. Results exhibited a trivial correlation (r = 0.028) and large variability (R2 = 0.00077) between RPE and PC% in comparison to PL and PC% (r = 0.50; R2 = 0.2502). The results indicate that an athlete's PC% does not correspond to how physically demanding they perceived a match to be. However, it is possible that performance variables may influence the RPE ratings a player provides.

Latent tuberculosis infection (LTBI), a product of exposure to Mycobacterium tuberculosis (Mtb), can lead to tuberculosis (TB) and further cause death if untreated. Fortunately, TB can be prevented with LTBI treatment. Targeting newly arrived visa holders for LTBI screening and treatment is an effective strategy for decreasing future TB burden. However, LTBI treatment completion rates are low, and researches had primarily focused on the nonrural U.S. setting. This study, using a retrospective cohort design under the epidemiological disease triangle framework evaluated (a) the treatment completion rates for 2 cohorts of visa holders (i.e., immigrants, N = 31 and refugees, N = 109) with LTBI residing in the rural setting using Pearson's chi-square analysis, (b) mean times on LTBI treatment using Kaplan-Meier survival analysis, and (c) predictors of time on treatment using Cox proportional hazard regression. Study findings revealed immigrants had higher treatment noncompletion rates over refugees (25.6% and 19.3%). The potential risk factors for noncompletion were being older than 24 years of age (HR = 0.18, p = 0.01). There were also significant interactions for the time on treatment between (a) being < 25 years old and visa type (HR = 0.23, p = 0.04), (b) being < 25 years and traveling longer (miles) to treatment facility (HR = 0.25, p = 0.03), or (c) being < 25 years and Mtb blood-test positive (HR = 0.35, p = 0.05). These findings suggest interventions targeting visa holders older than 24 years may increase the rate of treatment completion and decrease the future TB cases. Therefore, the study promotes social change by providing actionable, rural-population-specific information for the prioritization of visa holders at increased risk of experiencing LTBI treatment noncompletion.

In modern data and signal acquisition, one main challenge arises from the growing scale of data. The data acquisition devices, however, are often limited by physical and hardware constraints, precluding sampling with the desired rate and precision. It is thus of great interest to reduce the sensing complexity while retaining recovery resolution. And that is why we are interested in reconstructing a signal from a small number of randomly observed time domain samples. The main contributions of this thesis are as follows. First, we consider reconstructing a one-dimensional (1-D) spectrally sparse signal from a small number of randomly observed time-domain samples. The signal of interest is a linear combination of complex sinusoids at R distinct frequencies. The frequencies can assume any continuous values in the normalized frequency domain [0, 1). After converting the spectrally sparse signal into a low-rank Hankel structured matrix completion problem, we propose an efficient feasible point approach, named projected Wirtinger gradient descent (PWGD) algorithm, to efficiently solve this structured matrix completion problem. We give the convergence analysis of our proposed algorithms. We then apply this algorithm to a different formulation of structured matrix recovery: Hankel and Toeplitz mosaic structured matrix. The algorithms provide better recovery performance; and faster signal recovery than existing algorithms including atomic norm minimization (ANM) and Enhanced Matrix Completion (EMaC). We further accelerate our proposed algorithm by a scheme inspired by FISTA. Extensive numerical experiments are provided to illustrate the efficiency of our proposed algorithms. Different from earlier approaches, our algorithm can solve problems of very large dimensions very efficiently. Moreover, we extend our algorithms to signal recovery from noisy samples. Finally, we aim to reconstruct a two-dimension (2-D) spectrally sparse signal from a small size of randomly observed time-domain samples. We extend our algorithms to high-dimensional signal recovery from noisy samples and multivariate frequencies.

The ubiquitous roadblocks to university graduation have been investigated, identified, and interrogated for 7 decades, yet the mystery of retaining students to graduation continues to elude even the most prestigious universities. This researcher’s approach to increasing graduation began with the concession that increasingly, students may leave school at some point due to one or more of the retention issues that we recognize all too well—finances, illness, family problems, pregnancies, and other educational obstacles. However, leaving school does not mean that there is no going back. Student’s dropout status changes when they re-enroll in school; they take on new identities as stop-out students who forge their own nontraditional path to graduation. This work explored the lived experiences of this often-overlooked subset of university students—students who begin courses in higher education but then forgo their studies for a time before returning. These students are known in the literature as stop-out students, a cohort seldom acknowledged, studied, or desegregated from dropout statistics. An online survey was used to determine the demographics of the stop-out participants, and face-to-face, semi-structured interviews were then conducted to allow students to relate their experiences, in and out of school, in their own voices. Of particular interest was the effect of students’ perceived connections to faculty, staff, and/or administration as an influence in their decisions to return to school. The study was analyzed through the lens of care theory as a way to investigate how students’ persistence was affected by feelings of connection or caring. Only one of twelve interviewees had formed a relationship with a professor before he left school, and this relationship was maintained during his absence and renewed when he returned. The other interviewees acknowledged that they felt no specific connections to any person, office, or administration when they left. The stop-out population is one that higher education needs to acknowledge and support with targeted services. In many cases, they are only a few semesters from graduation. Rather than blocking their way when they run for the hills, we should be lighting their path back to success.

Senaste årens ökade internationalisering har gjort att antalet stora multinationella företag och koncerner ökat. Följden har blivit ett ökat behov av att kunna göra jämförbara tolkningar av finansiella rapporter för bolag i olika länder. Detta har lett till införandet av internationella redovisningsstandarder (IFRS/IAS), som gäller även för svenska koncerner från och med 2005. Då länderna historiskt haft olika värderingsprinciper finns risk för konflikt mellan det som varit traditionellt i Sverige, d.v.s. en mer försiktighetsinriktad syn på redovisningen, och matchnings-principen som är aktuell för dagens fokus på redovisning till verkligt värde. Problemfrågan är om de större verkstadsindustribolagen har en mer försiktighetsinriktad syn vid sin successiva vinst-avräkning. Stödjer metoden bolagen att visa en rättvisande bild av dess resultat och ställning? Syftet med uppsatsen har varit att söka svaret på vilken av försiktighets- eller matchnings-principen som väger tyngst för utvalda bolag när de gör successiv vinstavräkning. Motsvarar det kapitalmarknadens intresse med vinstavräkning till verkligt värde? För att söka svaren har strategin varit att göra en fallstudie där ett antal intervjuer gjorts med två bolag inom verkstadsindustribranschen. Då antalet direkta källor är begränsade bör inga generaliseringar göras av redovisat resultat. Uppsatsen baserar sig på redovisningsteorierna försiktighets- och matchningsprincipen samt den successiva vinstavräkningsmetoden. Försiktighetsprincipen innebär bl.a. att vinster inte ska tas upp förrän de har realiserats. Matchningsprincipen däremot innebär att intäkter och kostnader ska matchas och redovisas i samma period. Sistnämnda principen stödjer den successiva vinst-avräkningen som har till syfte att lyfta fram intäkter och kostnader i proportion till färdigställandet av varan/tjänsten. Sammantaget anser intervjuade bolag att den successiva vinstavräkningen hjälper dem att visa en rättvisande bild av bolagets resultat och ställning. Det framgår att koncernen, som det ena bolaget verkar i, historiskt haft en tradition att vara försiktiga. Analysen här är att försiktighetsprincipen fortfarande väger tyngst och att införandet av nya standarder antas först då det kulturella arvet mognat och hunnit ifatt. Men då omsättning endast till mindre del består av kundorders som avräknas successivt ger den totala externredovisning en rättvisande bild av bolaget. Slutsatser för det andra bolaget är att de tydligare använder matchningsprincipen vid värderingar för successiva vinstavräkningen. Tolkningen är att koncernen under en längre tid använt US-GAAP, vilka de internationella standarderna baseras på. Det medför att de redan värderar till verkligt värde och tillfredsställer därmed de privata kapitalmarknadsplacerarna.

This study utilized a combination of intrinsic and instrumental case studies to describe the experience of four dying persons and their significant relations, as they engaged in music therapy sessions designed with the goal of facilitating relationship completion. The four primary participants were inpatients of the Baycrest palliative care program who were diagnosed with a terminal illness and a life expectancy of less than six months. Two spouses who were involved in music therapy sessions were co-participants. I developed four case studies to represent each of the rich and detailed stories. Through the use of narrative research methods I was able to describe, interpret, and understand the complexity held within the multiple data sources that informed each case study. Data sources included: music created, utilized, recorded, and/or discussed in music therapy sessions; discussions during music therapy sessions; the researcher’s field notes; formal notes placed in the participant’s medical charts; the formal written assessment; transcriptions of audio-taped music therapy sessions; interviews; interview transcriptions; artistic pieces crafted by myself that emerged from the experiences of the participants as reflected in their interviews, and weekly participation in sessions which were verified by the participants; and other artistic material. The rich knowledge that emerged from the individual case studies informed a cross-case analysis where global themes were identified from a thematic analysis of participants’ experiences; and process motifs arising from the progression of participant engagement in music therapy are described. Global themes included: love; loss; gratitude; growth/transformation; courage/strength; and goodbye. The five process motifs that emerged were: (1) music therapy helps and was valued as a means of sharing the participants’ perceptions of their situation. (2) music therapy provides a safe place to become aware of, explore, and express feelings. (3) music enhances communication. (4) music therapy techniques provide creative avenues for self-expression. (5) music therapy provided a vehicle for revisiting and reminiscing. The thesis concludes with a summary of the knowledge revealed and a discussion of implications for music therapists and health care professionals, as well as a presentation of final thoughts and reflections on my role as researcher in this study.