Decision Support in Social Media and Cloud Computing

This cumulative dissertation examines applications of decision support in the field of social media and cloud computing. By the advent of Social Media, Big Data Analytics and Cloud Computing, new opportunities opening up in the field of decision support due to availability and ability to process new types of data sets. In this context, this dissertation introduces systems for the use of social media data for decisions and an approach for decision support in choosing a cloud computing provider. In this dissertation, the benefits of different Facebook profile data for use in product recommender systems will be analyzed. Two experiments are carried out, in which the recommendation quality is determined by user survey. In another part of this dissertation, structured stock recommendations of an online community are used to automatically derive and update a stock portfolio. So investment decisions in the stock market are supported by a regular recalculation of the community rating for individual stocks. An succeeding article on this topic develops a formalized model for the description of investment strategies to enable a portfolio management system that automatically follows a strategy parameterized by an investor. Finally, a cloud broker model is presented which offers price / performance-based decision support in identifying an appropriate IaaS provider on the market for public cloud services. In a fundamental part of the thesis an IT architecture design is proposed which allows the parallel use and evaluation of different solution approaches in an operative IT system. Statistical tests are used to identify the best performing approach(es) and prefer them quickly while in operation. Overall, this cumulative dissertation consists of an introduction and five published articles.