Where Next

• 1. Anna Monreale Fabio Pinelli Roberto TrasartiFosca Giannotti A. Monreale, F. Pinelli, R. Trasarti, F. Giannotti.WhereNext: a Location Predictor on Trajectory Pattern Mining . KDD 2009 Knowledge Discovery and Delivery Lab (ISTI-CNR&Univ. Pisa) www-kdd.isti.cnr.it

2.

• Wireless networks infrastructures are thenerves of our territory

• besides offering their services, they gather highly informativetracesabout the human mobile activities

• Miniaturization, wearability, pervasiveness will produce traces of increasing

• • positioning accuracy

• • semantic richness

3.

• From the analysis of the traces of our mobile phones it is possible to reconstruct our mobile behaviour, the way we collectively move

• This knowledge may help us improving decision-making in many mobility-related issues:

• • Planning traffic and public mobility systems in metropolitan areas;

• • Planning physical communication networks

• • Forecasting traffic-related phenomena

• • Organizing logistics systems

• • Prediction

4. 5.

• Predicting the next location of a trajectory can improve a large set of services such as:

• Navigational services.

• Trac management.

• Location-based advertising.

• Services Pre-fetching.

• Simulation.

? ? ? .4 .8 .35 6.

• How to realize this idea:

• Extract patterns fromall theavailable movementsin a certain area instead of on the individual history of an object;

• Using theseLocal movement patternsas predictive rules.

• Build a prediction tree as global model.

Trajectory dataset Local patterns Prediction Tree 7. Select the set of interesting trajectories Validation Evaluation Extract T-Patterns (A set of Local models) Merge T-Patterns (Global model) Use the Condensed model as predictor 8.

• The local pattern we use is theT-Pattern.It describes the common behavior of a group of users in space and time.

F. Giannotti, M. Nanni, F. Pinelli, and D. Pedreschi.Trajectory pattern mining . KDD 2007: 330-339. 9.

• Generatingall rulesfrom each T-pattern and using them to build a classifier is too expensive.

T-Pattern Rules 1 2 3 R 1 R 2 R 3 R 4 R 1 R 2 R 3 R 4 R 1 R 2 R 3 R 4 10.

• To avoid the rules generation the T-Pattern set is organized as a prefix tree.

• For Each nodev Ididenties the nodev

• Regiona spatial component of the T-Pattern

• Supportis the support of the T-pattern

• For Each edgej

• [a,b]correspond to the time interval nof the T-Pattern

11.

• Three steps:

• • Search for best match

• • Candidate generation

• • Make predictions

How to compute the Best Match? Best Match Prediction 12.

• The spatio-temporal distance computed between the segment of trajectory (bounded in time using the previous transition time) and the current node of the path.

Case a : The trajectory segment intersects the region of the node Case b : The enlarged trajectory segment intersects the region Case c : The enlarged trajectory segment doesnt intersect the region Wheretheth_tis the time tolerance window defined by the user. 13.

• The path score is the aggregation of all punctual scores along a path.

• TheBest Matchis the path having:

• • the maximum path score;

• • at least one admissible prediction.

10 min 15 min 8 min 10 min Punctual score: 1 Punctual Score: .58 Punctual Score: .8 11 min 16 min Path score .79 14.

• Averagegeneralizes distances between the trajectory and each node

• Sumis based on the concept of depth

• Maxis the optimistic one, the best punctual score is selected as path score

• Context-dependentaggregations can take into consideration other aspects of the problem.

15.

• The WhereNext algorithm can be tuned using its parameters: -th_t: time window tolerance

• -th_s : space window tolerance

• -th_score : minimum prediction score threshold

• -th_agg : the aggregation function used to compute the path score (Avg, Sum or Max)

16.

• It is very hard to understand which is the best set ofT-patterns we can use to build the our model:

• a big set ofT-patternsvery slow prediction.

• a small set of T-patternscoverage leaks

• For this reason we have defined a way to measure the prediction power of a T-Pattern set.

17.

• An evaluating function is defined to estimate thepredicting powerof a T-Pattern set.

• SpatialCoverage : the space coverage of the regions contained in the T-Patterns set;

• DatasetCoverage : measures how much the T-Pattern set represents the trajectories

• RegionSeparation : the precision of the regions in the T-Pattern set.

Model 1 Model 2 Testing the a priori evaluation 18. You are here 19.

• The results are evaluated using the following measures:

• Accuracy : rate of the correctly predicted locations (space and time) divided by the total number of trajectories to be predicted.

• Average Error : the average distance between the real trajectories in the predicted interval and the region predicted.

• Prediction rate : the number of trajectories which have a prediction divided by the total number of trajectories to be predicted.

Predicted Location Cut Original Predicted Location Cut Original Error 20.

• We used real life GPS dataset obtained from 17,000 vehicles in the urban area of the city of Milan.

Training set : 4000 trajectories between 7am and 10 am on WednesdayTest set : 500 trajectories between 7am and 10 am on Thursday. 21.

• Predictedvsth_score

Average Errorvsth_space 22.

• AccuracyvsAverage Error

Single UsersAccuracyandPrediction rate 23.

• A visual example of the application on Milan mobility data. The context is traffic management and we want to predict how the traffic will move in the city center.

• We have built a predictor on a good set ofT-patterns whichincludethe city gates of Milan.

Part of the GeoPKDD integrated platform.F. Giannotti, D. Pedreschi, and et al. Geopkdd:Geographic privacy-aware knowledge discovery and delivery(european project), 2008. 24.

• - Anew techniqueto predict the next locations of a trajectory based on previous movements of all the objects without considering any information about the users. - Thetime informationis used not only to order the events but is intrinsically equipped in the T-Patterns used to build the Prediction tree. - The user cantune the methodto obtain a good accuracy and prediction rate.

• - We are experimenting the methodin real worldapplications.

25. 26. Trajectories Dataset Regions of Interest T-PATTERNS 27. 28.

• The same exact spatial location (x,y) usually never occurs twice

• The same exact transition times usually do not occur twice

• Solution: allow approximation

• • a notion ofspatial neighborhood

• • a notion oftemporal tolerance

29.

• Two points match if one falls within aspatial neighborhood N()of the other

• Two transition times match if theirtemporal difference is

• Example:

30.

• Two points match if one falls within aspatial neighborhood N()of the other

• Two transition times match if theirtemporal difference is

• Example:

31.

• Two points match if one falls within aspatial neighborhood N()of the other

• Two transition times match if theirtemporal difference is

• Example:

32.

• T-pattern mining can be mapped to a density estimation problem over R 3n-1

• • 2 dimensions for each (x,y) in the pattern (2n)

• • 1 dimension for each transition (n-1)

• Density computed by

• • mapping each sub-sequence of n points of each input trajectory toR 3n-1

• • drawing an influence area for each point (composition ofN()and )

• Too computationally expensive, heuristics needed

• Our solution: a combination of sequential pattern mining and density-based clustering