No description available.

Nowadays, many real-world problems must deal with collections of high-dimensional data. High dimensional data usually have intrinsic low-dimensional representations, which are suited for subsequent analysis or processing. Therefore, finding low-dimensional representations is an essential step in many machine learning and data mining tasks. Low-rank and sparse modeling are emerging mathematical tools dealing with uncertainties of real-world data. Leveraging on the underlying structure of data, low-rank and sparse modeling approaches have achieved impressive performance in many data analysis tasks. Since the general rank minimization problem is computationally NP-hard, the convex relaxation of original problem is often solved. One popular heuristic method is to use the nuclear norm to approximate the rank of a matrix. Despite the success of nuclear norm minimization in capturing the low intrinsic-dimensionality of data, the nuclear norm minimizes not only the rank, but also the variance of matrix and may not be a good approximation to the rank function in practical problems. To mitigate above issue, this thesis proposes several nonconvex functions to approximate the rank function. However, It is often difficult to solve nonconvex problem. In this thesis, an optimization framework for nonconvex problem is further developed. The effectiveness of this approach is examined on several important applications, including matrix completion, robust principle component analysis, clustering, and recommender systems. Another issue associated with current clustering methods is that they work in two separate steps including similarity matrix computation and subsequent spectral clustering. The learned similarity matrix may not be optimal for subsequent clustering. Therefore, a unified algorithm framework is developed in this thesis. To capture the nonlinear relations among data points, we formulate this method in kernel space. Furthermore, the obtained continuous spectral solutions could severely deviate from the true discrete cluster labels, a discrete transformation is further incorporated in our model. Finally, our framework can simultaneously learn similarity matrix, kernel, and discrete cluster labels. The performance of the proposed algorithms is established through extensive experiments. This framework can be easily extended to semi-supervised classification.

Unemployment Insurance serves as a vital economic stabiliser, offering financial assistance and promoting workforce reintegration. In Sweden, occupation-specific unemployment funds, known as "Arbetslöshetskassan" (A-KASSAN), manage these claims. New complex challenges pertaining to A-KASSAN's decision-making process and unemployment insurance claims necessitate a holistic data analytics framework, innovative modelling approaches, and effective implementation strategies.  This study aims to establish a comprehensive approach to data analytics for unemployment insurance claims to provide a more accurate prediction model to aid A-KASSAN's decision-making. It accomplishes this through three main objectives: the development of a thorough framework employing management data analytics for claim analysis; advancement in modelling approaches to predict unemployment trends; and deliberation on effective strategies to visualise the developed solutions.  Drawing on Data Science, Computer Science, and Economics and Management Science, this study has crafted a four-tiered comprehensive framework encompassing descriptive, diagnostic, predictive, and prescriptive analytics. It has explored novel methodologies, formulated a model library, devised rules for result integration, and validated these through case studies. The model library showcases diverse models from Economic modelling, Statistical modelling, Big Data analytics with Machine Learning and Deep Learning, alongside hybrid modelling strategies. This study primarily concentrates on developing visualisation tools as an implementation strategy. In a summary, this study provides A-KASSAN with an approach to overcome two central issues: the lack of a comprehensive data analytics approach for unemployment insurance claims, including a framework and predictive modelling, and a dearth of visualisation solutions for management data analytics pertinent to these claims.

The purpose of this research is to understand how data analytics talent has been helping to improve firm performance. Given that firms have been hiring expensive and rare data talent aggressively to build their data analytics capabilities, the motivation of this research is to understand the kind impact this talent is having on firm performance. This research also examines the extent to which this talent has helped firms increase their revenue and improve their profitability and looks at this phenomenon by firm size and by industry membership. The research finds that the impact of data talent is higher in larger firms when compared to smaller firms. The research also finds that the impact of data talent on firm performance varies by industry membership, with some industries having a higher impact on improving revenue whereas some industries seeing impact on reducing costs. The research finally looks at explaining this difference in impact on revenue and cost with the help of the skills of the talent these firms are looking for. It finds that the type of skills requested by firms in data talent have a relationship with type of impact this talent can make on firm performance. There are important insights for Chief Data/Analytics Officers to align their organizations to contribute towards the performance of their firm, especially given the financial position of the firm. Some firms need to improve revenues whereas others have a need to cut costs. If a firm has a significant opportunity to reduce costs, but its data strategy is aligned towards improving revenue it becomes difficult to translate goals into action. The insight in this paper allows Chief Data/Analytics Officers to align their talent strategy towards the goals of the firm. / Business Administration/Strategic Management

Keeping track of requirements from eliciting data to making decision needs an effective path from data to decision [43]. Visualization science helps to create this path by extracting insights from flood of data. Model helps to shape the transformation of data to visualization. Defect Detection and Prevention model was created to assess quality assurance activities. We selected DDP and started enhancing user interactivity with requirements visualization over basic DDP with implementing a visual requirements analytics framework. By applying GQM table to our framework, we added six visualization features to the existing visual requirements visualization approaches. We applied this framework to technical and non-technical stakeholder scenarios to gain the operational insights of requirements-driven risk mitigation in practice. The combination of the first and second scenarios' result presented the multiple stakeholders scenario result which was a small number of strategies from kept tradespase with common mitigations that must deploy to the system.

Although past research finds that auditors support data analytics and artificial intelligence to enhance audit quality in their daily work, in reality, only a small number of audit firms, who innovated and invested in the two sophisticated technologies, utilize it in their auditing process. This paper analyzes three factors, including three individual theories, that may influence the adoption of data analytics and artificial intelligence in auditing: regulation (Institutional theory: explaining the catch-22 between the auditors and policymakers), knowledge barrier (Technology acceptance model's theory: explore the concept of ease of use), and people (algorithm aversion: a phenomenon that auditors believe in human decision makers more than technology). Among the three barriers, this paper focuses more on the people factor, which firms can start to overcome early. Past research has shown the existence of algorithm aversion in audit, so it is important to identify ways to decrease algorithm aversion. This study conducted a survey with four attributes: transparency-efficiency-trade-off, positive exposure, imperfect algorithm, and company's training. The study results shows that transparency-efficiency-trade-off can be a potential solution for decreasing algorithm aversion. When auditor firms implement transparency-efficiency-trade-off in their company training, auditors may give more trust to the technologies. The trust may lead to the increase of data analytics and artificial intelligence in audit.

Proporciona los pasos y procedimientos para acceder al recurso EndNote.

No description available.

No description available.

With the emergence of Big Data, data high in volume, variety, and velocity, new analysis techniques need to be developed to effectively use the data that is being collected. Knowledge discovery from databases is a larger methodology encompassing a process for gathering knowledge from that data. Analytics pair the knowledge with decision making to improve overall outcomes. Organizations have conclusive evidence that analytics provide competitive advantages and improve overall performance. This paper proposes a stand-alone methodology for data exploration. Data exploration is one part of the data mining process, used in knowledge discovery from databases and analytics. The goal of the methodology is to reduce the amount of time to gain meaningful information about a previously unanalyzed data set using tabular summaries and visualizations. The reduced time will enable faster implementation of analytics in an organization. Two case studies using a prototype implementation are presented showing the benefits of the methodology.