Offshoring of services will shape the economic landscape for the coming decades, and present opportunities and challenges for individuals, organizations, and countries. Many countries are jostling in the global services marketplace to take advantage of the emerging opportunities. The success of an organization's offshoring initiative depends critically on the selection of a right destination; a wrong choice may result in a failure and impose significant costs. This paper presents a framework to assess the attractiveness of a target country, which identifies various motivating, inhibiting, and facilitating factors that may influence its attractiveness. The framework is used to analyze the future attractiveness of India, a topic that continues to be debated. The relative attractiveness of other potential competitors for offshoring of services is also discussed. Contrary to other analysts’ opinions, we conclude that India is likely to sustain its leading position for a long time in the services sector, and that it is likely to emerge as a global hub for the delivery of services, analogous to China's role in the manufacturing sector.

Mainstream machine learning approaches to predictive analytics consistently prove their ability to perform well using a variety of datasets, although the task of identifying an optimally-performing machine learning approach for any given dataset becomes much less intuitive. Methods such as ensemble and transformation modeling have been developed to improve upon individual base learners and datasets with large degrees of variance. Despite the increased generalizability and flexibility of ensemble approaches, the cost often involves sacrificing inference for predictive ability. This paper introduces an alternative approach to ensemble modeling, combining the predictive ability of an ensemble framework with localized model construction through the incorporation of cluster analysis as a pre-processing technique. The workflow not only outperforms independent base learners and comparative ensemble methods, but also preserves local inferential capability by manipulating cluster parameters and maintaining interpretable relative importance values and non-transformed coefficients for the overall consideration of variable importance. This paper demonstrates the ensemble technique on a dataset to estimate rates of health insurance coverage across the state of Missouri, where the cluster pre-processing assists in understanding both local and global variable importance and interactions when predicting high concentration areas of low health insurance coverage based on demographic, socioeconomic, and geospatial variables.

Increasingly, decision makers are incorporating large quantities of interrelated data in their decision making. Decision support systems need to provide visualization tools to help decision makers glean trends and patterns that will help them design and evaluate alternative actions. While visualization software that might be incorporated into decision support systems is available, the literature does not provide sufficient guidelines for selecting among possible visualizations or their attributes. This paper describes a case study of the development of a visualization component to represent regional relationship data. It addresses the specific information goals of the target organization, various constraints that needed to be satisfied, and how the goals were achieved via a suitable choice of visualization technology and visualization algorithms. The development process highlighted the need for specific visualizations to be driven by the specific problem characteristics as much as general rules of visualization. Lessons learned during the process and how these lessons may be generalized to address similar requirements is presented.

In a previous study, Mueller et al. (ISPRS Int J Geo-Inf 8(1):13, 2019), presented a machine learning ensemble algorithm using K-means clustering as a preprocessing technique to increase predictive modeling performance. As a follow-on research effort, this study seeks to test the previously introduced algorithm's stability and sensitivity, as well as present an innovative method for the extraction of localized and state-level variable importance information from the original dataset, using a nontraditional method known as synthetic population generation. Through iterative synthetic population generation with similar underlying statistical properties to the original dataset and exploration of the distribution of health insurance coverage across the state of Missouri, we identified variables that contributed to decisions for clustering, variables that contributed most significantly to modeling health insurance distribution status throughout the state, and variables that were most influential in optimizing model performance, having the greatest impact on change-in-meansquared-error (MSE) measurements. Results suggest that cluster-based preprocessing approaches for machine learning algorithms can result in significantly increased performance, and also demonstrate how synthetic populations can be used for performance measurement to identify and test the extent to which variable statistical properties within a dataset can vary without resulting in significant performance loss.

No abstract