Scribd
Upload
26 views13 pages

002 What Is Prediction

Uploaded by

The Bens Mbr 17
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
26 views13 pages

002 What Is Prediction

Uploaded by

The Bens Mbr 17
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
You are on page 1/ 13
What is prediction? Jeffrey Leek Johns Hopkins Bloomberg School of Public Health The central dogma of prediction Probability/Sampling Training Set ° te, 1% *. oon, . . fy ite oe e Prediction function 10 e What can go wrong ‘16 bara The Parable of Google Flu: Traps in Big Data Analysis Sodan" a eh a ig Anat Ve Fey 201, Goole Fa “rents (G7) mde eins etn arene at Googe Seater etn i Bo ‘Scng tom weld ve pe ‘ower it GFT spe
  • input data -> features -> algorithm -> parameters -> evaluation SPAM Example question -> input data -> features -> algorithm -> parameters -> evaluation Start with a general question Can | automatically detect emails that are SPAM that are not? Make it concrete Can | use quantitative characteristics of the emails to classify them as SPAM/HAM? si SPAM Example question -> input data -> features -> algorithm -> parameters -> evaluation http:/rss.acs.unt.edu/Rdoc/library/kernlab/ntml/spam.html SPAM Example question -> input data -> features -> algorithm -> parameters -> evaluation Dear Jeff, Can you send me your address so I can send you the invitation? Thanks, Ben SPAM Example question -> input data -> features -> algorithm -> parameters -> evaluation Dear Jeff, Can you send me your address so | can send you the invitation? Thanks, Ben Frequency of you = 2/17 = 0.118 SPAM Example question -> input data -> features -> algorithm -> parameters -> evaluation Library (kernlab) data(spam) head spam) make address all num3d our over remove internet order mail receive 1 people report addresses 10.00 0.64 0.64 0.0.32 0.00 0.00 0,00 9.00 0.00 0.00 0.64 9.00 0.00 0.00 20.21 0,280.50 0.0.14 0.28 0.21 «0.07 0.00 0.94 0.21 0.79 0.65 0.21 «0.14 30.06 0,00 0.71 01.23 0419 0.19 0412 0-64 0.25 0.38 0.45 0412 0.00 © 1.75 40.00 0,00 0.00 0 0.63 0.00 0.31 «0.63 0.31 0.63 0.31 0.31 0.31 0.00 0.00 50.00 0.00 0.00 0.0.63 0.00 0.31 «0.63 (0.31 0.63 0.31 0.31 0.31 0.00 9.00 60.00 0.00 0.00 0.1.85 0.00 0.00 1.85 0.00 0.00 0.00 0.00 0.00 0.00 0.00 free business email you credit your font mum000 money hp hpl george nun650 lab labs telnet 10.32 0.00 1.29 1.93 0,00 0.96 0 0.00 0.00 0 0 0 9 0 0 0 20.14 0.07 0.28 3.47 0,001.59 0 0.43 0.43 0 0 0 0 0 0 0 30.06 0,06 1.03 1.36 0.32 0.51 0 1.16 0.06 0 0 0 0 0 0 0 40.31 0,00 0.00 3.18 0.00 0.31 9 0.00 0.00 0 0 o 0 0 0 oMMf SPAM Example question -> input data -> features -> algorithm -> parameters -> evaluation jonspam" ]), col="blue" main="" ,xlab="Frequency of ‘your’ Lines (density (spansyour [spanstype=" spam"]) , 'red") Density 2 L 0 2 4 6 a Frequency of 'your’ ton3 SPAM Example question -> input data -> features -> algorithm -> parameters -> evaluation Our algorithm Find a value C. + frequency of 'your' > C predict "spam" 44n3 SPAM Example question -> input data -> features -> algorithm -> parameters -> evaluation jonspam" ]), col="blue" main="",xlab="Frequency of ‘your’ Lines (density (spansyour [spanstype=" spam"]) , abline(v=0.5,col lack") 'red") Density 2 L Frequency of 'your’ tena SPAM Example question -> input data -> features -> algorithm -> parameters -> evaluation prediction <- ifelse(spamgyour > 0.5, "span", table (prediction, spam$type) /length(spam$type) ‘nonspam" ) prediction nonspam spam nonspam 0.4590 0.1017 spam 01469 0.2923 Accuracy= 0.459 + 0.292 = 0.751 19/13

    • You might also like

    pFad - Phonifier reborn

    Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

    Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


    Alternative Proxies:

    Alternative Proxy

    pFad Proxy

    pFad v3 Proxy

    pFad v4 Proxy