1 / 40

PARSYMONY: Scalable Parallel Data Mining

PARSYMONY: Scalable Parallel Data Mining. Alok N. Choudhary Northwestern University (ACK: Harsha Nagesh (Bell Labs) and Sanjay Goil (Sun). Outline. Overview of PARSIMONY MAFIA (Subspace clustering) pMAFIA (Parallel Subspace Clustering) Performance Results PARSIMONY

valiant
Télécharger la présentation

PARSYMONY: Scalable Parallel Data Mining

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. PARSYMONY: Scalable Parallel Data Mining Alok N. Choudhary Northwestern University (ACK: Harsha Nagesh (Bell Labs) and Sanjay Goil (Sun) choudhar@ece.nwu.edu 1

  2. Outline • Overview of PARSIMONY • MAFIA (Subspace clustering) • pMAFIA (Parallel Subspace Clustering) • Performance Results • PARSIMONY • multidimensional data analysis • parallel classification • Summary choudhar@ece.nwu.edu 2

  3. Overview of Knowledge Discovery Process choudhar@ece.nwu.edu 3

  4. PARSIMONY Overview choudhar@ece.nwu.edu 4

  5. MAFIA: Subspace Clustering for High-Dimensional Data Sets Clustering and subspace clustering Base MAFIA Algorithm pMAFIA (parallelization) Performance Results choudhar@ece.nwu.edu 5

  6. Clustering Multi-dimensional Dimensional Data Sets Determine range of attributes in each dimension of the cluster(s) choudhar@ece.nwu.edu 7

  7. Issues to be addressed • Basic algorithm computational optimizations • Scalability with database size (out of core data sets) • Scalability with the dimensionality of data • Efficient Parallelization • Recognition of arbitrary shaped clusters choudhar@ece.nwu.edu 8

  8. Related Work • Partition based Clustering: • User specified k representative points taken as cluster centers and points assigned to cluster centers • k-means, k-mediods, CLARANS (VLDB 94), BIRCH (SIGMOD 96),.. • Consider clustering partitioning of points • Hierarchical Clustering: • Each point is a cluster. Merge similar points together gradually. • CURE (SIGMOD 98) (use sampling) • Categorical Clustering: • Clustering of categorical data e.g automobile sales data: color, year, model, price, etc • Best suited for non-numerical data • CACTUS (KDD 99), STIRR (VLDB 98) choudhar@ece.nwu.edu 9

  9. Related Work • Density and Grid Based Clustering: • Clusters are high density regions than its surroundings • WaveCluster (VLDB 98), DBSCAN, CLIQUE (SIGMOD 98) • Number of subspaces is exponential in the data dimensionality • Multidimensional space divided into grids. The histogram in each hyper-rectangle is found. Grid regions with a significant histogram value are cluster regions. • Post-processing done to grow the connected cluster regions. • Fine grid size results in explosion in the number hyper-rectangles, coarser grids fail to detect clusters. • Correct Grid Size is very critical ! choudhar@ece.nwu.edu 10

  10. Subspace Clustering • Observation: “If a collection of points S is a cluster in a k-dimensional space, then S is also a part of a cluster in any (k-1) dimensional projection of the space” • Algorithm : Growing clusters…candidate dense units in any k dimensions are obtained by merging dense units in (k-1) dimensions which share any (k-2) dimensions. • Ex: ( {a1,b7,d9}, {b7,c8,d9} ) --> {a1,b7,c8,d9} • Candidate dense units are populated by a pass on the data set and the dense units are found out in each dimension. • Dense units found are combined to form candidate dense units. • Algorithm terminates when no more candidate dense units found. choudhar@ece.nwu.edu 12

  11. Automatic Grid fitting based on data distribution MAFIA : Merging of Adaptive Finite Intervals ! Optimal Bins in each dimension leads to very few units in the grid (candidate dense units) (a) : CLIQUE (b) : MAFIA Adaptive Grids (reducing computation in practice) choudhar@ece.nwu.edu 13

  12. Base MAFIA Algorithm • Divide each dimension into very fine regions. • Compute histogram in these regions along every dimension. • Set the value of a sliding window to the maximum histogram value in the window. • Adjacent units which have nearly same histogram values are merged together to form larger bins. • Threshold of each bin formed is computed automatically. • A bin having a histogram value much greater (by a factor 2~3) than that of equi-distribution of data is DENSE. choudhar@ece.nwu.edu 14

  13. choudhar@ece.nwu.edu 15

  14. MAFIA Algorithm (merging dense units) • Algorithm : Candidate dense units in any k dimensions are obtained by merging dense units in (k-1) dimensions which share any(k-2) dimensions. • Ex: ( {a1,c7,b8}, {c7,b8,d9} ) --> {a1,c7,b8,d9} choudhar@ece.nwu.edu 16

  15. CLIQUE: CDUs in any dimension k is formed by combining dense units of dimension (k-1) which share first (k-2) dimensions. MAFIA: CDUs in any dimension k is formed by combining dense units of dimension (k-1) which share any(k-2) dimensions. Huge CDU Set Non Cluster Dims reported CLIQUE Fixed Grids Fixed Grids first (k-2) algo Huge Search Space Data Set much reduced CDU Set Adaptive Grids Correct Cluster Dims reported any (k-2) algo MAFIA Reduced Search Space Dimensions aware of data distribution choudhar@ece.nwu.edu 17

  16. Parallel MAFIA • pMAFIA: Parallel Subspace Clustering • Scalable in data size and number of dimensions Grid and Density based clustering algorithm • Parallelization provides speedup for the subspace clustering algorithm. choudhar@ece.nwu.edu 18

  17. pMAFIA Algorithm: • Each processor reads part of the data in a Data Parallel fashion and constructs histogram in every dimension. • // Data read in chunks (out of core) of data • Reduce communication to obtain global histogram. All processors build Adaptive grids using the histogram. • // Each bin formed is a Candidate Dense Unit. • Current Dimension k = 1 • while (no more dense units found) • if( k > 1) • Build Candidate Dense units(); • Populate the candidate dense units in a data parallel fashion and in chunks (out-of-core) of B records. • Reduce communication to obtain global CDU population. • Identify the dense units (); • Build dense unit data structures(); // for the next higher dimension choudhar@ece.nwu.edu 19

  18. Current Dimension(k) = 3. Build-Candidate-Dense-Units choudhar@ece.nwu.edu 20

  19. Build Candidate Dense Units • Each dense unit is compared with every other dense unit to form CDUs, resulting in an O(Ndu2) algorithm.(Ndu-number of dense units) • For large values of Ndu CDUs built in parallel. • Processors 0,..,(p-1) work in parallel on parts of total Ndu dense units. Processor k compares dense units between Ni and Ni+1 with all the other dense units; for optimal task partitioning we have • Identical CDUs generated during the process need to be discarded. • Each generated CDU compared with every other CDU to identify similar ones resulting in O(Ncdu2) algorithm. choudhar@ece.nwu.edu 22

  20. Parallelization (Building CDUs) Total work done by Pi is shown choudhar@ece.nwu.edu 24

  21. pMAFIA : ANALYSIS • Data parallelism in populating the CDUs in every dimension : effective for massive data sets. • Gains of task parallelism realized when data contains large number of dense units (clusters). • A ‘k’ dimensional dense unit allocated just 2k bytes of memory, k bytes for dimensions and k for bin indices. • Data structures in form of linear arrays of bytes: communicate very small message buffers, space optimization. • Although bottom-up algorithm is exponential in the data dimension, for low subspace coverage, with use of Adaptive grids and parallel formulation very promising results. • k dimension of the highest dimension dense unit, we explore all possible subspaces of these k dimensions ==> O(ck) • For k passes over the data set O( (N/pB) * k*Tio), • N-total number of records, p-processors, • B- records per chunk, Tio-I/O access time for a block of B records. • Communication overhead results in O(Tcomm * S * p * k), • Tcomm - constant for communication, S- size of message exchanged,. • O(ck + (N/pB) * k*Ti + Tcomm * S * p * k ) choudhar@ece.nwu.edu 25

  22. pMAFIA Outline • Clustering and subspace clustering • Base MAFIA Algorithm • pMAFIA (parallelization) • Performance Results • Adaptivity performance • Scalability with data set size. • Scalability with data dimensionality. • Scalability with cluster dimensionality. • Some Real world data sets. choudhar@ece.nwu.edu 26

  23. Advantage of Adaptive Grids • 300,000 records in a 15 Dimension space with 1 cluster of 5 dimensions. • A speedup of 80 obtained over CLIQUE. • CLIQUE failed to produce results with our modified CDU generation algorithm even in 2 hours on 16 processors. • This relatively small data set mined in 32 seconds on 1 processor choudhar@ece.nwu.edu 29

  24. Scalability with Data Set Size • 20 Dimension data with 5 clusters in 5 different subspaces • data sets :up to 11.8 million records • Clusters detected in just about 3 minutes on 16 processors ! • Almost Linear with the increase in the data set size (because most time in scanning) choudhar@ece.nwu.edu 30

  25. Parallelization (on IBM SP2) • 30 Dimension data with 8.3 million records, 5 clusters each in a 6 dimension subspace. • Near linear speedups • Negligible Communication overheads (<1%) choudhar@ece.nwu.edu 31

  26. Data Dimensionality • 250,000 records, 3 clusters in different 5 dimensional subspaces. • Near linear behavior with data dimensionality : Algorithm depends on the maximum number of dimensions in a clusters and not on the data dimensionality. choudhar@ece.nwu.edu 32

  27. Cluster Dimensionality • 50 dimension data with 1 cluster, 650,000 records; cluster dimension from 3 to 10. • Behavior in line with the order of the algorithm : increases with subspace coverage of the cluster. choudhar@ece.nwu.edu 33

  28. Scalability on Movie Data • 72,916 users rated 1628 movies in 18 months: 2.8 Million ratings • 4D data: {user-id, movie-id, weight, score} • Discovered seven interesting 2d clusters! choudhar@ece.nwu.edu 34

  29. Other Data Sets • One day Ahead Prediction of DAX (German Stock Exchange) • DAX prediction data set was based on a 12 input time series like stock indices, bond indices, gold prices, etc • 22 dimensions with 2757 records: Major gains from task parallelism • Mined clusters in 8.16 seconds on 8 processors. • Unique clusters discovered in 3,4,5 and 6 dimensional subspaces. • Ionosphere data: (UCI repository) • Radar data collected in Goose Bay, Labrador • 34 dimension data, 351 records. • Discovered unique clusters only in 3 and 4 dimensional sub spaces. choudhar@ece.nwu.edu 35

  30. Summary and Conclusions for pMAFIA • MAFIA : unsupervised subspace clustering algorithm • Introduced the adaptive grids formulation • First parallel subspace clustering algorithm for massivedata sets • Incorporates both task and data parallelism. . • pMAFIA : Scalable and Parallel Implementation in data size, no of dimensions choudhar@ece.nwu.edu 37

  31. PARSIMONY Overview choudhar@ece.nwu.edu 38

  32. Sparse Data Structures and Representations choudhar@ece.nwu.edu 39

  33. choudhar@ece.nwu.edu 40

  34. Using OLAP Framework for choudhar@ece.nwu.edu 41

  35. choudhar@ece.nwu.edu 42

  36. choudhar@ece.nwu.edu 43

  37. choudhar@ece.nwu.edu 44

  38. choudhar@ece.nwu.edu 45

  39. choudhar@ece.nwu.edu 46

  40. choudhar@ece.nwu.edu 47

More Related