Probabilistic Mining of Socio-Geographic Routines From Mobile Phone Data

Abstract

There is relatively little work on the investigation of large-scale human data in terms of multimodality for human activity discovery. In this paper we suggest that human interaction data, or human proximity, obtained by mobile phone Bluetooth sensor data, can be integrated with human location data, obtained by mobile cell tower connections, to mine meaningful details about human activities from large and noisy datasets. We propose a model, called bag of multimodal behavior, that integrates the modeling of variations of location over multiple time-scales, and the modeling of interaction types from proximity. Our representation is simple yet robust to characterize real-life human behavior sensed from mobile phones, which are devices capable of capturing large-scale data known to be noisy and incomplete. We use an unsupervised approach, based on probabilistic topic models, to discover latent human activities in terms of the joint interaction and location behaviors of 97 individuals over the course of approximately a 10 month period using data from MIT's Reality Mining project. Some of the human activities discovered with our multimodal data representation include "going out from 7pm-midnight alone" and "working from 11am-5pm with 3-5 other people", further finding that this activity dominantly occurs on specific days of the week. Our methodology also finds dominant work patterns occurring on other days of the week. We further demonstrate the feasibility of the topic modeling framework to discover human routines to predict missing multimodal phone data on specific times of the day.

Survey on an Efficient Method for Finds Friends in Social Networks

“…Recommendation systems that try to suggest items (e.g., music, movie, and books) to users have become more and more popular in recent years. For instance, Amazon [2] recommends items to a user based on items the user previously visited, and items that other users are looking at. Netflix [4] and Rotten Tomatoes [5] recommend movies to a user based on the user's previous ratings and watching habits.…”

“…Recommendation systems constitute a large role in providing quality customized user experiences. The main challenge in developing relevant friend recommendations is due to the dynamic nature of humans' perception of friendship, which constitutes a cause for heterogeneity in social networks [1], [2]. It is usual and frequent for humans to change their view of friendship.…”

Friends Wall: A Semantic-based Friend Recommendation System for Social Networks

“…In this paper [4],there is relatively little work on the investigation of large-scale human data in terms of multimodality for human activity discovery. In this paper, we suggest that human interaction data, or human proximity, obtained by mobile phone Bluetooth sensor data, can be integrated with human location data, obtained by mobile cell tower connections, to mine meaningful details about human activities from large and noisy datasets.…”

Privacy-sensitive recognition of group conversational context with sociometers

“…Outdoor daily routines were discovered by Eagle and Pentland [12] using Principal Component Analysis (PCA) from mobile phone data on subjects' location, proximity, communication, and device usage behavior. Farrahi et al [13,14] extended the work using topic models, employing location and proximity data. Candia et al [3], used phone call data to study mean collective behavior of humans at large-scales.…”