Mining for Spatial Patterns Shashi Shekhar Department of

Mining for Spatial Patterns Shashi Shekhar Department of Computer Science University of Minnesota http: //www. cs. umn. edu/~shekhar Collaborators: U. of Minnesota: V. Kumar, G. Karypis, C. T. Lu, W. Wu, Y. Huang, V. Raju, P. Zhang, P. Tan, M. Steinbach NASA Ames Research Center: C. Potter California State University, Monterey Bay: S. Klooster This work was partially funded by NASA and Army High Performance Computing Center Shashi Shekhar Mining For Spatial Patterns 1

Background NSF workshop on GIS and DM (3/99) Spatial data - traffic, bird habitats, global climate, logistics, . . . For spatial patterns - outliers, location prediction, associations, sequential associations, clustering, trends, … Shashi Shekhar Mining For Spatial Patterns 2

Framework Problem statement: capture special needs Data exploration: maps, new methods Try reusing classical methods from data mining, spatial statistics If reuse is not possible, invent new methods Validation, Performance tuning Shashi Shekhar Mining For Spatial Patterns 3

Research Goals: • modeling of ecological data Ø event modeling Ø zone modeling. • finding spatio-temporal patterns Ø associations Ø predictive models. A key interest is finding connections between the ocean and the land. Shashi Shekhar Mining For Spatial Patterns 4

Sources of Earth Science Data Before 1950, very sparse, unreliable data. Since 1950, reliable global data. Ocean temperature and pressure are based on data from ships. Most land data, (solar, precipitation, temperature and pressure) comes from weather stations. Since 1981, data has been available from Earth orbiting satellites. FPAR, a measure related to plant Since 1999 TERRA, the flagship of the NASA Earth Observing System, is providing much more detailed data. Shashi Shekhar Mining For Spatial Patterns 5

Example Pattern: Teleconnections are the simultaneous variation in climate and related processes over widely separated points on the Earth. For example, El Nino is the anomalous warming of the eastern tropical region of the Pacific, and has been linked to various climate phenomena. Droughts in Australia and Southern Africa Heavy rainfall along the western coast of South America Milder winters in the Midwest Shashi Shekhar Mining For Spatial Patterns 6

Net Primary Production (NPP) is the net assimilation of atmospheric carbon dioxide (CO 2) into organic matter by plants. NPP is driven by solar radiation and can be constrained by precipitation and temperature. NPP is a key variable for understanding the global carbon cycle and ecological dynamics of the Earth. Keeping track of NPP is important because it includes the food source of humans and all other organisms. Sudden changes in the NPP of a region can have a direct impact on the regional ecology. An ecosystem model for predicting NPP, CASA (the Carnegie Ames Stanford Approach) provides a detailed view of terrestrial productivity. Shashi Shekhar Mining For Spatial Patterns 7

Benefits of Data Mining Data mining provides earth scientist with tools that allow them to spend more time choosing and exploring interesting families of hypotheses. However, statistics is needed to provide methods for determining the “statistical” significance of results. By applying the proposed data mining techniques, some of the steps of hypothesis generation and evaluation will be automated, facilitated and improved. Association rules provide a “new” framework for detecting relationships between events. Shashi Shekhar Mining For Spatial Patterns 8

Approaches Shashi Shekhar Mining For Spatial Patterns 9

Clustering Interested in relationships between regions, not “points. ” For land, clustering based on NPP or other variables, e. g. , precipitation, temperature. For ocean, clustering based on SST (Sea Surface Temperature). When “raw” NPP and SST are used, clustering can find seasonal patterns. Anomalous regions have plant growth patterns which reversed from those typically observed in the hemisphere in which they reside, and are easy to spot. Shashi Shekhar Mining For Spatial Patterns 10

Clustering El Nino Regions SNN clusters of SST that are highly correlated with El Nino indices. Shashi Shekhar Mining For Spatial Patterns 11

Spatial Association Rule Citation: Symp. On Spatial Databases 2001 Problem: Given a set of boolean spatial features find subsets of co-located features, e. g. (fire, drought, vegetation) Data - continuous space, partition not natural, no reference feature Classical data mining approach: association rules But, Look Ma! No Transactions!!! No support measure! Approach: Work with continuous data without transactionizing it! confidence = Pr. [fire at s | drought in N(s) and vegetation in N(s)] support: cardinality of spatial join of instances of fire, drought, dry veg. participation: min. fraction of instances of a features in join result new algorithm using spatial joins and apriori_gen filters Shashi Shekhar Mining For Spatial Patterns 12

Event Definition Convert the time series into sequence of events at each spatial location. Shashi Shekhar Mining For Spatial Patterns 13

Interesting Association Patterns Use domain knowledge to eliminate uninteresting patterns. A pattern is less interesting if it occurs at random locations. Approach: Partition the land area into distinct groups (e. g. , based on landcover type). For each pattern, find the regions for which the pattern can be applied. If the pattern occurs mostly in a certain group of land areas, then it is potentially interesting. If the pattern occurs frequently in all groups of land areas, then it is less interesting. Shashi Shekhar Mining For Spatial Patterns 14

Association Rules Intra-zone non-sequential Patterns FPAR-Hi NPP-Hi (support 10) Shrubland regions • Region corresponds to semi-arid grasslands, a type of vegetation, which is able to quickly take advantage of high precipitation than forests. • Hypothesis: FPAR-Hi events could be related to unusual precipitation conditions. Shashi Shekhar Mining For Spatial Patterns 15

Co-location Can you find co-location patterns from the following sample dataset? Answers: Shashi Shekhar and Mining For Spatial Patterns 16

Co-location Can you find co-location patterns from the following sample dataset? Shashi Shekhar Mining For Spatial Patterns 17

Co-location Spatial Co-location A set of features frequently co-located Given A set T of K boolean spatial feature types T={f 1, f 2, … , fk} A set P of N locations P={p 1, …, p. N } in a spatial frame work S, pi P is of some spatial feature in T A neighbor relation R over locations in S Reference Feature Centric Find Tc = subsets of T frequently co-located Objective Correctness Completeness Efficiency Constraints R is symmetric and reflexive Monotonic prevalence measure Window Centric Shashi Shekhar Mining For Spatial Patterns Event Centric 18

Co-location Comparison with association rules Association rules Co-location rules underlying space discrete sets continuous space item-types events /Boolean spatial features collections transactions neighborhoods prevalence measure support participation index conditional probability measure Pr. [ A in T | B in T ] Pr. [ A in N(L) | B at L ] Participation index Participation ratio pr(fi, c) of feature fi in co-location c = {f 1, f 2, …, fk}: fraction of instances of fi with feature {f 1, …, fi-1, fi+1, …, fk} nearby 2. Participation index = min{pr(fi, c)} Algorithm Hybrid Co-location Miner Shashi Shekhar Mining For Spatial Patterns 19

Spatial Co-location Patterns Dataset • Spatial feature A, B, C and their instances • Possible associations are (A, B), (B, C), etc. • Neighbor relationship includes following pairs: • A 1, B 1 • A 2, B 2 • B 1, C 1 • B 2, C 2 Shashi Shekhar Mining For Spatial Patterns 20

Spatial Co-location Patterns Dataset Partition approach[Yasuhiko, KDD 2001] • Support not well defined, i. e. not independent of execution trace • Has a fast heuristic which is hard to analyze for correctness/completeness Spatial feature A, B, C, and their instances Support A, B =2 B, C=2 Shashi Shekhar Mining For Spatial Patterns Support A, B=1 B, C=2 21

Spatial Co-location Patterns Dataset Reference feature approach [Han SSD 95] • C as reference feature to get transactions • Transactions: (B 1) (B 2) • Support (A, B) = Ǿ from Apriori algorithm Spatial feature A, B, C, and their instances Shashi Shekhar • Note: Neighbor relationship includes following pairs: • A 1, B 1 • A 2, B 2 • B 1, C 1 • B 2, C 2 Mining For Spatial Patterns 22

Spatial Co-location Patterns Dataset Spatial feature A, B, C, and their instances Our approach (Event Centric) • Neighborhood instead of transactions • Spatial join on neighbor relationship • Support Prevalence • Participation index = min. p_ratio • P_ratio(A, B)) = fraction of instance of A participating in join(A, B, neighbor) • Examples Support(A, B)=min(2/2, 3/3)=1 Support(B, C)=min(2/2, 2/2)=1 Shashi Shekhar Mining For Spatial Patterns 23

Spatial Co-location Patterns Dataset Partition approach Our approach Support(A, B)=min(2/2, 3/3)=1 Support(B, C)=min(2/2, 2/2)=1 Spatial feature A, B, C, and their instances Support A, B =2 B, C=2 Reference feature approach C as reference feature Transactions: (B 1) (B 2) Support (A, B) = Ǿ Support A, B=1 B, C=2 Shashi Shekhar Mining For Spatial Patterns 24

Spatial Outliers Spatial Outlier: A data point that is extreme relative to it neighbors Case Study: traffic stations different from neighbors [SIGKDD 2001] Data - space-time plot, distr. Of f(x), S(x) Distribution of base attribute: spatially smooth frequency distribution over value domain: normal Classical test - Pr. [item in population] is low Q? distribution of diff. [f(x), neighborhood agg{f(x)}] Insight: this statistic is distributed normally! Test: (z-score on the statistics) > 2 Performance - spatial join, clustering methods Shashi Shekhar Mining For Spatial Patterns 25

Spatial Outlier Detection Given A spatial graph G={V, E} A neighbor relationship (K neighbors) An attribute function : V -> R An aggregation function : : R k -> R A comparison function Confidence level threshold Statistic test function ST: R ->{T, F} Find O = {vi | vi V, vi is a spatial outlier} Objective Correctness: The attribute values of vi is extreme, compared with its neighbors Computational efficiency Constraints and ST are algebraic aggregate functions of and Computation cost dominated by I/O op. Shashi Shekhar Mining For Spatial Patterns 26

Spatial Outlier Detection Test 1. Choice of Spatial Statistic S(x) = [f(x)–E y N(x)(f(y))] Theorem: S(x) is normally distributed if f(x) is normally distributed 2. Test for Outlier Detection | (S(x) - s) / s | > Hypothesis I/O cost determined by clustering efficiency f(x) Shashi Shekhar Mining For Spatial Patterns S(x) 27

Spatial Outlier Detection Results 1. CCAM achieves higher clustering efficiency (CE) 2. CCAM has lower I/O cost 3. High CE => low I/O cost 4. Big Page => high CE CE value Cell-Tree Shashi Shekhar CCAM Mining For Spatial Patterns I/O cost Z-order 28

A Unified Approach Spatial Outliers • Tests : quantitative, graphical • Results: • Computation = spatial self-join • Tests: algebraic functions of join • Join predicate: neighbor relations • I/O-cost: f(clustering efficiency) • Our algorithm is I/O-efficient for Algebric tests Scatter Plot Original Data Our Approach Shashi Shekhar Mining For Spatial Patterns 29

Graphical Spatial Tests Moran Scatter Plot Original Data Variogram Cloud Shashi Shekhar Mining For Spatial Patterns 30

Location Prediction Citations: IEEE Tran. on Multimedia 2002, SIAM DM Conf. 2001, SIGKDD DMKD 2000 Problem: predict nesting site in marshes given vegetation, water depth, distance to edge, etc. Data - maps of nests and attributes spatially clustered nests, spatially smooth attributes Classical method: logistic regression, decision trees, bayesian classifier but, independence assumption is violated ! Misses autocorrelation ! Spatial auto-regression (SAR), Markov random field bayesian classifier Open issues: spatial accuracy vs. classification accurary Open issue: performance - SAR learning is slow! Shashi Shekhar Mining For Spatial Patterns 31

Location Prediction Given: 1. Spatial Framework 2. Explanatory functions: 3. A dependent class: 4. A family of function mappings: Find: Classification model: Nest locations Distance to open water Objective: maximize classification_accuracy Constraints: Spatial Autocorrelation exists Vegetation durability Shashi Shekhar Mining For Spatial Patterns Water depth 32

Motivation and Framework Shashi Shekhar Mining For Spatial Patterns 33

Solution Procedures • Spatial Autoregression Model (SAR) • y = Wy + X + • W models neighborhood relationships • models strength of spatial dependencies • error vector • Solutions • and - can be estimated using ML or Bayesian stat. • e. g. , spatial econometrics package uses Bayesian approach using sampling-based Markov Chain Monte Carlo (MCMC) method. • Likelihood-based estimation requires O(n 3) ops. • Other alternatives – divide and conquer, sparse matrix, LU decomposition, etc. Shashi Shekhar Mining For Spatial Patterns 34

Evaluation Linear Regression Spatial model is better Shashi Shekhar Mining For Spatial Patterns 35

Solution Procedures • Markov Random Field based Bayesian Classifiers • Pr(li | X, Li) = Pr(X|li, Li) Pr(li | Li) / Pr (X) • Pr(li | Li) can be estimated from training data • Li denotes set of labels in the neighborhood of si excluding labels at si • Pr(X|li, Li) can be estimated using kernel functions • Solutions • stochastic relaxation [Geman] • Iterated conditional modes [Besag] • Graph cut [Boykov] Shashi Shekhar Mining For Spatial Patterns 36

Comparison • • SAR can be rewritten as y = (QX) + Q • where Q = (I- W)-1 which can be viewed as a spatial smoothing operation. • This transformation shows that SAR is similar to linear logistic model, and thus suffers with same limitations – i. e. , SAR model assumes linear separability of classes in transformed feature space SAR model also make more restrictive assumptions about the distribution of features and class shapes than MRF The relationship between SAR and MRF are analogous to the relationship between logistic regression and Bayesian classifiers. Our experimental results shows that MRF model yields better spatial and classification accuracies than SAR predictions. Shashi Shekhar Mining For Spatial Patterns 37

MRF vs. SAR Confusion Matrix: Spatial Confusion Matrix: Shashi Shekhar Mining For Spatial Patterns 38

Experiment Design Shashi Shekhar Mining For Spatial Patterns 39

Prediction Maps(Learning) Actual Nest Sites (Real Learning) MRF-P Prediction (ADNP=3. 36) NZ=85 MRF-GMM Prediction (ADNP=5. 88) NZ=138 SAR Prediction (ADNP=9. 80) NZ=140 Shashi Shekhar NZ=130 Mining For Spatial Patterns 40

Prediction Maps(Testing) Actual Nest Sites (Real Testing) MRF-P Prediction (ADNP=2. 84) Actual Nest Sites (Real Learning) NZ=30 NZ=80 MRF-GMM Prediction (ADNP=3. 35) NZ=76 Shashi Shekhar SAR Prediction (ADNP=8. 63) NZ=80 Mining For Spatial Patterns 41

Conclusion and Future Directions Spatial domains may not satisfy assumptions of classical methods data: auto-correlation, continuous geographic space patterns: global vs. local, e. g. spatial outliers vs. outliers data exploration: maps and albums Open Issues patterns: hot-spots, blobology (shape), spatial trends, … metrics: spatial accuracy(predicted locations), spatial contiguity(clusters) spatio-temporal dataset scale and resolutions sentivity of patterns geo-statistical confidence measure for mined patterns Shashi Shekhar Mining For Spatial Patterns 42

Reference 1. S. Shekhar, S. Chawla, S. Ravada, A. Fetterer, X. Liu and C. T. Liu, “Spatial Databases: Accomplishments and Research Needs”, IEEE Transactions on Knowledge and Data Engineering, Jan. -Feb. 1999. 2. S. Shekhar and Y. Huang, “Discovering Spatial Co-location Patterns: a Summary of Results”, In Proc. of 7 th International Symposium on Spatial and Temporal Databases (SSTD 01), July 2001. 3. S. Shekhar, C. T. Lu, P. Zhang, "Detecting Graph-based Spatial Outliers: Algorithms and Applications“, the Seventh ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2001. 4. S. Shekhar, C. T. Lu, P. Zhang, “Detecting Graph-based Saptial Outlier”, Intelligent Data Analysis, To appear in Vol. 6(3), 2002 5. S. Shekhar, S. Chawla, the book “Spatial Database: Concepts, Implementation and Trends”, Prentice Hall, 2002 6. S. Chawla, S. Shekhar, W. Wu and U. Ozesmi, “Extending Data Mining for Spatial Applications: A Case Study in Predicting Nest Locations”, Proc. Int. Confi. on 2000 ACM SIGMOD Workshop on Research Issues in Data Mining and Knowledge Discovery (DMKD 2000), Dallas, TX, May 14, 2000. 7. S. Chawla, S. Shekhar, W. Wu and U. Ozesmi, “Modeling Spatial Dependencies for Mining Geospatial Data”, First SIAM International Conference on Data Mining, 2001. 8. S. Shekhar, P. R. Schrater, R. R. Vatsavai, W. Wu, and S. Chawla, “Spatial Contextual Classification and Prediction Models for Mining Geospatial Data”, To Appear in IEEE Transactions on Multimedia, 2002. 9. S. Shekhar, V. Kumar, P. Tan. M. Steinbach, Y. Huang, P. Zhang, C. Potter, S. Klooster, “Mining Patterns in Earth Science Data”, IEEE Computing in Science and Engineering (Submitted) Shashi Shekhar Mining For Spatial Patterns 43

Reference 10. S. Shekhar, C. T. Lu, P. Zhang, “A Unified Approach to Spatial Outliers Detection”, IEEE Transactions on Knowledge and Data Engineering (Submitted) 11. S. Shekhar, C. T. Lu, X. Tan, S. Chawla, Map Cube: A Visualization Tool for Spatial Data Warehouses, as Chapter of Geographic Data Mining and Knowledge Discovery. Harvey J. Miller and Jiawei Han (eds. ), Taylor and Francis, 2001, ISBN 0 -415 -23369 -0. 12. S. Shekhar, Y. Huang, W. Wu, C. T. Lu, What's Spatial about Spatial Data Mining: Three Case Studies , as Chapter of Book: Data Mining for Scientific and Engineering Applications. V. Kumar, R. Grossman, C. Kamath, R. Namburu (eds. ), Kluwer Academic Pub. , 2001, ISBN 1 -4020 -0033 -2 13. Shashi Shekhar and Yan Huang , Multi-resolution Co-location Miner: a New Algorithm to Find Co-location Patterns in Spatial Datasets, Fifth Workshop on Mining Scientific Datasets (SIAM 2 nd Data Mining Conference), April 2002 Shashi Shekhar Mining For Spatial Patterns 44