Collective Traffic Prediction with Partially Observed Traffic - PowerPoint PPT Presentation

Motivation Related Works CTP Method Experiments Summary Collective Traffic Prediction with Partially Observed Traffic History using Location-Based Social Media Xinyue Liu, Xiangnan Kong, Yanhua Li Worcester Polytechnic Institute February 22, 2017 1 / 34

Motivation Related Works CTP Method Experiments Summary About me 2 / 34

Motivation Related Works CTP Method Experiments Summary About me 3 / 34

Motivation Related Works CTP Method Experiments Summary About me ◦ I only know Python (2), and it is great. ◦ I think JavaScript, Ruby, Haskell... are cool, but I am too lazy to learn them. ◦ I hate C++. 4 / 34

Motivation Related Works CTP Method Experiments Summary My Research Interests ◦ Social Network Analysis [CIKM’16, SDM’17b] ◦ Recommender Systems [SDM’16] ◦ Brain Network [SDM’17a, IJCNN17] 5 / 34

Motivation Related Works CTP Method Experiments Summary Overview Motivation 1 Related Works 2 CTP Method 3 Experiments 4 Summary 5 6 / 34

Motivation Related Works CTP Method Experiments Summary Why Traffic Prediction? ◦ Excessive traffic causes travel delays, resource wasting, and pollution. ◦ In 2011, traffic congestion costs urban Americans 5.5 billion hours of travel delay, 2.9 billion gallons of extra fuel, for a total congestion cost of $121 billion . 7 / 34

Motivation Related Works CTP Method Experiments Summary Why (Location-Based) Social Media? Location Associations Temporal Data “Traffic jam on Storrow Drive, Boston, Massachusetts” 8 AM 4 PM 11 PM Semantic Data Traffic Condition Sensor Traffic Networks Location-Based Social Media ◦ Location-Based Social Media (LBSM) is popular, can be used as mobile sensors. ◦ Semantic and spatial information from social media can be helpful. 8 / 34

Motivation Related Works CTP Method Experiments Summary Challenges ◦ Lack of historical traffic data in partial regions. In real-world road systems, only a small fraction of the road segments are deployed with sensors. It is difficult to predict traffic without traffic history. ◦ Sparsity of LBSM information at fine granularity. Table: Average # of tweets in each region under different spatiotemporal resolutions Temporal Resolution Spatial Resolution Ave. #Tweets 12 hours 1 × 1 47,113 1 hour 1 × 1 3,926 1 hour 2 × 2 1,306 1 hour 3 × 3 554 1 hour 4 × 4 389 1 hour 30 × 30 15 9 / 34

Motivation Related Works CTP Method Experiments Summary Conventional Methods ◦ Auto Regression [Smith and Demetsky, 1997, Journal of Transportation Engineering] ◦ Tweet Semantics [He et al.,2013, IJCAI] 10 / 34

Motivation Related Works CTP Method Experiments Summary Auto Regression [Smith and Demetsky, 1997] Prediction spatio-temporal dependencies Historical Traffic Data t time ◦ v ( t ) = α + β 1 v ( t − 1) + β 2 v ( t − 2) g g g ◦ Fail to work for locations without traffic history. 11 / 34

Motivation Related Works CTP Method Experiments Summary Tweet Semantics [He et al.,2013] Social Media Prediction a a a b b c c c e e d d e Historical Traffic Data t time ◦ Consider each location independently. ◦ Extract tweet semantics as bag-of-words feature for each location during a 12-hour time window. ◦ Build an auto regression-like model using both traffic history and tweet semantics. ◦ Fail to work for locations without traffic history. 12 / 34

Motivation Related Works CTP Method Experiments Summary Illustration of CTP [Our Method] a b c d e road network time Local-based Social Media a b Prediction spatio-temporal c dependencies congestion e d regions without any sensor Historical Traffic Data t time ◦ Incorporate LBSM information at finer spatiotemporal granularity. ◦ Consider different locations collectively. ◦ It works for locations without traffic history! 13 / 34

Motivation Related Works CTP Method Experiments Summary Social Media Semantic Vectors 14 / 34

Motivation Related Works CTP Method Experiments Summary Spatio-temporal Dependencies: I t-1 v i ( t − 1) v q v j ( t − 1) ( t − 1) t v p ( t − 1) v i ( t ) v q v j ( t ) ( t ) v p ( t ) ◦ Same as the traffic history in auto regression model. 15 / 34

Motivation Related Works CTP Method Experiments Summary Spatio-temporal Dependencies: II t-1 v i ( t − 1) v q v j ( t − 1) ( t − 1) t v p ( t − 1) v i ( t ) v q v j ( t ) ( t ) v p ( t ) ◦ Spatial dependency within a time window. 16 / 34

Motivation Related Works CTP Method Experiments Summary Spatio-temporal Dependencies: III t-1 v i ( t − 1) v q v j ( t − 1) ( t − 1) t v p ( t − 1) v i ( t ) v q v j ( t ) ( t ) v p ( t ) ◦ Spatial dependency across time windows. 17 / 34

Motivation Related Works CTP Method Experiments Summary CTP Method Training t-2 t-1 ($) 𝑤 " ($) 𝑤 & ($) 𝑤 ' … Response LBSM Semantics ◦ assume time lag = 2 for the simplicity here. ◦ response variable (average speed, total traffic flow, etc). 18 / 34

Motivation Related Works CTP Method Experiments Summary CTP Method Training t-2 t-1 t-2 t-1 ($) 𝑤 " ($) 𝑤 & ($) 𝑤 ' … Response LBSM Semantics Dependency I (Traffic History) 19 / 34

Motivation Related Works CTP Method Experiments Summary CTP Method Training t-2 t-1 t-2 t-1 ($%&) 𝑤 " ($) ($%() 𝑤 " 𝑤 " ($%&) 𝑤 ) ($%() ($) 𝑤 ) 𝑤 ) ($%&) 𝑤 * ($%() ($) 𝑤 * 𝑤 * Retrieve the historical data Response LBSM Semantics Dependency I (Traffic History) 20 / 34

Motivation Related Works CTP Method Experiments Summary CTP Method Training t-2 t-1 t-2 t-1 t ($) 𝑤 " ($) 𝑤 & ($) 𝑤 ' … Response LBSM Semantics Dependency I (Traffic History) Dependency II (Neighbors’ Traffic) 21 / 34

Motivation Related Works CTP Method Experiments Summary CTP Method Training t-2 t-1 t-2 t-1 t t-2 t-1 ($) 𝑤 " ($) 𝑤 & ($) 𝑤 ' … Response LBSM Semantics Dependency I (Traffic History) Dependency II (Neighbors’ Traffic) Dependency III (Neighbors’ Traffic History) 22 / 34

Motivation Related Works CTP Method Experiments Summary CTP Method Training t-2 t-1 t-2 t-1 t t-2 t-1 ($) 𝑤 6 ($) 𝑤 " ($) 𝑤 * Compute using an aggregation function (e.g. average) Response • Response = Speed, aggregation function = AVG. ($) = 50, 𝑤 * ($) = 45 , 𝑤 , ($) and 𝑤 - ($) are unobserved. • 𝑤 " LBSM Semantics • The De Dependency-II II Fea eature for node A at time t is: (1) + / 2 (1) ) Dependency I (/ 0 • = 47.5 3 Dependency II Dependency III 23 / 34

Motivation Related Works CTP Method Experiments Summary CTP Method Training (only observed) t-2 t-1 t-2 t-1 t t-2 t-1 … Bootstrap t-1 t+1 t-1 t t-1 t t 0 … Response 0 0 0 (unobserved regions) LBSM Semantics … Dependency I Dependency II Dependency III 24 / 34

Motivation Related Works CTP Method Experiments Summary CTP Method Training (only observed) t-2 t-1 t-2 t-1 t t-2 t-1 … Bootstrap t-1 t+1 t-1 t t-1 t t 0 … Response 0 0 0 (unobserved regions) LBSM Semantics … Dependency I Dependency II Dependency III 25 / 34

Motivation Related Works CTP Method Experiments Summary CTP Method Training (only observed) t-2 t-1 t-2 t-1 t t-2 t-1 … Iterative Inference t-1 t+1 t-1 t t-1 t t … Response Keep updating 0 0 (unobserved regions) LBSM Semantics … Dependency I Keep updating Dependency II Dependency III 26 / 34

Motivation Related Works CTP Method Experiments Summary Dataset ◦ Traffic Data Collect from the California Performance Measurement System(PeMS) between October 19 and November 28, 2014. 31,102,272 entries of traffic records. ◦ LBSM Data Collect tweets from the same area during the same time range using the Twitter streaming API. This collection results in a total number of 2,648,446 tweets. 27 / 34

Motivation Related Works CTP Method Experiments Summary Compared Methods ◦ TDO [Smith and Demetsky, 1997]: Auto regression model using traffic history. ◦ TDO-floor [——–]: Similar to TDO , except it uses full traffic history. ◦ TwSeO : A degenerated version of [He et al. 2013], using tweets semantics. 28 / 34

Motivation Related Works CTP Method Experiments Summary Experimental Setting ◦ Partition the data into two parts, with the beginning (1 − 1 u ) as the training set and the remaining 1 u as the test set ( u = 3 , . . . , 7). ◦ k -fold cross-validation is used to randomly sample 1 / k regions as unobserved ( k = 2 , 3 , 4 , 5). ◦ Root Mean Square Error (RMSE) is used to evaluate the performance. 29 / 34

Motivation Related Works CTP Method Experiments Summary Results our method low ower Is Is be better ◦ TDO-floor performs the best by using full traffic history. ◦ The proposed CTP outperforms TDO and TwSeO. ◦ The result shows the effectiveness of incorporating tweets semantics into the collective inference model. 30 / 34

Motivation Related Works CTP Method Experiments Summary The effect of r low ower our method Is Is be better Sparser Information in LBSM Figure: Test Ratio = 1/7 ( u = 7) 31 / 34