Get More From Your Data with Data Analytics

Transcription

1 Get More From Your Data with Data Analytics Francesca Perino 2015 The MathWorks, Inc. 1

2 2

3 3

4 4

5 Buildings have thermodynamic properties u t α 2 u x u y u z 2 = 0 5

6 Temperatures change 6

7 Electricity demand varies 7

8 Humans have comfort bounds 8

9 u t α 2 u x u y u z 2 = 0 9

10 10

11 Traits of Data Analytics applications 1. Diverse and/or Big Data 2. Advanced Algorithms 3. Deployment 11

12 Why MATLAB? 3 Develop embedded systems with analytics powered functionality 1 DATA Engineering, Scientific, and Field Business and Transactional Analytics that increasingly require both business and engineering data 4 Develop analytics to run on both enterprise and embedded platforms Embedded Systems 2 Enable Domain Experts to be Data Scientists Business Systems Data Scientist 12

13 Why MATLAB? 3 Develop embedded systems with analytics powered functionality 1 DATA Engineering, Scientific, and Field Business and Transactional Analytics that increasingly require both business and engineering data 4 Develop analytics to run on both enterprise and embedded platforms Smarter Embedded Systems 2 Enable Domain Experts to be Data Scientists Business Systems Data Scientist 13

14 Business and Engineering Data Business and Transactional Data Engineering, Scientific, and Field Data Repositories Databases (SQL) NoSQL Hadoop File I/O Text Spreadsheet XML Web Sources HTML Mapping Financial datafeeds RESTful JSON File I/O Text Spreadsheet XML CDF/HDF Image Audio Video Geospatial Communication Protocols CAN (Controller Area Network) DDS (Data Distribution Service) OPC (OLE for Process Control) XCP (explicit Control Protocol) Real-Time Sources Sensors GPS Instrumentation Cameras Communication systems Machines (embedded systems) No matter what industry our client is in, and no matter what data they ask us to analyze text, audio, images, or video MATLAB enables us to provide clear results faster. Dr. G Subrahmanya VRK Rao, Cognizant 14

15 Accessing Data 15

16 Datastore HDFS Node Data Map Reduce Node Data Map Reduce Node Data Map Reduce 16

17 Table 17

18 Preprocessing Data 18

19 Signal Processing cheby2 filter rms pwelch periodogram xcov findpeaks movmean movstd 19

20 Image Processing Perform image processing, analysis, visualization, and algorithm development Image display and exploration Image enhancement Image analysis Morphological operations Image registration Geometric transformation ROI-based processing 20

21 Feature Engineering Extracting Information from Data Data type Common Techniques for Deriving Features Sensor data Peak analysis Pulse and transition metrics Spectral measurements (power, bandwidth, mean frequency, median frequency) Image and video data Bag of visual words HOG (Histogram of Oriented Gradients) Minimum Eigenvalue algorithm Local feature descriptors Edge detection Transactional data Decomposing timestamps into components (day, month, day of week, etc.) Calculation of aggregate values 21

22 Why MATLAB? 3 Develop embedded systems with analytics powered functionality 1 DATA Engineering, Scientific, and Field Business and Transactional Analytics that increasingly require both business and engineering data 4 Develop analytics to run on both enterprise and embedded platforms Smarter Embedded Systems 2 Enable Domain Experts to be Data Scientists Business Systems Data Scientist 22

23 Enabling Domain Experts to be Data Scientists Machine Learning Statistics Image Processing Apps Language Neural Networks Signal Processing Optimization Control Systems Financial Modeling Symbolic Computing MATLAB has helped accelerate our R&D and deployment with its robust numerical algorithms, extensive visualization and analytics tools, reliable optimization routines, support for object-oriented programming, and ability to run in the cloud with our production Java applications. Borislav Savkovic, BuildingIQ 23

24 Apps - Classification Learner app 24

25 Language - MATLAB Execution Engine Redesigned execution engine runs MATLAB code faster All MATLAB code can be JIT compiled A platform for future improvements The examples of all FSDA functions have been monitored under R2012a, R2014b, R2015a, R2015b. Results seem to indicate that, generally, on computationally intensive mathematical/statistical function R2015b yields remarkable time improvements. 25

26 Why MATLAB? 3 Develop embedded systems with analytics powered functionality 1 DATA Engineering, Scientific, and Field Business and Transactional Analytics that increasingly require both business and engineering data 4 Develop analytics to run on both enterprise and embedded platforms Smarter Embedded Systems 2 Enable Domain Experts to be Data Scientists Business Systems Data Scientist 26

27 Smarter Embedded Systems RESEARCH REQUIREMENT S DESIGN C, C++ Environment Models Physical Components Algorithms IMPLEMENTATION VHDL, Verilog Structured Text TEST AND VERIFICATION Airbus Battery management Sonova Hearing implants GM Climate control Weinmann Transport ventilator Festo Industrial robots ABB Smart Grid controller MCU DSP FPGA ASIC PLC INTEGRATION manroland Printing presses FLIR Thermal imaging Daimler Cruise controller 27

28 MATLAB Code Generation 28

29 C Code Generation 29

30 Why MATLAB? 3 Develop embedded systems with analytics powered functionality 1 DATA Engineering, Scientific, and Field Business and Transactional Analytics that increasingly require both business and engineering data 4 Develop analytics to run on both enterprise and embedded platforms Smarter Embedded Systems 2 Enable Domain Experts to be Data Scientists Business Systems Data Scientist 30

31 Where does the processing happen? Devices Business Systems Visibility Self, Maybe Neighbors Entire System Power Battery (Low) Grid (Unlimited) Processing Microprocessor (Low) CPU s (Unlimited) Storage Memory (Minimum Buffer) HD s (Unlimited) Updates Difficult Easy 31

32 Why not transfer all the data? 1. Data privacy concerns 2. Cost of network/storage 3. Power required to transmit data from device (for wireless) 4. Response time 32

33 Splitting computation 33

34 Smart Systems: Wearable Healthcare Technology 34

35 Deploying Algorithms to Enterprise Systems MATLAB MATLAB Compiler MATLAB Compiler SDK Standalone Application Excel Add-in C/C ++.NET MATLAB Production Server Hadoop Python Java MATLAB Compiler enables sharing MATLAB programs without integration programming MATLAB Compiler SDK provides implementation and platform flexibility for software developers MATLAB Production Server provides the most efficient development path for secure and scalable web and enterprise applications 35

36 Enterprise Integration Forecasting Model 36

37 MATLAB Differentiators 3 Develop embedded systems with analytics powered functionality 1 DATA Engineering, Scientific, and Field Business and Transactional Analytics that increasingly require both business and engineering data 4 Develop analytics to run on both enterprise and embedded platforms Smarter Embedded Systems 2 Enable Domain Experts to be Data Scientists Business Systems Data Scientist 37

38 Learn More 38