Quantifying Reciprocity in Social Networks

In this paper we propose a new reciprocity index for quantifying social relationships based on mobile phone call detail records and Twitter blogs. We use this reciprocity index to measure the level of reciprocity between users. This work is useful for detecting unwanted calls (e.g., spam) and product marketing. For validation of our results, we used actual call logs of 100 users collected at MIT by the Reality Mining Project group for a period of 8 months and Twitter blogs of 460 users collected by the Network Security team at UNT for a period of 12 months. The experimental results show that our model achieves results with high accuracy.