Understanding predictability and exploration in human mobility (original) (raw)

The influence of temporal and spatial features on the performance of next-place prediction algorithms

Proceedings of the 2013 ACM international joint conference on Pervasive and ubiquitous computing - UbiComp '13, 2013

Several algorithms to predict the next place visited by a user have been proposed in the literature. The accuracy of these algorithms -measured as the ratio of the number of correct predictions and the number of all computed predictions -is typically very high. In this paper, we show that this good performance is due to the high predictability intrinsic in human mobility. We also show that most algorithms fail to correctly predict transitions, i.e. situations in which users move between different places. To this end, we analyze the performance of 18 prediction algorithms focusing on their ability to predict transitions. We run our analysis on a data set of mobility traces of 37 users collected over a period of 1.5 years. Our results show that even algorithms achieving an overall high accuracy are unable to reliably predict the next location of the user if this is different from the current one. Building upon our analysis we then present a novel next-place prediction algorithm that can both achieve high overall accuracy and reliably predict transitions. Our approach combines all the 18 algorithms considered in our analysis and achieves its good performance at the cost of a higher computational and memory overhead.

Next Place Prediction using Mobile Data

Recently, location-based applications and services for mobile users have attracted significant attention. In this context, one challenging problem is predicting the future location of a mobile user given his or her current location and associated metadata. Solving this problem enables many interesting applications such as location-aware mobile advertisements, traffic warnings, etc. In this paper, we present an approach based on user-specific decision trees learned from each user's history. The classification tree is built ...

User location forecasting at points of interest

Proceedings of the 2012 RecSys workshop on Personalizing the local mobile experience - LocalPeMA '12, 2012

Predicting the location of a mobile user in the near future can be used for a very large number of user-centered or crowd-centered ubiquitous applications. It is convenient for the discussion to think in terms of discrete locations driven by Points of Interest (POI) instead of absolute positions. We postulate that POI sequences are Markovian once the data is clustered by day of the week and time of the day. To prove our hypothesis we used a public dataset, used in a previous work [15]. In that paper the authors were able to predict the location of a user with 90% to 70% accuracy in five minutes and one hour time windows, respectively. With our approach, using Hidden Markov Models, we are able to predict the next POIs within seven hours without significant accuracy decrease. This result enables a large number of potential applications where the aggregate data of a single users conform the behavior of the crowd.

Context Specific Next Location Prediction

2014

The rapid evolution and proliferation of mobile devices, mobile Internet access and social networks facilitate the development of intelligent services that build on the knowledge of the next location of a mobile user. The prediction of the next location of a mobile user, among others based on spatial, temporal, and in particular social influence factors, is the subject of this dissertation.

Predicting the location of mobile users

Proceedings of the 2009 international conference on Pervasive services - ICPS '09, 2009

Mobile context-aware applications experience a constantly changing environment with increased dynamicity. In order to work efficiently, the location of mobile users needs to be predicted and properly exploited by mobile applications. We propose a spatial context model, which deals with the location prediction of mobile users. Such model is used for the classification of the users' trajectories through Machine Learning (ML) algorithms. Predicting spatial context is treated through supervised learning. We evaluate our model in terms of prediction accuracy w.r.t. specific prediction parameters. The proposed model is also compared with other ML algorithms for location prediction. Our findings are very promising for the efficient operation of mobile context-aware applications.

MyWay: Location prediction via mobility profiling

Information Systems, 2017

Forecasting the future positions of mobile users is a valuable task allowing to operate efficiently a myriad of different applications which need this type of information. We propose MyWay, a prediction system which exploits the individual systematic behaviors modeled by mobility profiles to predict human movements. MyWay provides three strategies: the individual strategy uses only the user individual mobility profile, the collective strategy takes advantage of all users individual systematic behaviors, and the hierarchical strategy that is a combination of the previous two. A key point is that MyWay only requires the sharing of individual mobility profiles, a concise representation of the user's movements, instead of raw trajectory data revealing the detailed movement of the users. We evaluate the prediction performances of our proposal by a deep experimentation on large real-world data. The results highlight that the synergy between the individual and collective knowledge is the key for a better prediction and allow the system to outperform the state-of-art methods.

Where do we go? ONTHEWAY: A prediction system for spatial locations

2006

In ubiquitous computing we need to know the present context in order to interact properly with the nearby smart elements. When we are moving outdoors, mobile devices take a very important role because they provide us with a link between the world outside and ourselves through means of intelligent interfaces. There are a lot of situations in which it would be very useful to know or foresee the future context, i.e. as a geographic environment, in which we could find ourselves in a near future, and at the same time being able to use that information from our devices. Therefore we must preview this location with enough precision and time and be able to use this information from our mobile device. In our "OnTheWay" system, we used GPS technology and databases made of past paths taken by a person, in order to predict the next location, once we had begun a new course, comparing the new one with those ones stored. The results were amazing: from the data collected about paths travelled during a month and five days, we got the actual destination in 98% of cases, when we have only made a 30,35% of the total path. Therefore, including statistic and semantic information will allow us to upgrade our results, due to the sedentary human behaviour, the small number of frequently visited locations and the fact that the paths used to arrive to these locations are usually the same.

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

Proceedings of the Third ACM SIGSPATIAL International Workshop on Mobile Geographic Information Systems, 2014

Predicting the next location of people from their mobile phone logs has become an active research area. Due to two main reasons this problem is very challenging: the log data is very large and there are variety of granularity levels for specifying the spatial and the temporal attributes. In this work, we focus on predicting the next location change of the user and when this change occurs. Our method has two steps, namely clustering the spatial data into larger regions and grouping temporal data into time intervals to get higher granularity levels, and then, applying sequential pattern mining technique to extract frequent movement patterns to predict the change of the region of the user and its time frame. 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.

Predicting Future Locations and Arrival Times of Individuals

2011

Abstract This work has two objectives: a) to predict people's future locations, and b) to predict when they will be at given locations. Current location-based applications react to the user's current location. The progression from location-awareness to location-prediction can enable the next generation of proactive, context-predicting applications.

Far Out: Predicting Long-Term Human Mobility

Proceedings of the AAAI Conference on Artificial Intelligence

Much work has been done on predicting where is one going to be in the immediate future, typically within the next hour. By contrast, we address the open problem of predicting human mobility far into the future, a scale of months and years. We propose an efficient nonparametric method that extracts significant and robust patterns in location data, learns their associations with contextual features (such as day of week), and subsequently leverages this information to predict the most likely location at any given time in the future. The entire process is formulated in a principled way as an eigendecomposition problem. Evaluation on a massive dataset with more than 32,000 days worth of GPS data across 703 diverse subjects shows that our model predicts the correct location with high accuracy, even years into the future. This result opens a number of interesting avenues for future research and applications.