Building an IT Taxonomy with Co-occurrence Analysis, Hierarchical Clustering, and Multidimensional Scaling

Different information technologies (ITs) are related in complex ways. How can the relationships among a large number of ITs be described and analyzed in a representative, dynamic, and scalable way? In this study, we employed co-occurrence analysis to explore the relationships among 50 information technologies discussed in six magazines over ten years (1998-2007). Using hierarchical clustering and multidimensional scaling, we have found that the similarities of the technologies can be depicted in hierarchies and two-dimensional plots, and that similar technologies can be classified into meaningful categories. The results imply reasonable validity of our approach for understanding technology relationships and building an IT taxonomy. The methodology that we offer not only helps IT practitioners and researchers make sense of numerous technologies in the iField but also bridges two related but thus far largely separate research streams in iSchools - information management and IT management