Spatio-temporal Rule Mining: Issues and Techniques

Slides:



Advertisements
Similar presentations
Association Rule and Sequential Pattern Mining for Episode Extraction Jonathan Yip.
Advertisements

Office of SA to CNS GeoIntelligence Introduction Data Mining vs Image Mining Image Mining - Issues and Challenges CBIR Image Mining Process Ontology.
Frequent Itemset Mining Methods. The Apriori algorithm Finding frequent itemsets using candidate generation Seminal algorithm proposed by R. Agrawal and.
A distributed method for mining association rules
3/3/20081 Data Warehousing and Data Mining. 3/3/20082 Why Data Mining? — Potential Applications Database analysis and decision support –Market analysis.
Spatial Database Systems. Spatial Database Applications GIS applications (maps): Urban planning, route optimization, fire or pollution monitoring, utility.
COMP5318 Knowledge Discovery and Data Mining
Association Rules Spring Data Mining: What is it?  Two definitions:  The first one, classic and well-known, says that data mining is the nontrivial.
LOGO Association Rule Lecturer: Dr. Bo Yuan
Mining Long Sharable Patterns in Trajectories of Moving Objects Győző Gidofalvi and Torben Bach Pedersen Arrrrgggg, all this spatio- temporal data from.
Data Mining Techniques Cluster Analysis Induction Neural Networks OLAP Data Visualization.
Weekly Report Start learning GPU Ph.D. Student: Leo Lee date: Sep. 18, 2009.
© Prentice Hall1 DATA MINING TECHNIQUES Introductory and Advanced Topics Eamonn Keogh (some slides adapted from) Margaret Dunham Dr. M.H.Dunham, Data Mining,
Dieter Pfoser, LBS Workshop1 Issues in the Management of Moving Point Objects Dieter Pfoser Nykredit Center for Database Research Aalborg University, Denmark.
1 A DATA MINING APPROACH FOR LOCATION PREDICTION IN MOBILE ENVIRONMENTS* by Gökhan Yavaş Feb 22, 2005 *: To appear in Data and Knowledge Engineering, Elsevier.
Application of Apriori Algorithm to Derive Association Rules Over Finance Data Set Presented By Kallepalli Vijay Instructor: Dr. Ruppa Thulasiram.
Mining Sequences. Examples of Sequence Web sequence:  {Homepage} {Electronics} {Digital Cameras} {Canon Digital Camera} {Shopping Cart} {Order Confirmation}
Tang: Introduction to Data Mining (with modification by Ch. Eick) I: Introduction to Data Mining A.Short Preview 1.Initial Definition of Data Mining 2.Motivation.
USpan: An Efficient Algorithm for Mining High Utility Sequential Patterns Authors: Junfu Yin, Zhigang Zheng, Longbing Cao In: Proceedings of the 18th ACM.
Data Mining Techniques Sequential Patterns. Sequential Pattern Mining Progress in bar-code technology has made it possible for retail organizations to.
Privacy Preserving Data Mining on Moving Object Trajectories Győző Gidófalvi Geomatic ApS Center for Geoinformatik Xuegang Harry Huang Torben Bach Pedersen.
Data Mining CS157B Fall 04 Professor Lee By Yanhua Xue.
Efficient Data Mining for Calling Path Patterns in GSM Networks Information Systems, accepted 5 December 2002 SPEAKER: YAO-TE WANG ( 王耀德 )
Mining Multidimensional Sequential Patterns over Data Streams Chedy Raїssi and Marc Plantevit DaWak_2008.
Sequential Pattern Mining
Expert Systems with Applications 34 (2008) 459–468 Multi-level fuzzy mining with multiple minimum supports Yeong-Chyi Lee, Tzung-Pei Hong, Tien-Chin Wang.
3-1 Data Mining Kelby Lee. 3-2 Overview ¨ Transaction Database ¨ What is Data Mining ¨ Data Mining Primitives ¨ Data Mining Objectives ¨ Predictive Modeling.
Temporal Database Paper Reading R 資工碩一 馬智釗 Efficient Mining Strategy for Frequent Serial Episodes in Temporal Database, K Huang, C Chang.
1 Knowledge Discovery from Transportation Network Data Paper Review Jiang, W., Vaidya, J., Balaporia, Z., Clifton, C., and Banich, B. Knowledge Discovery.
Academic Year 2014 Spring Academic Year 2014 Spring.
Mining Sequential Patterns © Tan,Steinbach, Kumar Introduction to Data Mining 4/18/2004 Slides are adapted from Introduction to Data Mining by Tan, Steinbach,
Data Mining  Association Rule  Classification  Clustering.
Chap 6: Association Rules. Rule Rules!  Motivation ~ recent progress in data mining + warehousing have made it possible to collect HUGE amount of data.
Optimization of Association Rules Extraction Through Exploitation of Context Dependent Constraints Arianna Gallo, Roberto Esposito, Rosa Meo, Marco Botta.
Chapter 3 Data Mining: Classification & Association Chapter 4 in the text box Section: 4.3 (4.3.1),
Differential Analysis on Deep Web Data Sources Tantan Liu, Fan Wang, Jiedan Zhu, Gagan Agrawal December.
Mining Sequential Patterns With Item Constraints
Data Mining – Intro.
What Is Cluster Analysis?
Sequential Pattern Mining
MIS2502: Data Analytics Advanced Analytics - Introduction
DATA MINING © Prentice Hall.
Data Mining Association Analysis: Basic Concepts and Algorithms
New ideas on FP-Growth and batch incremental mining with FP-Tree
Data Mining: Introduction
Sequential Pattern Mining Using A Bitmap Representation
Physical Database Design and Performance
Data Mining: Concepts and Techniques
A Research Oriented Study Report By :- Akash Saxena
Data Mining: Concepts and Techniques
Data Mining: Introduction
Techniques for Finding Patterns in Large Amounts of Data: Applications in Biology Vipin Kumar William Norris Professor and Head, Department of Computer.
Online Frequent Episode Mining
Datamining : Refers to extracting or mining knowledge from large amounts of data Applications : Market Analysis Fraud Detection Customer Retention Production.
Data Mining: Introduction
Market Basket Analysis and Association Rules
Data Mining: Introduction
Association Rule Mining
Data Mining Association Rules: Advanced Concepts and Algorithms
Transactional data Algorithm Applications
I don’t need a title slide for a lecture
Association Rule Mining
Mining Sequential Patterns
Market Basket Analysis and Association Rules
©Jiawei Han and Micheline Kamber
Data Mining: Introduction
Mining Path Traversal Patterns with User Interaction for Query Recommendation 龚赛赛
Data Mining for Finding Connections of Disease and Medical and Genomic Characteristics Vipin Kumar William Norris Professor and Head, Department of Computer.
Mining Long, Sharable Patterns in Trajectories of Moving Objects
15-826: Multimedia Databases and Data Mining
Presentation transcript:

Spatio-temporal Rule Mining: Issues and Techniques Győző Gidófalvi Geomatic ApS Center for Geoinformatik and Torben Bach Pedersen Aalborg University 13-11-2018 DaWaK 2005

Outline Why mine spatio-temporal data? Frequent pattern mining background Frequent itemset mining / association rules Sequential patterns Taxonomy of spatio-temporal data Some examples: STM, HUR, INFATI, DMI How to mine spatio-temporal rules? Pivoting -> spatio-temporal baskets Some examples: INFATI, HUR, DMI Mining long, common patterns in trajectories of moving objects Conclusions and future directions 13-11-2018 DaWaK 2005

Why Mine Spatio-temporal Data? Spatio-temporal data is being collected at enormous speeds (Tbyte/hour) Remote sensors on satellites Telescope scanning the skies Location data received from mobile devices Data needs to be analysed for various purposes Cataloging, classification, segmentation Scientific hypothesis formulation Study complex systems with autonomous mobile entities Aid the management, storage, and retrieval of spatio-temporal data Hidden information in data can be used to provide customized Location-Based Services (LBS) and Location-Based Advertising (LBA) 13-11-2018 DaWaK 2005

Frequent Pattern Mining First proposed by Agrawal and Sirkant for the analysis of customer purchase behaviour Frequent itemsets: Discover items are bought together by customers frequently? {bread, peanut butter, jelly} Association rules: Discover a possible causal relationship between the items in such a frequent itemset? {bread, peanut butter} -> {jelly} (within trans.) Sequential patterns: Discover sequences of items or itemsets that are frequent in sequences of transactions? {Star Wars Episode I} -> {Episode II} -> {Episode III} (between transactions) Episodes: Discover periodic patterns in a long sequence? Patterns in other structures: trees, graphs,… How do we extend these to the spatio-temporal domain? EX: {Strøget,noon,businessman} -> {cafe} 13-11-2018 DaWaK 2005

Frequent Pattern Mining Cont… Approaches to frequent itemset mining and association rules: Apriori: bottom-up, generate-and-test frequent itemsets BFS traversal of search space Pruning using support monotonicity of itemsets Projection-based (FP-growth): generate frequent itemset prefixes and extend the prefix by mining the prefix- projected database DFS traversal of search space Many other variants employing sophisticates in-memory data structures and representations of the data. Restrictions on frequent itemsets Closed frequent itemsets Maximal frequent itemsets 13-11-2018 DaWaK 2005

Taxonomy of Spatio-temporal Data Examples of spatio-temporal data: Space Time Man (STM): activities performed by mobile users at particular times and locations HUR1: number of passengers getting on/off busses at particular times and locations HUR2: Personal chip cards recording travels of individuals DMI: periodic atmospheric measurements like temperature, humidity, and pressure for 5 km grid cells INFATI: day–to–day movements of 20 private cars on the road network of Aalborg Criteria for categorization of spatio-temporal data: Are the measured entities mobile or immobile? Are the attribute values of the measured entities static or dynamic? 13-11-2018 DaWaK 2005

How to Mine Spatio-temporal Rules? Knowledge extractable by association rules is about dependencies between items within baskets. -> Need to construct spatio-temporal baskets. Pivoting is the process of grouping a set of records based on one or more attributes (pivoting attributes) and assigning the values of an another attribute (pivoted attribute) to groups or baskets. Spatio-temporal rules that can be mined from spatio-temporal baskets can be either implicit or explicit. 13-11-2018 DaWaK 2005

Illustration of Pivoting INFATI pivoting example: pivoting attributes are “Location” and “Time”, pivoted attribute is “CarID” 13-11-2018 DaWaK 2005

Spatio-temporally Restricted vs. Unrestricted 13-11-2018 DaWaK 2005

Explicit Spatio-temporal Rule Mining 1 13-11-2018 DaWaK 2005

Explicit Spatio-temporal Rule Mining 2 13-11-2018 DaWaK 2005

DMI: Dynamic Attributes of Immobile Entities 13-11-2018 DaWaK 2005

Mining Long, Common Patterns (LCP) in Trajectories of Moving Objects Trajectories of moving objects contain regularities or patterns These patterns can be used in indexing, tracking, and LBS LBS example: intelligent rideshare application Find common routes for a set of commuters and suggest rideshare possibilities to them Unique requirements: Patterns should rather be long than frequent Patterns should be shareable, i.e.: common Unique challenges: Patterns are extremely long Interesting patterns have relatively low support Not all sub-patterns are interesting 13-11-2018 DaWaK 2005

Method to Mine LCP in Trajectories Pre-processing: Identify trips, i.e.: gaps Map date-time domain to time-of-day domain Substitute noisy GPS measurements with spatio-temporal regions Use / exploit unique requirements: Prune search space if extractable patterns are doomed to be short Define unique support measure: n-support – # of transactions satisfying an itemset if the number of distinct objects associated with those transactions >= n, 0 otherwise It can be shown that interesting patterns are closed frequent itemsets In current work, a projection-based FIM algorithm is being extended to meet and use and meet these requirements Illustrative example: 13-11-2018 DaWaK 2005

Pre-processed Example Trajectory Database 13-11-2018 DaWaK 2005

Extracted Long Common Patterns 13-11-2018 DaWaK 2005

Conclusions and Future Directions Today: Taxonomy of spatio-temporal data Pivoting to obtain spatio-temporal baskets Mining explicit and implicit spatio-temporal rules Spatio-temporally restricted vs. unrestricted mining Mining long, common patterns in trajectories Tomorrow: Incorporate spatio-temporal indexes in spatio-temporal rule mining or vice versa Incorporate various spatio-temporal space partitioning methods into mining 13-11-2018 DaWaK 2005

Thank you for your attention! Questions? 13-11-2018 DaWaK 2005