Presented by: Mi Tian, Deepan Sanghavi, Dhaval Dholakia

Slides:



Advertisements
Similar presentations
15th CTI Workshop, July 26, Smart Itinerary Recommendation based on User-Generated GPS Trajectories Hyoseok Yoon 1, Y. Zheng 2, X. Xie 2 and W.
Advertisements

Xiaolei Li, Zhenhui Li, Jiawei Han, Jae-Gil Lee. 1. Motivation 2. Anomaly Definitions 3. Algorithm 4. Experiments 5. Conclusion.
Wall Building for RTS Games Patrick Schmid. Age of Empires.
ADAPTIVE FASTEST PATH COMPUTATION ON A ROAD NETWORK: A TRAFFIC MINING APPROACH Hector Gonzalez, Jiawei Han, Xiaolei Li, Margaret Myslinska, John Paul Sondag.
An Interactive-Voting Based Map Matching Algorithm
Evaluating “find a path” reachability queries P. Bouros 1, T. Dalamagas 2, S.Skiadopoulos 3, T. Sellis 1,2 1 National Technical University of Athens 2.
Mining Compressed Frequent- Pattern Sets Dong Xin, Jiawei Han, Xifeng Yan, Hong Cheng Department of Computer Science University of Illinois at Urbana-Champaign.
Diversity Maximization Under Matroid Constraints Date : 2013/11/06 Source : KDD’13 Authors : Zeinab Abbassi, Vahab S. Mirrokni, Mayur Thakur Advisor :
Quality Aware Privacy Protection for Location-based Services Zhen Xiao, Xiaofeng Meng Renmin University of China Jianliang Xu Hong Kong Baptist University.
Presented by: GROUP 7 Gayathri Gandhamuneni & Yumeng Wang.
Patch to the Future: Unsupervised Visual Prediction
Constructing Popular Routes from Uncertain Trajectories Authors of Paper: Ling-Yin Wei (National Chiao Tung University, Hsinchu) Yu Zheng (Microsoft Research.
Constructing Popular Routes from Uncertain Trajectories Ling-Yin Wei 1, Yu Zheng 2, Wen-Chih Peng 1 1 National Chiao Tung University, Taiwan 2 Microsoft.
Critical Analysis Presentation: T-Drive: Driving Directions based on Taxi Trajectories Authors of Paper: Jing Yuan, Yu Zheng, Chengyang Zhang, Weilei Xie,
1 Discrete Structures & Algorithms Graphs and Trees: II EECE 320.
Influence and Correlation in Social Networks Aris Anagnostopoulos Ravi Kumar Mohammad Mahdian.
1 Efficient Placement and Dispatch of Sensors in a Wireless Sensor Network Prof. Yu-Chee Tseng Department of Computer Science National Chiao-Tung University.
Trip Planning Queries F. Li, D. Cheng, M. Hadjieleftheriou, G. Kollios, S.-H. Teng Boston University.
Minimum Spanning Trees. Subgraph A graph G is a subgraph of graph H if –The vertices of G are a subset of the vertices of H, and –The edges of G are a.
Distance Indexing on Road Networks A summary Andrew Chiang CS 4440.
Reducing Uncertainty of Low-sampling-rate Trajectories Kai Zheng, Yu Zheng, Xing Xie, Xiaofang Zhou University of Queensland & Microsoft Research Asia.
IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS 2007 (TPDS 2007)
Message-Optimal Connected Dominating Sets in Mobile Ad Hoc Networks Paper By: Khaled M. Alzoubi, Peng-Jun Wan, Ophir Frieder Presenter: Ke Gao Instructor:
1 A Bayesian Method for Guessing the Extreme Values in a Data Set Mingxi Wu, Chris Jermaine University of Florida September 2007.
Energy-Aware Scheduling with Quality of Surveillance Guarantee in Wireless Sensor Networks Jaehoon Jeong, Sarah Sharafkandi and David H.C. Du Dept. of.
Boundary Recognition in Sensor Networks by Topology Methods Yue Wang, Jie Gao Dept. of Computer Science Stony Brook University Stony Brook, NY Joseph S.B.
Finding Top-k Shortest Path Distance Changes in an Evolutionary Network SSTD th August 2011 Manish Gupta UIUC Charu Aggarwal IBM Jiawei Han UIUC.
Maximizing the lifetime of WSN using VBS Yaxiong Zhao and Jie Wu Computer and Information Sciences Temple University.
1 Utilizing Shared Vehicle Trajectories for Data Forwarding in Vehicular Networks IEEE INFOCOM MINI-CONFERENCE Fulong Xu, Shuo Gu, Jaehoon Jeong, Yu Gu,
CS223 Advanced Data Structures and Algorithms 1 Maximum Flow Neil Tang 3/30/2010.
Resource Allocation in Network Virtualization Jie Wu Computer and Information Sciences Temple University.
Trajectory Data Mining Dr. Yu Zheng Lead Researcher, Microsoft Research Chair Professor at Shanghai Jiao Tong University Editor-in-Chief of ACM Trans.
Trajectory Data Mining
Author: Haoyu Song, Murali Kodialam, Fang Hao and T.V. Lakshman Publisher/Conf. : IEEE International Conference on Network Protocols (ICNP), 2009 Speaker:
Computer Science 1 Using Clustering Information for Sensor Network Localization Haowen Chan, Mark Luk, and Adrian Perrig Carnegie Mellon University
Mobile Sensor Deployment for a Dynamic Cluster-based Target Tracking Sensor Network Niaoning Shan and Jindong Tan Department of Electrical and Computter.
1 Finding Spread Blockers in Dynamic Networks (SNAKDD08)Habiba, Yintao Yu, Tanya Y., Berger-Wolf, Jared Saia Speaker: Hsu, Yu-wen Advisor: Dr. Koh, Jia-Ling.
U of Minnesota DIWANS'061 Energy-Aware Scheduling with Quality of Surveillance Guarantee in Wireless Sensor Networks Jaehoon Jeong, Sarah Sharafkandi and.
Efficient Placement and Dispatch of Sensors in a Wireless Sensor Network You-Chiun Wang, Chun-Chi Hu, and Yu-Chee Tseng IEEE Transactions on Mobile Computing.
Optimal Relay Placement for Indoor Sensor Networks Cuiyao Xue †, Yanmin Zhu †, Lei Ni †, Minglu Li †, Bo Li ‡ † Shanghai Jiao Tong University ‡ HK University.
::Network Optimization:: Minimum Spanning Trees and Clustering Taufik Djatna, Dr.Eng. 1.
1 Link Privacy in Social Networks Aleksandra Korolova, Rajeev Motwani, Shubha U. Nabar CIKM’08 Advisor: Dr. Koh, JiaLing Speaker: Li, HueiJyun Date: 2009/3/30.
Cost Effective Mobile and Static Road Side Unit Deployment for Vehicular Adhoc Networks Presenter: Yesenia Velasco (Senior in Computer Science) Department.
Ning Jin, Wei Wang ICDE 2011 LTS: Discriminative Subgraph Mining by Learning from Search History.
Network Analyst. Network A network is a system of linear features that has the appropriate attributes for the flow of objects. A network is typically.
Managing Massive Trajectories on the Cloud
Prof. Yu-Chee Tseng Department of Computer Science
Cohesive Subgraph Computation over Large Graphs
T-Share: A Large-Scale Dynamic Taxi Ridesharing Service
Lecture 11 Graph Algorithms
Feeling-based location privacy protection for LBS
Surviving Holes and Barriers in Geographic Data Reporting for
DS595/CS525 Team#2 - Mi Tian, Deepan Sanghavi, Dhaval Dholakia
Maximal Independent Set
Dynamic Coverage In Wireless Ed-Hoc Sensor Networks
Mining Spatio-Temporal Reachable Regions over Massive Trajectory Data
Mining the Most Influential k-Location Set from Massive Trajectories
3.1 Clustering Finding a good clustering of the points is a fundamental issue in computing a representative simplicial complex. Mapper does not place any.
Spanning Trees Longin Jan Latecki Temple University based on slides by
Effective Social Network Quarantine with Minimal Isolation Costs
Efficient Evaluation of k-NN Queries Using Spatial Mashups
Finding Fastest Paths on A Road Network with Speed Patterns
Coverage and Distinguishability in Traffic Flow Monitoring
On Discovery of Traveling Companions from Streaming Trajectories
Graph Theory Unit 2.
Diversified Top-k Subgraph Querying in a Large Graph
Spanning Trees Longin Jan Latecki Temple University based on slides by
Maximum Flow Neil Tang 4/8/2008
Donghui Zhang, Tian Xia Northeastern University
A Neural Network for Car-Passenger matching in Ride Hailing Services.
Presentation transcript:

Presented by: Mi Tian, Deepan Sanghavi, Dhaval Dholakia Planning Bike Paths based on Sharing-Bikes’ Trajectories Jie Bao, Tianfu He, Sijie Ruan, Yanhua Li, Yingcai Wu, Yu Zheng Presented by: Mi Tian, Deepan Sanghavi, Dhaval Dholakia

Outline Introduction Methodology Experiments Conclusion Pre-processing Greedy Network Expansion Experiments Effectiveness and Case Studies System Deployment Conclusion

Introduction Cycling – Green transportation mode Benefits Effective bike paths Constraints: Budget Limitation Construction Convenience Bike Path Utilization

Motivation Introduction Earlier data mining methods drawbacks Importance of bike paths Incorporate real world constraints using Mobike user trajectories Benefits of Mobike trajectories data: Realistic Travel Demands Rich Travel Information

Problem Definition Two components: Pre-processing and Path planning Introduction Problem Definition Two components: Pre-processing and Path planning Road network graph G = (V , E) V -> Intersections and E -> Road segments Goal: To discover a subset of edges E′ ⊆ E, that follows three criterion: construction budget constraint connectivity constraint maximal usage benefit

Maximal Usage Benefits Introduction Maximal Usage Benefits Goal: Maximize usage of deployed bike paths which includes: Facilitate as many users as possible Cover more continuous route segments Example: Bike travels e1 → e2 → e3 2 planned paths cover same distance Plan 2 is preferred due to longer continuous path Serving more users vs longer continuous trips conflicts Flexible Score function Set of Continuous Road Segments Beneficial Score Normalize Si Tuning Parameter

Problem Statement

Introduction System Overview Explain an overview of the system

Methodology

Pre-Processing Methodology Trajectory Parsing Trajectory Map-Matching Remove GPS outlier Trajectory Map-Matching Bike path to road segment Inverted Index Construction Given road segment, find corresponding trajectories quickly

Idea Methodology Two interesting spatial patterns from visualization Spatial hotspots Star like trajectories Idea of a greedy algorithm Start from hotspots Expand out Until reach the budget Questions What road segments are GOOD? What road segments do we START from? What road segments do we EXPAND to?

What are GOOD roads? Methodology Beneficial Score For a set of edges E’ .g indicates: The level of continuous coverage of bike trajectories Because longer continuous path is more valuable for bike riders in reality Normalized length Overlapped road segments Larger a means longer continuous length coverage is preferred (a = 1 means we don’t care about it)

Greedy Network Expansion Methodology Greedy Network Expansion Maintain a list of candidate segments Start from the initial segments Add their neighbor segments to the list Pick one with the highest beneficial score per cost Add its neighbor segments, and update budget Repeat pick… until reach budget

Where do we START? Methodology Top-K Initialization Analysis Pick K segments with the highest beneficial score per cost Analysis High rank segments tend to be close to each other End up covering less areas (especially when budget B is large)

Where do we START? (cont.) Methodology Where do we START? (cont.) Spatial Clustering Initialization Pick K clusters from the highest ranked segments Analysis Covers more areas when budget is large May not expand important areas enough when budget is small

The Algorithm Methodology Constraints, parameters, and Pre-processing results Initialization (top-k or spatial clustering) Update budget as expand Terminate condition Choose one adjacent segment with maximum score/cost (from all associated trajectories) Expand the candidate list

Experiments

Dataset Statistics Experiments Road Network: Shanghai (Source- Bing Map) Intersections: 333,766 Road Segments: 440,922 Mobike Dataset: One month dataset (September 2016) 13,063 unique users 3791 Bikes 230,303 trajectories

Dataset Statistic Bike Trip Length Distribution Bike Trip Duration Distribution

Dataset Statistics Bike Trip Temporal Distribution Traversed Edges

Effectiveness Study Different k-values Different Total Budgets Experiments Effectiveness Study Different k-values Different Total Budgets

Effectiveness Study

Experiments Case Study

Experiments System Deployment

Conclusion Conclusion Data Driven approach to plan bike paths NP-hard problem Proposed greedy network expansion algorithm

Questions? Mi Tian (mtian AT wpi DOT edu) Deepan Sanghavi (dasanghavi AT wpi DOT edu) Dhaval Dholakia ( dddholakia AT wpi DOT edu)