Proximity Measurement between Communication Objects

Many countries including the US and ED oblige telecommunication service providers to retain communication logs for a certain amount of time. The retained data are used for the purpose of the investigation, detection, or prosecution of serious crimes, but of huge size. In order to efficiently extract information such as a correlation between criminals and suspects, we must eliminate unnecessary data that occupy a large portion of communication logs. In this paper, we propose how to measure the proximity between communication objects using communication logs. The proximity let the collected data be analyzed efficiently: analyzing the data in the decreasing order of proximities or removing the data with low proximities before analyzing. The experimental results show that there is a correlation between proximities of our proposed measurement and estimation by people.