The companies’ support phase, as all of business’ functional areas and components, went through a heavy and rapid digitalization which has unlocked the availability of an unprecedented amount of data. Unlike other relevant business areas and components, the support phase seems to have experienced fewer improvements attributable to Data Science and machine learning. By focusing on two well-known problems of these two fields, Time Series Analysis and Regression Analysis, this project aims at understanding which techniques are applicable within the support phase and how these can improve the effectiveness and pro-activeness of this area. The goal within this project is to apply them to improve the handling of support tickets, the digital entity used to track issues and requests within support systems. Through the use of Time Series Analysis, we aim at forecasting the volume of tickets to be expected in a near-future time frame. Using Regression Analysis we intend to estimate the resolution time of a newly submitted ticket. The results produced by the two tasks were satisfactory. On one hand, the Time Series task produced accurate results and the models could be directly employed and bring some added value to help Elvenite’s support team. On the other hand, while the Regression Analysis results were not as good, they nonetheless proved that the task’s aim is achievable through improvements on both the data used and the models applied. Finally, both tasks successfully showcased how to investigate and evaluate the application of such techniques within the support phase of a business. / Supportfasen, likväl samtliga andra delar av företags affärsfunktionella områden och komponenter, har genomgått en intensiv och snabb digitalisering som har öppnat upp tillgången till en enastående mängd data. Till skillnad från andra relevanta affärsområden och komponenter verkar supportfasen ha upplevt färre förbättringar som kan attribueras till Datavetenskap och maskininlärning. Projektet syftar till att förstå det ovanstående genom att fokusera på två välkända tekniker: tidsserieanalys och regressionsanalys. Det är följaktligen viktigt att undersöka vilka metoder från föregående nämnda områden som är användbara inom supportsystemen, samt hur dessa kan förbättra effektiviteten och proaktiviteten inom området. Det genomgående målet för projektet är att tillämpa analysmetoderna för att förbättra hanteringen av supportbiljetter. Supportbiljetter är den digitala enheten som används för att spåra frågor och förfrågningar inom supportsystem. Genom att använda tidsserieanalys eftersträvas att prognostisera volymen av biljetter som kan förväntas inom en snar framtid. Regressionsanalys användas för att tillhandahålla en uppskattad tid för en nyanländ biljett att bli löst, baserat på lösningstiden för tidigare lösta liknande biljetter. De två tillvägagångssätten gav olika, men tillfredställande resultat. Till att börja med anses tidsserieanalysen vara tillfredsställande och kan vara av värde samt hjälp för Elvenites supportteam. Dessvärre var resultaten från regressionsanalysen inte lika optimala och modellerna skulle behöva förbättras ytterligare före de appliceras i verkligheten. De båda teknikerna kunde ändock framgångsrikt bevisa och påvisa hur man kan undersöka samt utvärdera liknande metoder inom supportfasen av ett företag.

No description available.

No description available.

Implementing Big Data Analytics (BDA) has been widely known as a major source of competitiveness and innovation. While previous research suggests several process models and identifies critical factors for the successful implementation of BDA, there is a lack of understanding of how this organizational process is realized by its primary recipients, that is, Data Science Professionals (DSPs) whose innovation with BDA technologies stands at the core of big data-driven innovation. In particular, far less understood are the motivational and contextual factors that derive DSPs’ innovation with BDA technologies. This study proposes that commitment is the force that can attach DSPs to the BDA implementation process and motivate them to engage in innovative behaviors. It also introduces two organizational mechanisms, namely, BDA communication reciprocity and BDA leader theme-specific reputation, that can be employed to develop this constructive force in DSPs. Inspired by this, a theoretical model was developed based on the assertions of Commitment in Workplace Theory and the literature on creativity in organizations to assess the impact of DSPs’ commitment to BDA implementation and organizational context on their innovation with BDA technologies. This study theorizes that communication reciprocity and leader theme-specific reputation influence the three components of DSPs’ commitment (affective, continuance, and normative) to BDA implementation through their perceived participation in organizational decision-making and positive uncertainty, which, in turn, derive DSP’s innovation with BDA technologies. To further enrich the theorization, the moderating role of DSPs’ competency on the effect of DSPs’ components of commitment on their innovation with BDA technologies is investigated. Predictions were tested following an experimental vignette methodology with 240 subjects where the two organizational mechanisms were manipulated. Results indicate that organizational mechanisms provoke mediating psychological perceptions, though with varying strengths. In addition, results suggest that DSPs’ innovation with BDA technologies is primarily rooted in their affective and continuance commitments, and DSPs’ competency interacts with DSPs’ affective commitment to affect their innovation with BDA technologies. This research enhances the theoretical understanding of the role of commitment and organizational context in fostering DSPs’ innovation with BDA technologies. The results of this study also offer suggestions for information systems implementation practitioners on the effectiveness of organizational mechanisms that facilitate big data-driven innovation. / Thesis / Doctor of Philosophy (PhD)

Computing courses struggle to retain introductory students, especially as learner demographics have expanded to include more diverse majors, backgrounds, and career interests. Motivational contexts for these courses must extend beyond short-term interest to empower students and connect to learners' long-term goals, while maintaining a scaffolded experience. To solve ongoing problems such as student retention, methods should be explored that can engage and motivate students. I propose Data Science as an introductory context that can appeal to a wide range of learners. To test this hypothesis, my work uses two educational theories — the MUSIC Model of Academic Motivation and Situated Learning Theory — to evaluate different components of a student's learning experience for their contribution to the student's motivation. I analyze existing contexts that are used in introductory computing courses, such as game design and media computation, and their limitations in regard to educational theories. I also review how Data Science has been used as a context, and its associated affordances and barriers. Next, I describe two research projects that make it simple to integrate Data Science into introductory classes. The first project, RealTimeWeb, was a prototypical exploration of how real-time web APIs could be scaffolded into introductory projects and problems. RealTimeWeb evolved into the CORGIS Project, an extensible framework populated by a diverse collection of freely available "Pedagogical Datasets" designed specifically for novices. These datasets are available in easy-to-use libraries for multiple languages, various file formats, and also through accessible web-based tools. While developing these datasets, I identified and systematized a number of design issues, opportunities, and concepts involved in the preparation of Pedagogical Datasets. With the completed technology, I staged a number of interventions to evaluate Data Science as an introductory context and to better understand the relationship between student motivation and course outcomes. I present findings that show evidence for the potential of a Data Science context to motivate learners. While I found evidence that the course content naturally has a stronger influence on course outcomes, the course context is a valuable component of the course's learning experience. / Ph. D.

<p dir="ltr">This dissertation encompasses three studies in distinct yet impactful domains: B2B marketing, real-time video super-resolution (VSR), and smart office document routing systems. In the B2B marketing sphere, the study addresses the extended buying cycle by developing an algorithm for customer data aggregation and employing a CatBoost model to predict potential purchases with 91% accuracy. This approach enables the identification of high-potential<br>customers for targeted marketing campaigns, crucial for optimizing marketing efforts.<br>Transitioning to multimedia enhancement, the dissertation presents a lightweight recurrent network for real-time VSR. Developed for applications requiring high-quality video with low latency, such as video conferencing and media playback, this model integrates an optical flow estimation network for motion compensation and leverages a hidden space for the propagation of long-term information. The model demonstrates high efficiency in VSR. A<br>comparative analysis of motion estimation techniques underscores the importance of minimizing information loss.<br>The evolution towards smart office environments underscores the importance of an efficient document routing system, conceptualized as an online class-incremental image classification challenge. This research introduces a one-versus-rest parametric classifier, complemented by two updating algorithms based on passive-aggressiveness, and adaptive thresholding methods to manage low-confidence predictions. Tested on 710 labeled real document<br>images, the method reports a cumulative accuracy rate of approximately 97%, showcasing the effectiveness of the chosen aggressiveness parameter through various experiments.</p>

<p dir="ltr">Reinforcement learning (RL) has been a rapidly evolving field of research over the past years, enhancing developments in areas such as artificial intelligence, healthcare, and education, to name a few. Regardless of the success of RL, its inherent online learning nature presents obstacles for its real-world applications, since in many settings, online data collection with the latest learned policy can be expensive and/or dangerous (such as robotics, healthcare, and autonomous driving). This challenge has catalyzed research into offline RL, which involves reinforcement learning from previously collected static datasets, without the need for further online data collection. However, most existing offline RL methods depend on two key assumptions: unconfoundedness and positivity (also known as the full-coverage assumption), which frequently do not hold in the context of static datasets. </p><p dir="ltr">In the first part of this dissertation, we simultaneously address these two challenges by proposing a novel policy learning algorithm: PESsimistic CAusal Learning (PESCAL). We utilize the mediator variable based on Front-Door Criterion, to remove the confounding bias. Additionally, we adopt the pessimistic principle to tackle the distributional shift problem induced by the under-coverage issue. This issue refers to the mismatch of distributions between the action distributions induced by candidate policies, and the policy that generates the observational data (known as the behavior policy). Our key observation is that, by incorporating auxiliary variables that mediate the effect of actions on system dynamics, it is sufficient to learn a lower bound of the mediator distribution function, instead of the Q-function, to partially mitigate the issue of distributional shift. This insight significantly simplifies our algorithm, by circumventing the challenging task of sequential uncertainty quantification for the estimated Q-function. Moreover, we provide theoretical guarantees for the algorithms we propose, and demonstrate their efficacy through simulations, as well as real-world experiments utilizing offline datasets from a leading ride-hailing platform.</p><p dir="ltr">In the second part of this dissertation, in contrast to the first part, which approaches the distributional shift issue implicitly by penalizing the value function as a whole, we explicitly constrain the learned policy to not deviate significantly from the behavior policy, while still enabling flexible adjustment of the degree of constraints. Building upon the offline reinforcement learning algorithm, TD3+BC \cite{fujimoto2021minimalist}, we propose a model-free actor-critic algorithm with an adjustable behavior cloning (BC) term. We employ an ensemble of networks to quantify the uncertainty of the estimated value function, thus addressing the issue of overestimation. Moreover, we introduce a method that is both convenient and intuitively simple for controlling the degree of BC, through a Bernoulli random variable based on the user-specified confidence level for different offline datasets. Our proposed algorithm, named Ensemble-based Actor Critic with Adaptive Behavior Cloning (EABC), is straightforward to implement, exhibits low variance, and achieves strong performance across all D4RL benchmarks.</p>

<p dir="ltr">The success of machine learning techniques in widespread applications has taught us that with respect to accuracy, the more data, the better the model. However, for fairness, data quality is perhaps more important than quantity. Existing studies have considered the impact of data preprocessing on the accuracy of ML model tasks. However, the impact of preprocessing on the fairness of the downstream model has neither been studied nor well understood. Throughout this thesis, we conduct a systematic study of how data quality issues and data preprocessing steps impact model fairness. Our study evaluates several preprocessing techniques for several machine learning models trained over datasets with different characteristics and evaluated using several fairness metrics. It examines different data preparation techniques, such as changing categories into numbers, filling in missing information, and smoothing out unusual data points. The study measures fairness using standards that check if the model treats all groups equally, predicts outcomes fairly, and gives similar chances to everyone. By testing these methods on various types of data, the thesis identifies which combinations of techniques can make the models both accurate and fair.The empirical analysis demonstrated that preprocessing steps like one-hot encoding, imputation of missing values, and outlier treatment significantly influence fairness metrics. Specifically, models preprocessed with median imputation and robust scaling exhibited the most balanced performance across fairness and accuracy metrics, suggesting a potential best practice guideline for equitable ML model preparation. Thus, this work sheds light on the importance of data preparation in ML and emphasizes the need for careful handling of data to support fair and ethical use of ML in society.</p>

This research paper focuses on fraud detection in the financial industry using Generative Adversarial Networks (GANs) in conjunction with Uni and Multi Variate Bayesian Model with Shrinkage Priors (BMSP). The problem addressed is the need for accurate and advanced fraud detection techniques due to the increasing sophistication of fraudulent activities. The methodology involves the implementation of GANs and the application of BMSP for variable selection to generate synthetic fraud samples for fraud detection using the augmented dataset. Experimental results demonstrate the effectiveness of the BMSP GAN approach in detecting fraud with improved performance compared to other methods. The conclusions drawn highlight the potential of GANs and BMSP for enhancing fraud detection capabilities and suggest future research directions for further improvements in the field.

Stroke is a leading cause of long-term disability, affecting thousands of individuals annually and significantly impairing their mobility, independence, and quality of life. Traditional methods for assessing motor impairments are often costly and invasive, creating substantial barriers to effective rehabilitation. This thesis explores the use of DeepLabCut (DLC), a deep-learning-based pose estimation tool, to extract clinically meaningful kinematic features from video data of stroke survivors with upper-extremity (UE) impairments. To conduct this investigation, a specialized protocol was developed to tailor DLC for analyzing movements characteristic of UE impairments in stroke survivors. This protocol was validated through comparative analysis using peak acceleration (PA), mean squared jerk (MSJ), and area under the curve (AUC) as kinematic features. These features were extracted from the DLC output and compared to those derived from the assumed ground-truth data from IMU sensors worn by the participants. The accuracy of this analysis was quantified using percent mean squared error (PMSE) between each IMU sensor and DLC. PMSE analysis indicates that DLC-based kinematic features capture aspects of both accelerometer and gyroscope for the control participant. PA (8.78%) and AUC (3.28%) align more closely with the gyroscope, while MSJ (5.20%) demonstrates greater agreement with the accelerometer. On the other hand, for the stroke participant, DLC estimations for all kinematic features predominantly reflect data from the accelerometer. Across all datasets, AUC has the smallest PMSE values, suggesting that, based on our data, motor effort and energy expenditure in the tasks are best represented by DLC. Additionally, PMSE values for the stroke dataset are higher than those for the control, highlighting DLC's limitations in accurately detecting finer details of motion data in individuals with UE impairments. The results indicate that DLC reasonably estimates kinematic data for both participants, although further refinement of the methods is necessary to enhance the analysis of stroke data.