Intragroup density predicting intergroup tie strength within open-source-software collaboration network

Abstract

There have been many studies in the literature discussing intra- and inter-cluster ties within sociological systems denoted by strong and weak ties, social and biological systems represented by community structure, and organizational systems signified by strong and loose couplings; where inter-cluster ties are mostly considered weak or loose. Loose couplings lead to dissemination or retrieval of information, spread of viruses, new product adoption, more and their prediction is significant for knowledge management, organizational innovation, epidemics of contagious diseases, and viral product design. Therefore in this paper, we investigate how to predict inter-cluster tie strength, and propose that inter-cluster tie strength can be predicted from determination of intra-cluster density. In the model design section, we provide the hypothesis and discuss logical and analytical reasoning; in the empirical section, we alternatively examine the relationship between intra-group density and inter-group tie strength via examining open-source-software (OSS) project collaboration data collected from SourceForge repository.