Predicting the change of location of mobile phone users

Abstract

Predicting the location of people from their mobile phone logs is becoming an attractive research area. Due to two main reasons this problem is very challenging: the log data is very large and there is a variety of granularity levels both for specifying the location and the time, especially with low granularity level it becomes much more complicated to define common user behaviour patterns. In this work, rather than determining the next location of a person, we focus on the predicting the location of a person when it changes. We employed a two phase method; which first clusters the data to obtain a higher granularity level, and then extracts frequent sequential patterns corresponding to location changes. We have validated our results with real data obtained from one of the largest mobile phone operators in Turkey. Our results are very encouraging, and we have obtained very high accuracy results in predicting the change of location of mobile phone users.

Predicting the next location change and time of change for mobile phone users

“…Predicting the time when change in location occurs is the basic novelty in the problem definition, which is not the case in our previous work [10]. Rather than predicting exact time of the action, it is aimed to find the time interval that action takes place in.…”

“…For example for 100 clusters, at the end of this process, the largest cluster contains 656 base stations and the smallest cluster contains only 6 base stations. Graphical representation of the clusters for this case can be found in our previous work [10].…”

“…In our recent work [10], we aimed to predict the user's next location for two cases. The first one is predicting the next action's (call, sms, gprs connection) location and time and the second one is predicting the next location of the user when he/she changes his/her current location by using only spatial data.…”

“…In this work, similar to the second problem of [10], we again focus on the prediction of the user's next location and time when he/she changes his/her current location. Rather than just using spatial information as in [10], by taking also temporal information into consideration, we get more precise prediction results.…”

“…Rather than just using spatial information as in [10], by taking also temporal information into consideration, we get more precise prediction results.…”