Exact data mining from in- exact data Nick Freris

Size: px
Start display at page:

Download "Exact data mining from in- exact data Nick Freris"

Transcription

1 Exact data mining from in- exact data Nick Freris Qualcomm, San Diego October 10, 2013

2 Introduc=on (1) Informa=on retrieval is a large industry.. Biology, finance, engineering, marke=ng, vision/graphics, etc...but data are hardly ever maintained in original form Compression Rights protection Original Quantized Watermarking

3 Preserva=on of the mining outcome (2) this talk: provable guarantees for distanced- based mining Optimal distance estimation K-means preservation D=11 ˆD 2 [11, 12]

4 Op=mal distance es=ma=on (3) This work presents op=mal es=ma=on of Euclidean distance in the compressed domain. This result cannot be further improved D = 11.5 ˆD 2 [11, 12] Our approach is applicable on any orthonormal data compression basis (Fourier, Wavelets, Chebyshev, PCA, etc.) Our method allows up to 57% bewer distance es=ma=on 80% less computa=on effort 128 : 1 compression efficiency

5 Mo=va=on/Benefits Time- series data are customarily compressed in order to: Save storage space Reduce transmission bandwidth Remove noise Achieve faster processing / data analysis (4) Distance es=ma=on has various data analy=cs applica=ons Clustering / Classifica=on Anomaly detec=on Similarity search Now we can do all this very efficiently directly on the compressed data!

6 Previous work vs current approach First coeffs + error Best coeffs + error Best coeffs + error + constraints à Op=mal (our method) (5) Approach 1 Approach 2: (only one compressed) Our approach

7 (6) applica=ons Similarity Search/Grouping/Clustering find seman=cally similar searches, deduce associa=ons dynamic inventory op=miza=on Five key supply chain areas Chief supply chain officer IBM s dynamic inventory integrated supply chain Smarter supply chain Retail Supply Management Business consultant supply chain management QUERY: Supply Chain Results with similar query pawerns

8 (6) Similarity Search/Grouping/Clustering applica=ons Adver=sing/recommenda=ons Buying keywords that are cheaper but exhibit same demand pawern Requests Query: business services Query: business analytics Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

9 (6) Burst Detec=on Discovery of important events applica=ons Query: Acquisition Feb 10: Intelliden Inc. Acquisition Requests May 10: Sterling Commerce Acquisition (1.4B) Jan Feb Mar Apr May Jun Jul Aug Sep Okt Nov Dec

10 (7) underlying opera=on: k- NN search K- Nearest Neighbor (k- NN) similarity search Issues that arise: How to compress data? How to speedup search We have to es=mate =ght bounds on the distance metric using just the compressed representa=on

11 (8) similarity search problem query D = 7.3 Distance Linear Scan: Objec've: Compare the query with all sequences in DB and return the k most similar sequences to the query. D = 10.2 D = 11.8 D = 17 D = 22

12 (8) speeding up similarity search simplified DB Candidate Superset original DB Final Answer set Upper / lower bounds on distance Verify against original DB simplified query keyword 1 keyword 2 keyword 3 keyword

13 (9) compressing weblog data Use Euclidean distance to match =me- series. But how should we compress the data? Query: analytics and optimization 1 year span The data is smooth and highly periodic, so we can use Fourier decomposition. Instead of using the first Fourier coefficients we can use the best ones instead. Query: consulting services

14 (10) Discrete Fourier Transform (DFT) decomposi=on of =me- series into sinusoids Every signal can be represented as a superposition of sines and cosines X(f) i i i i i i i i i i i i i i i i i i i i i i i i x(n) Jean Baptiste Fourier ( )

15 (10) spectrum P(f) X(f) x(n) i Keep just some frequencies of periodogram i i i i i i Sequence: i i i i i i Periodogram: i i i i i f i i i i i i

16 (10) spectrum P(f) X(f) x(n) i Keep just some frequencies of periodogram i i i i i i Sequence: i i i i i i Periodogram: i i i i i f i i day period i i i i

17 (10) spectrum P(f) X(f) x(n) i Keep just some frequencies of periodogram i i i i i i Sequence: i i i i i i Periodogram: i i i i i f i i day period i i i i

18 (10) spectrum P(f) X(f) x(n) i Keep just some frequencies of periodogram i i i i i i Sequence: i i i i i i Periodogram: i i i i i Reconstruction: f i i i i i i

19 (11) similarity Search Approximate the Euclidean distance using DFT First 5 Coefficients +symmetric ones x(n) X(f) i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i

20 (11) similarity Search Approximate the Euclidean distance using DFT Best 5 Coefficients + symmetric ones x(n) X(f) i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i

21 (12) objec=ve Calculate the =ghtest possible upper/lower bounds using the coefficients with the highest energy This will result in bewer pruning of the search space - > faster search

22 (13) x X Q q i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i

23 (13) Find best k coefficients of X (k=4) Magnitude vector X X Q i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i Q

24 (13) Find best 4 coefficients of T X X Q i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i Q

25 (13) minpower Iden=fy smallest magnitude (à power) X X Q i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i Q

26 (13) minpower The remaining powers are less than minpower X X Q i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i Q

27 (13) e 2 We keep also the sum of squares of the remaining powers X X Q i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i Q

28 (13) Calculate distance from k known coeffs X X Q i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i Q

29 (13) minpower Op=mize distance from remaining coefficients X X Q i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i Q Op=miza=on Generaliza=on

30 (14) water- filling d 2 Q X = -α Q d 1

31 (14) water- filling d 2 Q e Boundary condi=on on total energy: d d 22 < e 2 X = -α Q d 1

32 (14) water- filling d 2 Q minp e Boundary condi=on on each dimension: d 1 < minpower, d 2 < minpower X = -α Q d 1 minp

33 (14) water- filling d 2 Q minp e We hit one of our constraints X = -α Q d 1 minp

34 (14) water- filling d 2 Q minp e We start moving on the other dimension un=l we use all our energy X = -α Q d 1 minp

35 (14) in N- dimensions Q X minpower Available Energy

36 (14) in N- dimensions Q X minpower Available Energy

37 (14) in N- dimensions Q X minpower Available Energy

38 (14) in N- dimensions Q X minpower Available Energy

39 (14) in N- dimensions Q X minpower Available Energy

40 (14) in N- dimensions Q X minpower Available Energy

41 (14) in N- dimensions Q X minpower Available Energy

42 (14) in N- dimensions Q X minpower Available Energy

43 (14) in N- dimensional space Q X minpower Available Energy

44 (15) When both sequences are compressed the op=mal distance es=ma=on can be solved using a double waterfilling process. single waterfilling double waterfilling

45 the bounds are op=mally =ght (16) Theorem (Freris et al. 2012): The computa=on of lower and upper bounds given the aforemen=oned compressed representa=ons can be solved exactly using double water- filling problem. The lower and upper bounds are op=mally =ght; no =ghter solu=ons can be provided Sketch of proof: - convex re- parametriza=on - KKT condi=ons - proper=es of op=mal solu=ons

46 Double Water- filling algorithm (17) Intui=on: Water- fill for the discarded coefficients of the two vectors separately....using the appropriate energy alloca=on Zero- overhead algorithm Complexity: Θ(nlong)

47 Experiments (18) Unica database. IBM web traffic for year of 2010 Marke=ng/Adwords recommenda=on Analysis/Storage of weblog queries (1TB of data per month) GBS: Scheduling adver=sing campaigns / pricing BUSINESS DYNAMICS IBM YIN YANG OF FINANCIAL DISRUPTION EINSURANCE CUSTOMER EXPERIENCE. IBM GLOBAL BUSINESS ANDREW STEVENS BUSINESS CONSULTING GLENN FINCH IBM AMERICA MEDIA PLAYER INDUSTRY STRATEGIE ENTREPRISE RENTABILIT

48 our analy=c solu=on is 300x faster than convex op=mizers (19)

49 Our algorithm converges very fast: in 2-3 itera=ons irrespec=ve of the compression rate (20)

50 the derived LB/UB are 20% bewer than state- of- art (21)

51 when searching in a DB this results in up to 80% data pruning the previous (10-20%) improvement in distance es=ma=on can significantly reduce the search space when searching for k- NN results (22) We retrieve 20%-80% fewer sequences than other approaches

52 Conclusions (23) Increasing data sizes is a perennial problem for data analysis ª It is important to support mining directly on the compressed data We have shown an exact algorithm for obtaining op=mally =ght Euclidean bounds on compressed representa=ons ª Mining directly on compressed data with provable guarantees Many generaliza=ons: Cosine Similarity (text documents): cos(x,y) = 1 - L 2 (x,y) 2 /2 Correla=on (financial analysis): corr(x,y) = 1 - L 2 (x,y) 2 /2 (for normalized signals x,y) Dynamic Time Warping (flexible similarity metric):

53 Compression with provable K- means preserva=on

54 Scope of this work (24) Mul=- bit compression and inver=ble transforma=on of high- dimensional data Ø with provable K- means preserva=on cluster 1 cluster 2 cluster 3 cluster 1 cluster 2 cluster 3 identical clustering results K-means" K-means" original data Original data quantized data Compressed data

55 Applica=ons (25) Reduced storage requirements Reduced processing =me Data hiding / anonymiza=on Original values are hidden via an inver=ble cluster contrac=on Encoder- decoder scheme Shape preserva=on High- quality data reconstruc=on Extends applicability to other mining opera=ons such as Nearest- Neighbor search, data visualiza=on, etc. Tunable compression level based on number of allocated bits in order to achieve good storage/quality trade- off

56 Overview (26) 1. Pre- clustering 2. Quan=za=on per cluster and dimension (one or mul=ple bits) 3. Data contrac=on transforma=on 4. Post- quan=za=on clustering 5. Decoding using key (contrac=on parameter)

57 Our method effec=vely compresses the data (27) 1. MMSE quan=za=on per cluster and dimension (one or mul=ple bits) Preserves mean and reduces variance and dynamic range Example: 1- bit MMSE quan=za=on Quan=zer:

58 (28) How does our scheme apply to high- dimensional data? These are the points for one dimension and for one cluster of objects. Dimension d (or time instance )d) Process is repeated for all dimensions and for all clusters. We have one quan=zer per class

59 (29) We can also support Data- Hiding with our scheme 2. Data contrac=on transforma=on Inver=ble Preserves centroids and reduces variance and dynamic range BeWer cluster separa=on

60 Results (30) Theore=cal guarantee on cluster preserva=on for sufficiently contracted clusters (sufficiently small a) Analy=cal calcula=on of op=mal contrac=on coefficient (a) K(K- 1) calcula=ons, K: number of clusters Lower distor=on (MSE) than MPQ which decreases with the number of allocated bits Mul=- bit alloca=on (per cluster) Generaliza=on of MMSE quan=zers for mul=ple bits Efficient algorithm for op=mal alloca=on given storage constraints

61 (31) We preserve 100% the clustering structure Stock values 2100 stocks during a period of three years Clustered in 3,5,8 clusters Cluster preserva=on was always 100% Shape preserva=on

62 Mul=- Bit quan=za=on improves quality up to 96%! MSE due to quan=za=on decreases with number of bits and number of clusters (32) 37%, 70% and 96% over MPQ using MMSE with 1, 2, 4 bits 76%, 84% using 5, 8 clusters over 3 clusters

63 Compress > 8x without loss in clustering accuracy (33) Compression efficiency: 1.8x 8x reduc=on

64 Conclusions (34) Efficient K- means preserving compression of high- dimensional data Data hiding (encoder- decoder data transforma=on) Tunable compression level (storage / distor=on trade- off) Theore=cal guarantees Our scheme is the first one which always guarantees 100% cluster preserva=on Efficient bit alloca=on algorithm for op=mal distor=on given storage constraints Experiments 100% cluster preserva=on Excellent shape preserva=on even for 1 bit quan=zers High compression efficiency Extensions Our approach works with any quan=za=on scheme Example: Minimum Absolute Error Quan=za=on (MAE)

65 1. Distributed es=ma=on in sensor networks Other research (35) 2. Randomized algorithms Ax = b 3. Compressed sensing 4. Clock synchroniza=on Convergence lab (CSL, Univ. Illinois) 5. Wireless resource alloca=on 6. Video streaming SmartSense, EPFL

66 Thank you! Ques=ons

Cri$ques Ø 5 cri&ques in total Ø Each with 6 points

Cri$ques Ø 5 cri&ques in total Ø Each with 6 points Cri$ques Ø 5 cri&ques in total Ø Each with 6 points 1 Distributed Applica$on Alloca$on in Shared Sensor Networks Chengjie Wu, You Xu, Yixin Chen, Chenyang Lu Shared Sensor Network Example in San Francisco

More information

GAMINGRE 8/1/ of 7

GAMINGRE 8/1/ of 7 FYE 09/30/92 JULY 92 0.00 254,550.00 0.00 0 0 0 0 0 0 0 0 0 254,550.00 0.00 0.00 0.00 0.00 254,550.00 AUG 10,616,710.31 5,299.95 845,656.83 84,565.68 61,084.86 23,480.82 339,734.73 135,893.89 67,946.95

More information

SKA machine learning perspec1ves for imaging, processing and analysis

SKA machine learning perspec1ves for imaging, processing and analysis 1 SKA machine learning perspec1ves for imaging, processing and analysis Slava Voloshynovskiy Stochas1c Informa1on Processing Group University of Geneva Switzerland with contribu,on of: D. Kostadinov, S.

More information

Lecture 8: Principal Component Analysis; Kernel PCA

Lecture 8: Principal Component Analysis; Kernel PCA Lecture 8: Principal Component Analysis; Kernel PCA Lester Mackey April 23, 2014 Stats 306B: Unsupervised Learning Sta306b April 19, 2011 Principal components: 16 PCA example: digit data 130 threes, a

More information

Wavelets & Mul,resolu,on Analysis

Wavelets & Mul,resolu,on Analysis Wavelets & Mul,resolu,on Analysis Square Wave by Steve Hanov More comics at http://gandolf.homelinux.org/~smhanov/comics/ Problem set #4 will be posted tonight 11/21/08 Comp 665 Wavelets & Mul8resolu8on

More information

Image Data Compression. Dirty-paper codes Alexey Pak, Lehrstuhl für Interak<ve Echtzeitsysteme, Fakultät für Informa<k, KIT

Image Data Compression. Dirty-paper codes Alexey Pak, Lehrstuhl für Interak<ve Echtzeitsysteme, Fakultät für Informa<k, KIT Image Data Compression Dirty-paper codes 1 Reminder: watermarking with side informa8on Watermark embedder Noise n Input message m Auxiliary informa

More information

Computation of Large Sparse Aggregated Areas for Analytic Database Queries

Computation of Large Sparse Aggregated Areas for Analytic Database Queries Computation of Large Sparse Aggregated Areas for Analytic Database Queries Steffen Wittmer Tobias Lauer Jedox AG Collaborators: Zurab Khadikov Alexander Haberstroh Peter Strohm Business Intelligence and

More information

Average 175, , , , , , ,046 YTD Total 1,098,649 1,509,593 1,868,795 1,418, ,169 1,977,225 2,065,321

Average 175, , , , , , ,046 YTD Total 1,098,649 1,509,593 1,868,795 1,418, ,169 1,977,225 2,065,321 AGRICULTURE 01-Agriculture JUL 2,944-4,465 1,783-146 102 AUG 2,753 6,497 5,321 1,233 1,678 744 1,469 SEP - 4,274 4,183 1,596 - - 238 OCT 2,694 - - 1,032 340-276 NOV 1,979-5,822 637 3,221 1,923 1,532 DEC

More information

Average 175, , , , , , ,940 YTD Total 944,460 1,284,944 1,635,177 1,183, ,954 1,744,134 1,565,640

Average 175, , , , , , ,940 YTD Total 944,460 1,284,944 1,635,177 1,183, ,954 1,744,134 1,565,640 AGRICULTURE 01-Agriculture JUL 2,944-4,465 1,783-146 102 AUG 2,753 6,497 5,321 1,233 1,678 744 1,469 SEP - 4,274 4,183 1,596 - - 238 OCT 2,694 - - 1,032 340-276 NOV 1,979-5,822 637 3,221 1,923 1,532 DEC

More information

Lecture 17: Face Recogni2on

Lecture 17: Face Recogni2on Lecture 17: Face Recogni2on Dr. Juan Carlos Niebles Stanford AI Lab Professor Fei-Fei Li Stanford Vision Lab Lecture 17-1! What we will learn today Introduc2on to face recogni2on Principal Component Analysis

More information

Lecture 17: Face Recogni2on

Lecture 17: Face Recogni2on Lecture 17: Face Recogni2on Dr. Juan Carlos Niebles Stanford AI Lab Professor Fei-Fei Li Stanford Vision Lab Lecture 17-1! What we will learn today Introduc2on to face recogni2on Principal Component Analysis

More information

WHEN IS IT EVER GOING TO RAIN? Table of Average Annual Rainfall and Rainfall For Selected Arizona Cities

WHEN IS IT EVER GOING TO RAIN? Table of Average Annual Rainfall and Rainfall For Selected Arizona Cities WHEN IS IT EVER GOING TO RAIN? Table of Average Annual Rainfall and 2001-2002 Rainfall For Selected Arizona Cities Phoenix Tucson Flagstaff Avg. 2001-2002 Avg. 2001-2002 Avg. 2001-2002 October 0.7 0.0

More information

Least Square Es?ma?on, Filtering, and Predic?on: ECE 5/639 Sta?s?cal Signal Processing II: Linear Es?ma?on

Least Square Es?ma?on, Filtering, and Predic?on: ECE 5/639 Sta?s?cal Signal Processing II: Linear Es?ma?on Least Square Es?ma?on, Filtering, and Predic?on: Sta?s?cal Signal Processing II: Linear Es?ma?on Eric Wan, Ph.D. Fall 2015 1 Mo?va?ons If the second-order sta?s?cs are known, the op?mum es?mator is given

More information

CS 6140: Machine Learning Spring What We Learned Last Week 2/26/16

CS 6140: Machine Learning Spring What We Learned Last Week 2/26/16 Logis@cs CS 6140: Machine Learning Spring 2016 Instructor: Lu Wang College of Computer and Informa@on Science Northeastern University Webpage: www.ccs.neu.edu/home/luwang Email: luwang@ccs.neu.edu Sign

More information

Randomized Kaczmarz Nick Freris EPFL

Randomized Kaczmarz Nick Freris EPFL Randomized Kaczmarz Nick Freris EPFL (Joint work with A. Zouzias) Outline Randomized Kaczmarz algorithm for linear systems Consistent (noiseless) Inconsistent (noisy) Optimal de-noising Convergence analysis

More information

Unsupervised Learning: K- Means & PCA

Unsupervised Learning: K- Means & PCA Unsupervised Learning: K- Means & PCA Unsupervised Learning Supervised learning used labeled data pairs (x, y) to learn a func>on f : X Y But, what if we don t have labels? No labels = unsupervised learning

More information

Cosmological N-Body Simulations and Galaxy Surveys

Cosmological N-Body Simulations and Galaxy Surveys Cosmological N-Body Simulations and Galaxy Surveys Adrian Pope, High Energy Physics, Argonne Na3onal Laboratory, apope@anl.gov CScADS: Scien3fic Data and Analy3cs for Extreme- scale Compu3ng, 30 July 2012

More information

Bias/variance tradeoff, Model assessment and selec+on

Bias/variance tradeoff, Model assessment and selec+on Applied induc+ve learning Bias/variance tradeoff, Model assessment and selec+on Pierre Geurts Department of Electrical Engineering and Computer Science University of Liège October 29, 2012 1 Supervised

More information

Seman&cs with Dense Vectors. Dorota Glowacka

Seman&cs with Dense Vectors. Dorota Glowacka Semancs with Dense Vectors Dorota Glowacka dorota.glowacka@ed.ac.uk Previous lectures: - how to represent a word as a sparse vector with dimensions corresponding to the words in the vocabulary - the values

More information

REPORT ON LABOUR FORECASTING FOR CONSTRUCTION

REPORT ON LABOUR FORECASTING FOR CONSTRUCTION REPORT ON LABOUR FORECASTING FOR CONSTRUCTION For: Project: XYZ Local Authority New Sample Project Contact us: Construction Skills & Whole Life Consultants Limited Dundee University Incubator James Lindsay

More information

Streaming - 2. Bloom Filters, Distinct Item counting, Computing moments. credits:www.mmds.org.

Streaming - 2. Bloom Filters, Distinct Item counting, Computing moments. credits:www.mmds.org. Streaming - 2 Bloom Filters, Distinct Item counting, Computing moments credits:www.mmds.org http://www.mmds.org Outline More algorithms for streams: 2 Outline More algorithms for streams: (1) Filtering

More information

Predic've Analy'cs for Energy Systems State Es'ma'on

Predic've Analy'cs for Energy Systems State Es'ma'on 1 Predic've Analy'cs for Energy Systems State Es'ma'on Presenter: Team: Yingchen (YC) Zhang Na/onal Renewable Energy Laboratory (NREL) Andrey Bernstein, Rui Yang, Yu Xie, Anthony Florita, Changfu Li, ScoD

More information

In Centre, Online Classroom Live and Online Classroom Programme Prices

In Centre, Online Classroom Live and Online Classroom Programme Prices In Centre, and Online Classroom Programme Prices In Centre Online Classroom Foundation Certificate Bookkeeping Transactions 430 325 300 Bookkeeping Controls 320 245 225 Elements of Costing 320 245 225

More information

CSCI 360 Introduc/on to Ar/ficial Intelligence Week 2: Problem Solving and Op/miza/on

CSCI 360 Introduc/on to Ar/ficial Intelligence Week 2: Problem Solving and Op/miza/on CSCI 360 Introduc/on to Ar/ficial Intelligence Week 2: Problem Solving and Op/miza/on Professor Wei-Min Shen Week 13.1 and 13.2 1 Status Check Extra credits? Announcement Evalua/on process will start soon

More information

FEB DASHBOARD FEB JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC

FEB DASHBOARD FEB JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC Positive Response Compliance 215 Compliant 215 Non-Compliant 216 Compliant 216 Non-Compliant 1% 87% 96% 86% 96% 88% 89% 89% 88% 86% 92% 93% 94% 96% 94% 8% 6% 4% 2% 13% 4% 14% 4% 12% 11% 11% 12% JAN MAR

More information

Jayalath Ekanayake Jonas Tappolet Harald Gall Abraham Bernstein. Time variance and defect prediction in software projects: additional figures

Jayalath Ekanayake Jonas Tappolet Harald Gall Abraham Bernstein. Time variance and defect prediction in software projects: additional figures Jayalath Ekanayake Jonas Tappolet Harald Gall Abraham Bernstein TECHNICAL REPORT No. IFI-2.4 Time variance and defect prediction in software projects: additional figures 2 University of Zurich Department

More information

COMP 562: Introduction to Machine Learning

COMP 562: Introduction to Machine Learning COMP 562: Introduction to Machine Learning Lecture 20 : Support Vector Machines, Kernels Mahmoud Mostapha 1 Department of Computer Science University of North Carolina at Chapel Hill mahmoudm@cs.unc.edu

More information

A Framework for Protec/ng Worker Loca/on Privacy in Spa/al Crowdsourcing

A Framework for Protec/ng Worker Loca/on Privacy in Spa/al Crowdsourcing A Framework for Protec/ng Worker Loca/on Privacy in Spa/al Crowdsourcing Nov 12 2014 Hien To, Gabriel Ghinita, Cyrus Shahabi VLDB 2014 1 Mo/va/on Ubiquity of mobile users 6.5 billion mobile subscrip/ons,

More information

Computer Vision. Pa0ern Recogni4on Concepts Part I. Luis F. Teixeira MAP- i 2012/13

Computer Vision. Pa0ern Recogni4on Concepts Part I. Luis F. Teixeira MAP- i 2012/13 Computer Vision Pa0ern Recogni4on Concepts Part I Luis F. Teixeira MAP- i 2012/13 What is it? Pa0ern Recogni4on Many defini4ons in the literature The assignment of a physical object or event to one of

More information

Introduction to Forecasting

Introduction to Forecasting Introduction to Forecasting Introduction to Forecasting Predicting the future Not an exact science but instead consists of a set of statistical tools and techniques that are supported by human judgment

More information

Parameter Es*ma*on: Cracking Incomplete Data

Parameter Es*ma*on: Cracking Incomplete Data Parameter Es*ma*on: Cracking Incomplete Data Khaled S. Refaat Collaborators: Arthur Choi and Adnan Darwiche Agenda Learning Graphical Models Complete vs. Incomplete Data Exploi*ng Data for Decomposi*on

More information

The Spectrum of Broadway: A SAS

The Spectrum of Broadway: A SAS The Spectrum of Broadway: A SAS PROC SPECTRA Inquiry James D. Ryan and Joseph Earley Emporia State University and Loyola Marymount University Abstract This paper describes how to use the sophisticated

More information

CSE P 501 Compilers. Value Numbering & Op;miza;ons Hal Perkins Winter UW CSE P 501 Winter 2016 S-1

CSE P 501 Compilers. Value Numbering & Op;miza;ons Hal Perkins Winter UW CSE P 501 Winter 2016 S-1 CSE P 501 Compilers Value Numbering & Op;miza;ons Hal Perkins Winter 2016 UW CSE P 501 Winter 2016 S-1 Agenda Op;miza;on (Review) Goals Scope: local, superlocal, regional, global (intraprocedural), interprocedural

More information

Compressing Kinetic Data From Sensor Networks. Sorelle A. Friedler (Swat 04) Joint work with David Mount University of Maryland, College Park

Compressing Kinetic Data From Sensor Networks. Sorelle A. Friedler (Swat 04) Joint work with David Mount University of Maryland, College Park Compressing Kinetic Data From Sensor Networks Sorelle A. Friedler (Swat 04) Joint work with David Mount University of Maryland, College Park Motivation Motivation Computer Science Graphics: Image and video

More information

Time Series Analysis

Time Series Analysis Time Series Analysis A time series is a sequence of observations made: 1) over a continuous time interval, 2) of successive measurements across that interval, 3) using equal spacing between consecutive

More information

Corn Basis Information By Tennessee Crop Reporting District

Corn Basis Information By Tennessee Crop Reporting District UT EXTENSION THE UNIVERSITY OF TENNESSEE INSTITUTE OF AGRICULTURE AE 05-13 Corn Basis Information By Tennessee Crop Reporting District 1994-2003 Delton C. Gerloff, Professor The University of Tennessee

More information

Wolf-Tilo Balke Silviu Homoceanu Institut für Informationssysteme Technische Universität Braunschweig

Wolf-Tilo Balke Silviu Homoceanu Institut für Informationssysteme Technische Universität Braunschweig Multimedia Databases Wolf-Tilo Balke Silviu Homoceanu Institut für Informationssysteme Technische Universität Braunschweig http://www.ifis.cs.tu-bs.de 14 Indexes for Multimedia Data 14 Indexes for Multimedia

More information

Monthly Trading Report July 2018

Monthly Trading Report July 2018 Monthly Trading Report July 218 Figure 1: July 218 (% change over previous month) % Major Market Indicators 2 2 4 USEP Forecasted Demand CCGT/Cogen/Trigen Supply ST Supply Figure 2: Summary of Trading

More information

Compression. What. Why. Reduce the amount of information (bits) needed to represent image Video: 720 x 480 res, 30 fps, color

Compression. What. Why. Reduce the amount of information (bits) needed to represent image Video: 720 x 480 res, 30 fps, color Compression What Reduce the amount of information (bits) needed to represent image Video: 720 x 480 res, 30 fps, color Why 720x480x20x3 = 31,104,000 bytes/sec 30x60x120 = 216 Gigabytes for a 2 hour movie

More information

ISO Lead Auditor Lean Six Sigma PMP Business Process Improvement Enterprise Risk Management IT Sales Training

ISO Lead Auditor Lean Six Sigma PMP Business Process Improvement Enterprise Risk Management IT Sales Training Training Calendar 2014 Public s (ISO LSS PMP BPI ERM IT Sales Training) www.excelledia.com (ISO, LSS, PMP, BPI, ERM, IT, Sales Public s) 1 Schedule Registration JANUARY FEBRUARY 2 days 26 JAN 27 JAN 3

More information

Founda'ons of Large- Scale Mul'media Informa'on Management and Retrieval. Lecture #3 Machine Learning. Edward Chang

Founda'ons of Large- Scale Mul'media Informa'on Management and Retrieval. Lecture #3 Machine Learning. Edward Chang Founda'ons of Large- Scale Mul'media Informa'on Management and Retrieval Lecture #3 Machine Learning Edward Y. Chang Edward Chang Founda'ons of LSMM 1 Edward Chang Foundations of LSMM 2 Machine Learning

More information

Pre-Calc Chapter 1 Sample Test. D) slope: 3 4

Pre-Calc Chapter 1 Sample Test. D) slope: 3 4 Pre-Calc Chapter 1 Sample Test 1. Use the graphs of f and g to evaluate the function. f( x) gx ( ) (f o g)(-0.5) 1 1 0 4. Plot the points and find the slope of the line passing through the pair of points.

More information

BUSI 460 Suggested Answers to Selected Review and Discussion Questions Lesson 7

BUSI 460 Suggested Answers to Selected Review and Discussion Questions Lesson 7 BUSI 460 Suggested Answers to Selected Review and Discussion Questions Lesson 7 1. The definitions follow: (a) Time series: Time series data, also known as a data series, consists of observations on a

More information

Monthly Trading Report Trading Date: Dec Monthly Trading Report December 2017

Monthly Trading Report Trading Date: Dec Monthly Trading Report December 2017 Trading Date: Dec 7 Monthly Trading Report December 7 Trading Date: Dec 7 Figure : December 7 (% change over previous month) % Major Market Indicators 5 4 Figure : Summary of Trading Data USEP () Daily

More information

Mr. XYZ. Stock Market Trading and Investment Astrology Report. Report Duration: 12 months. Type: Both Stocks and Option. Date: Apr 12, 2011

Mr. XYZ. Stock Market Trading and Investment Astrology Report. Report Duration: 12 months. Type: Both Stocks and Option. Date: Apr 12, 2011 Mr. XYZ Stock Market Trading and Investment Astrology Report Report Duration: 12 months Type: Both Stocks and Option Date: Apr 12, 2011 KT Astrologer Website: http://www.softwareandfinance.com/magazine/astrology/kt_astrologer.php

More information

Lecture Prepared By: Mohammad Kamrul Arefin Lecturer, School of Business, North South University

Lecture Prepared By: Mohammad Kamrul Arefin Lecturer, School of Business, North South University Lecture 15 20 Prepared By: Mohammad Kamrul Arefin Lecturer, School of Business, North South University Modeling for Time Series Forecasting Forecasting is a necessary input to planning, whether in business,

More information

Improve Forecasts: Use Defect Signals

Improve Forecasts: Use Defect Signals Improve Forecasts: Use Defect Signals Paul Below paul.below@qsm.com Quantitative Software Management, Inc. Introduction Large development and integration project testing phases can extend over many months

More information

Colorado PUC E-Filings System

Colorado PUC E-Filings System Colorado PUC E-Filings System Attachment A.1 RES Summary Total Acquired Non-Distributed Generation Distributed Generation Retail Distributed Generation Carry Forward Previous Carry Forward Total Carry

More information

Euro-indicators Working Group

Euro-indicators Working Group Euro-indicators Working Group Luxembourg, 9 th & 10 th June 2011 Item 9.4 of the Agenda New developments in EuroMIND estimates Rosa Ruggeri Cannata Doc 309/11 What is EuroMIND? EuroMIND is a Monthly INDicator

More information

ISO Lead Auditor Lean Six Sigma PMP Business Process Improvement Enterprise Risk Management IT Sales Training

ISO Lead Auditor Lean Six Sigma PMP Business Process Improvement Enterprise Risk Management IT Sales Training Training Calendar 2014 Public s (ISO LSS PMP BPI ERM IT Sales Training) (ISO, LSS, PMP, BPI, ERM, IT, Sales Public s) 1 Schedule Registration JANUARY ) FEBRUARY 2 days 26 JAN 27 JAN 3 days 28 JAN 30 JAN

More information

Colorado PUC E-Filings System

Colorado PUC E-Filings System Attachment A.1 RES Summary Colorado PUC E-Filings System Total RECs Acquired Non-Distributed Generation Distributed Generation Retail Distributed Generation Carry Forward Previous Carry Forward Total Carry

More information

ISO Lead Auditor Lean Six Sigma PMP Business Process Improvement Enterprise Risk Management IT Sales Training

ISO Lead Auditor Lean Six Sigma PMP Business Process Improvement Enterprise Risk Management IT Sales Training Training Calendar 2014 Public s (ISO LSS PMP BPI ERM IT Sales Training) (ISO, LSS, PMP, BPI, ERM, IT, Sales Public s) 1 Schedule Registration JANUARY IMS ) FEBRUARY 2 days 26 JAN 27 JAN 3 days 28 JAN 30

More information

STAD68: Machine Learning

STAD68: Machine Learning STAD68: Machine Learning Russ Salakhutdinov Department of Statistics! rsalakhu@utstat.toronto.edu! h0p://www.cs.toronto.edu/~rsalakhu/ Lecture 1 Evalua;on 3 Assignments worth 40%. Midterm worth 20%. Final

More information

Be#er Generaliza,on with Forecasts. Tom Schaul Mark Ring

Be#er Generaliza,on with Forecasts. Tom Schaul Mark Ring Be#er Generaliza,on with Forecasts Tom Schaul Mark Ring Which Representa,ons? Type: Feature- based representa,ons (state = feature vector) Quality 1: Usefulness for linear policies Quality 2: Generaliza,on

More information

NSP Electric - Minnesota Annual Report Peak Demand and Annual Electric Consumption Forecast

NSP Electric - Minnesota Annual Report Peak Demand and Annual Electric Consumption Forecast Page 1 of 5 7610.0320 - Forecast Methodology NSP Electric - Minnesota Annual Report Peak Demand and Annual Electric Consumption Forecast OVERALL METHODOLOGICAL FRAMEWORK Xcel Energy prepared its forecast

More information

Annual Average NYMEX Strip Comparison 7/03/2017

Annual Average NYMEX Strip Comparison 7/03/2017 Annual Average NYMEX Strip Comparison 7/03/2017 To Year to Year Oil Price Deck ($/bbl) change Year change 7/3/2017 6/1/2017 5/1/2017 4/3/2017 3/1/2017 2/1/2017-2.7% 2017 Average -10.4% 47.52 48.84 49.58

More information

Scalable Algorithms for Distribution Search

Scalable Algorithms for Distribution Search Scalable Algorithms for Distribution Search Yasuko Matsubara (Kyoto University) Yasushi Sakurai (NTT Communication Science Labs) Masatoshi Yoshikawa (Kyoto University) 1 Introduction Main intuition and

More information

Distributed Generation. Retail Distributed Generation

Distributed Generation. Retail Distributed Generation Attachment A.1 RES Summary Total RECs Acquired NonDistributed Generation Distributed Generation Retail Distributed Generation Carry Forward to 2017 Previous Carry Forward to 2017 Total Carry Forward to

More information

Lecture Prepared By: Mohammad Kamrul Arefin Lecturer, School of Business, North South University

Lecture Prepared By: Mohammad Kamrul Arefin Lecturer, School of Business, North South University Lecture 15 20 Prepared By: Mohammad Kamrul Arefin Lecturer, School of Business, North South University Modeling for Time Series Forecasting Forecasting is a necessary input to planning, whether in business,

More information

Computing & Telecommunications Services

Computing & Telecommunications Services Computing & Telecommunications Services Monthly Report September 214 CaTS Help Desk (937) 775-4827 1-888-775-4827 25 Library Annex helpdesk@wright.edu www.wright.edu/cats/ Table of Contents HEAT Ticket

More information

Statistical Models for Rainfall with Applications to Index Insura

Statistical Models for Rainfall with Applications to Index Insura Statistical Models for Rainfall with Applications to April 21, 2008 Overview The idea: Insure farmers against the risk of crop failure, like drought, instead of crop failure itself. It reduces moral hazard

More information

Introduction to Course

Introduction to Course .. Introduction to Course Oran Kittithreerapronchai 1 1 Department of Industrial Engineering, Chulalongkorn University Bangkok 10330 THAILAND last updated: September 17, 2016 COMP METH v2.00: intro 1/

More information

Supplementary Questions for HP Chapter Sketch graphs of the following functions: (a) z = y x. (b) z = ye x. (c) z = x 2 y 2

Supplementary Questions for HP Chapter Sketch graphs of the following functions: (a) z = y x. (b) z = ye x. (c) z = x 2 y 2 Supplementary Questions for HP Chapter 17 1. Sketch graphs of the following functions: (a) z = y x (b) z = ye x (c) z = x y. Find the domain and range of the given functions: (a) f(x, y) = 36 9x 4y (b)

More information

Networks. Can (John) Bruce Keck Founda7on Biotechnology Lab Bioinforma7cs Resource

Networks. Can (John) Bruce Keck Founda7on Biotechnology Lab Bioinforma7cs Resource Networks Can (John) Bruce Keck Founda7on Biotechnology Lab Bioinforma7cs Resource Networks in biology Protein-Protein Interaction Network of Yeast Transcriptional regulatory network of E.coli Experimental

More information

BASICS OF COMPRESSION THEORY

BASICS OF COMPRESSION THEORY BASICS OF COMPRESSION THEORY Why Compression? Task: storage and transport of multimedia information. E.g.: non-interlaced HDTV: 0x0x0x = Mb/s!! Solutions: Develop technologies for higher bandwidth Find

More information

Design of a Weather-Normalization Forecasting Model

Design of a Weather-Normalization Forecasting Model Design of a Weather-Normalization Forecasting Model Final Briefing 09 May 2014 Sponsor: Northern Virginia Electric Cooperative Abram Gross Jedidiah Shirey Yafeng Peng OR-699 Agenda Background Problem Statement

More information

Founda'ons of Large- Scale Mul'media Informa'on Management and Retrieval. Lecture #4 Similarity. Edward Chang

Founda'ons of Large- Scale Mul'media Informa'on Management and Retrieval. Lecture #4 Similarity. Edward Chang Founda'ons of Large- Scale Mul'media Informa'on Management and Retrieval Lecture #4 Similarity Edward Y. Chang Edward Chang Foundations of LSMM 1 Edward Chang Foundations of LSMM 2 Similar? Edward Chang

More information

IMAGE COMPRESSION-II. Week IX. 03/6/2003 Image Compression-II 1

IMAGE COMPRESSION-II. Week IX. 03/6/2003 Image Compression-II 1 IMAGE COMPRESSION-II Week IX 3/6/23 Image Compression-II 1 IMAGE COMPRESSION Data redundancy Self-information and Entropy Error-free and lossy compression Huffman coding Predictive coding Transform coding

More information

CAISO Participating Intermittent Resource Program for Wind Generation

CAISO Participating Intermittent Resource Program for Wind Generation CAISO Participating Intermittent Resource Program for Wind Generation Jim Blatchford CAISO Account Manager Agenda CAISO Market Concepts Wind Availability in California How State Supports Intermittent Resources

More information

Vector Quantization and Subband Coding

Vector Quantization and Subband Coding Vector Quantization and Subband Coding 18-796 ultimedia Communications: Coding, Systems, and Networking Prof. Tsuhan Chen tsuhan@ece.cmu.edu Vector Quantization 1 Vector Quantization (VQ) Each image block

More information

UNCLASSIFIED. Environment, Safety and Health (ESH) Report

UNCLASSIFIED. Environment, Safety and Health (ESH) Report Environment, Safety and Health (ESH) Report Perfect Days On a Perfect Day nobody is hurt, we receive no community complaints, we have no security breaches, we do not breach our environmental permit conditions,

More information

Bellman s Curse of Dimensionality

Bellman s Curse of Dimensionality Bellman s Curse of Dimensionality n- dimensional state space Number of states grows exponen

More information

Half-Pel Accurate Motion-Compensated Orthogonal Video Transforms

Half-Pel Accurate Motion-Compensated Orthogonal Video Transforms Flierl and Girod: Half-Pel Accurate Motion-Compensated Orthogonal Video Transforms, IEEE DCC, Mar. 007. Half-Pel Accurate Motion-Compensated Orthogonal Video Transforms Markus Flierl and Bernd Girod Max

More information

Dates and Prices ICAEW - Manchester In Centre Programme Prices

Dates and Prices ICAEW - Manchester In Centre Programme Prices Dates and Prices ICAEW - Manchester - 2019 In Centre Programme Prices Certificate Level GBP ( ) Intensive Accounting 690 Assurance 615 Law 615 Business, Technology and Finance 615 Mangement Information

More information

Applied Time Series Analysis FISH 507. Eric Ward Mark Scheuerell Eli Holmes

Applied Time Series Analysis FISH 507. Eric Ward Mark Scheuerell Eli Holmes Applied Time Series Analysis FISH 507 Eric Ward Mark Scheuerell Eli Holmes Introduc;ons Who are we? Who & why you re here? What are you looking to get from this class? Days and Times Lectures When: Tues

More information

Florida Courts E-Filing Authority Board. Service Desk Report March 2019

Florida Courts E-Filing Authority Board. Service Desk Report March 2019 Florida Courts E-Filing Authority Board Service Desk Report March 219 Customer Service Incidents March 219 Status January 219 February 219 March 219 Incidents Received 3,261 3,51 3,118 Incidents Worked

More information

Salem Economic Outlook

Salem Economic Outlook Salem Economic Outlook November 2012 Tim Duy, PHD Prepared for the Salem City Council November 7, 2012 Roadmap US Economic Update Slow and steady Positives: Housing/monetary policy Negatives: Rest of world/fiscal

More information

Application of Computer in Chemistry SSC Prof. Mohamed Noor Hasan Dr. Hasmerya Maarof Department of Chemistry

Application of Computer in Chemistry SSC Prof. Mohamed Noor Hasan Dr. Hasmerya Maarof Department of Chemistry Application of Computer in Chemistry SSC 3533 Prof. Mohamed Noor Hasan Dr. Hasmerya Maarof Department of Chemistry Outline Fields of applica:on Examples Types of computer Programming languages 2 Introduction

More information

The World Bank Haiti Business Development and Investment Project (P123974)

The World Bank Haiti Business Development and Investment Project (P123974) Public Disclosure Authorized LATIN AMERICA AND CARIBBEAN Haiti Trade & Competitiveness Global Practice IBRD/IDA Specific Investment Loan FY 2013 Seq No: 9 ARCHIVED on 29-Dec-2016 ISR26542 Implementing

More information

MODERN video coding standards, such as H.263, H.264,

MODERN video coding standards, such as H.263, H.264, 146 IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY, VOL. 16, NO. 1, JANUARY 2006 Analysis of Multihypothesis Motion Compensated Prediction (MHMCP) for Robust Visual Communication Wei-Ying

More information

Issues Concerning Dimensionality and Similarity Search

Issues Concerning Dimensionality and Similarity Search Issues Concerning Dimensionality and Similarity Search Jelena Tešić, Sitaram Bhagavathy, and B. S. Manjunath Electrical and Computer Engineering Department University of California, Santa Barbara, CA 936-956

More information

Record date Payment date PID element Non-PID element. 08 Sep Oct p p. 01 Dec Jan p 9.85p

Record date Payment date PID element Non-PID element. 08 Sep Oct p p. 01 Dec Jan p 9.85p 2017/18 Record date Payment date PID element Non-PID element 08 Sep 17 06 Oct 17 9.85p - 9.85p 01 Dec 17 05 Jan 18-9.85p 9.85p 09 Mar 18 06 Apr 18 9.85p - 9.85p Final 22 Jun 18 27 Jul 18 14.65p - 14.65p

More information

by Randall Munroe xkcd: A webcomic of romance, sarcasm, math, and language. (

by Randall Munroe xkcd: A webcomic of romance, sarcasm, math, and language. ( Tensor Decomposi-on xkcd by Randall Munroe xkcd: A webcomic of romance, sarcasm, math, and language. (http://xkcd.com) With notation and figures borrowed from M. Alex O. Vasilescu, TensorTextures: Multilinear

More information

Forecasting the Canadian Dollar Exchange Rate Wissam Saleh & Pablo Navarro

Forecasting the Canadian Dollar Exchange Rate Wissam Saleh & Pablo Navarro Forecasting the Canadian Dollar Exchange Rate Wissam Saleh & Pablo Navarro Research Question: What variables effect the Canadian/US exchange rate? Do energy prices have an effect on the Canadian/US exchange

More information

Winter Season Resource Adequacy Analysis Status Report

Winter Season Resource Adequacy Analysis Status Report Winter Season Resource Adequacy Analysis Status Report Tom Falin Director Resource Adequacy Planning Markets & Reliability Committee October 26, 2017 Winter Risk Winter Season Resource Adequacy and Capacity

More information

Sales Analysis User Manual

Sales Analysis User Manual Sales Analysis User Manual Confidential Information This document contains proprietary and valuable, confidential trade secret information of APPX Software, Inc., Richmond, Virginia Notice of Authorship

More information

A NEW BASIS SELECTION PARADIGM FOR WAVELET PACKET IMAGE CODING

A NEW BASIS SELECTION PARADIGM FOR WAVELET PACKET IMAGE CODING A NEW BASIS SELECTION PARADIGM FOR WAVELET PACKET IMAGE CODING Nasir M. Rajpoot, Roland G. Wilson, François G. Meyer, Ronald R. Coifman Corresponding Author: nasir@dcs.warwick.ac.uk ABSTRACT In this paper,

More information

Sparse Solutions of Systems of Equations and Sparse Modelling of Signals and Images

Sparse Solutions of Systems of Equations and Sparse Modelling of Signals and Images Sparse Solutions of Systems of Equations and Sparse Modelling of Signals and Images Alfredo Nava-Tudela ant@umd.edu John J. Benedetto Department of Mathematics jjb@umd.edu Abstract In this project we are

More information

Monthly Magnetic Bulletin

Monthly Magnetic Bulletin BRITISH GEOLOGICAL SURVEY Ascension Island Observatory Monthly Magnetic Bulletin December 2008 08/12/AS Crown copyright; Ordnance Survey ASCENSION ISLAND OBSERVATORY MAGNETIC DATA 1. Introduction Ascension

More information

Table 01A. End of Period End of Period End of Period Period Average Period Average Period Average

Table 01A. End of Period End of Period End of Period Period Average Period Average Period Average SUMMARY EXCHANGE RATE DATA BANK OF ZAMBIA MID-RATES Table 01A Period K/USD K/GBP K/ZAR End of Period End of Period End of Period Period Average Period Average Period Average Monthly January 6.48 6.46 9.82

More information

Waveform-Based Coding: Outline

Waveform-Based Coding: Outline Waveform-Based Coding: Transform and Predictive Coding Yao Wang Polytechnic University, Brooklyn, NY11201 http://eeweb.poly.edu/~yao Based on: Y. Wang, J. Ostermann, and Y.-Q. Zhang, Video Processing and

More information

CIMA Professional 2018

CIMA Professional 2018 CIMA Professional 2018 Interactive Timetable Version 16.25 Information last updated 06/08/18 Please note: Information and dates in this timetable are subject to change. A better way of learning that s

More information

Operations Management

Operations Management Operations Management Chapter 4 Forecasting PowerPoint presentation to accompany Heizer/Render Principles of Operations Management, 7e Operations Management, 9e 2008 Prentice Hall, Inc. 4 1 Outline Global

More information

Identification of Bursts in a Document Stream

Identification of Bursts in a Document Stream Identification of Bursts in a Document Stream Toshiaki FUJIKI 1, Tomoyuki NANNO 1, Yasuhiro SUZUKI 1 and Manabu OKUMURA 2 1 Interdisciplinary Graduate School of Science and Engineering, Tokyo Institute

More information

GCSE 4241/02 GEOGRAPHY (Specification B) HIGHER TIER UNIT 1

GCSE 4241/02 GEOGRAPHY (Specification B) HIGHER TIER UNIT 1 Surname Centre Number Candidate Number Other Names 0 GCSE 4241/02 GEOGRAPHY (Specification B) HIGHER TIER UNIT 1 S15-4241-02 P.M. TUESDAY, 19 May 2015 1 hour For s use Question Maximum Mark Mark Awarded

More information

High Dimensional Search Min- Hashing Locality Sensi6ve Hashing

High Dimensional Search Min- Hashing Locality Sensi6ve Hashing High Dimensional Search Min- Hashing Locality Sensi6ve Hashing Debapriyo Majumdar Data Mining Fall 2014 Indian Statistical Institute Kolkata September 8 and 11, 2014 High Support Rules vs Correla6on of

More information

Vector Quantization Encoder Decoder Original Form image Minimize distortion Table Channel Image Vectors Look-up (X, X i ) X may be a block of l

Vector Quantization Encoder Decoder Original Form image Minimize distortion Table Channel Image Vectors Look-up (X, X i ) X may be a block of l Vector Quantization Encoder Decoder Original Image Form image Vectors X Minimize distortion k k Table X^ k Channel d(x, X^ Look-up i ) X may be a block of l m image or X=( r, g, b ), or a block of DCT

More information

CIMA Professional

CIMA Professional CIMA Professional 201819 Birmingham Interactive Timetable Version 3.1 Information last updated 12/10/18 Please note: Information and dates in this timetable are subject to change. A better way of learning

More information

CIMA Professional

CIMA Professional CIMA Professional 201819 Manchester Interactive Timetable Version 3.1 Information last updated 12/10/18 Please note: Information and dates in this timetable are subject to change. A better way of learning

More information

EXAMINATIONS OF THE ROYAL STATISTICAL SOCIETY

EXAMINATIONS OF THE ROYAL STATISTICAL SOCIETY EXAMINATIONS OF THE ROYAL STATISTICAL SOCIETY GRADUATE DIPLOMA, 011 MODULE 3 : Stochastic processes and time series Time allowed: Three Hours Candidates should answer FIVE questions. All questions carry

More information