1 A Systematic Approach to Manage Information Quality for Supporting Software Package Selection Claudia Ayala, Xavier Franch Technical University of Catalunya UPC-Campus Nord (Omega), 08034 Barcelona, Spain {cayala,franch}@lsi.upc.edu Abstract. Software package selection plays a crucial role in the deployment of software systems. One of its main current problems is how to deal with the vast amount of unstructured, incomplete, evolvable and widespread information that highly increases the risks of taking a wrong decision. It could be said that nowadays there are not satisfactory solutions to that problem. In this paper, we propose an approach to systematically tackle these information quality problems by stating a reference model embracing quality indicators that facilitate the collection, storage, retrieval, analysis and reuse of information in a quality assurance environment. We illustrate it with a scenario of use that shows how this reference model may be used to support software package selection decision-making. 1. Introduction Nowadays, the amount of information available about software packages (SP) is vast and still growing. To select SP (i.e., the process of deciding the appropriate SP to be integrated in a component-based system [1]), decision-makers have to face not only the current diversity of SPs types available in the marketplace, but also the great deal of widespread, heterogeneous, and unstructured information describing each of them [2], [3]. The quality of this information largely determines the quality of the decisions made, and ultimately affects the quality of the whole software system and its devel- opment [3], [4]. Since SP selectors must rely on the information for their decision- making processes, ensuring Information Quality (IQ) is a critical success factor. Over time, librarians and other information professionals have developed a set of criteria to be used to evaluate IQ based on careful experts’ examination (e.g., author- ity, format, scope, etc.) [5]; however, these criteria are too general and do not provide much guidance to the particular problem of SP selection. Therefore, more specialized approaches are needed. Some recent approaches propose the use of automatic or semiautomatic search engines to identify SP, e.g., [6],[7],[8]; but, to the best of our knowledge, they have not reached a generalized consensus on their utility for the community and do not provide a systematic approach for managing IQ. Therefore, IQ is still a critical open issue from the SP selection perspective [2], [3], [4],[9]. The goal of our research is twofold. On the one hand, to develop a comprehensive framework that states those IQ aspects that are important to perform an informed SP selection, and how these aspects are feasibly gauged using a quality model schema that hierarchically describes the IQ aspects, and provides metrics to assess the value