Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

BigML Education - Time Series

233 views

Published on

Learn how to analyze time-ordered historical data to forecast future behavior using BigML's Time Series.

Published in: Data & Analytics
  • Be the first to comment

  • Be the first to like this

BigML Education - Time Series

  1. 1. BigML Education Time Series July 2017
  2. 2. BigML Education Program 2Time Series In This Video • Definition and discussion of time series data • Outlining different ways to model time series data • Exploration of the BigML time series model • Evaluation and comparison of time series models
  3. 3. BigML Education Program 3Time Series Machine Learning Data Color Mass Type red 11 pen green 45 apple red 53 apple yellow 0 pen blue 2 pen green 422 pineapple yellow 555 pineapple blue 7 pen Discovering patterns within data: • Color = “red” Mass < 100 • Type = “pineapple” Color ≠ “blue” • Color = “blue” PPAP = “pen”
  4. 4. BigML Education Program 4Time Series Machine Learning Data Color Mass Type red 53 apple blue 2 pen red 11 pen blue 7 pen green 45 apple yellow 555 pineapple green 422 pineapple yellow 0 pen Patterns valid despite reshuffling • Color = “red” Mass < 100 • Type = “pineapple” Color ≠ “blue” • Color = “blue” PPAP = “pen”
  5. 5. BigML Education Program 5Time Series Time Series Data Year Pineapple Harvest 1986 50,74 1987 22,03 1988 50,69 1989 40,38 1990 29,80 1991 9,90 1992 73,93 1993 22,95 1994 139,09 1995 115,17 1996 193,88 1997 175,31 1998 223,41 1999 295,03 2000 450,53 Pineapple Harvest Tons 0 125 250 375 500 Year 1986 1988 1990 1992 1994 1996 1998 2000 Trend
  6. 6. BigML Education Program 6Time Series Time Series Forecasts Use the data from the past to predict the future
  7. 7. BigML Education Program 7Time Series Time Series Data Year Pineapple Harvest 1986 139,09 1987 175,31 1988 9,91 1989 22,95 1990 450,53 1991 73,93 1992 40,38 1993 22,03 1994 295,03 1995 50,74 1996 29,8 1997 223,41 1998 115,17 1999 193,88 2000 50,69 Pineapple Harvest Tons 0 125 250 375 500 Year 1986 1988 1990 1992 1994 1996 1998 2000 Patterns invalid after shuffling
  8. 8. BigML Education Program 8Time Series Exponential Smoothing
  9. 9. BigML Education Program 9Time Series Exponential Smoothing Weight 0 0,05 0,1 0,15 0,2 Lag 1 3 5 7 9 11 13
  10. 10. BigML Education Program 10Time Series Trendy 0 12,5 25 37,5 50 Time Apr May Jun Jul y 0 50 100 150 200 Time Apr May Jun Jul Additive Multiplicative
  11. 11. BigML Education Program 11Time Series Seasonalityy 0 30 60 90 120 Time 1 4 7 10 13 16 19 y 0 35 70 105 140 Time 1 4 7 10 13 16 19 Additive Multiplicative
  12. 12. BigML Education Program 12Time Series Errory 0 150 300 450 600 Time 1 4 7 10 13 16 19 y 0 125 250 375 500 Time 1 4 7 10 13 16 19 Additive Multiplicative
  13. 13. BigML Education Program 13Time Series Model Types None Additive Multiplicative None A,N,N M,N,N A,N,A M,N,A A,N,M M,N,M Additive A,A,N M,A,N A,A,A M,A,A A,A,M M,A,M Additive + Damped A,Ad,N M,Ad,N A,Ad,A M,Ad,A A,Ad,M M,Ad,M Multiplicative A,M,N M,M,N A,M,A M,M,A A,M,M M,M,M Multiplicative + Damped A,Md,N M,Md,N A,Md,A M,Md,A A,Md,M M,Md,M M,N,A Multiplicative Error No Trend Additive Seasonality
  14. 14. BigML Education Program 14Time Series Linear Splitting Year Pineapple Harvest 1986 139,09 1987 175,31 1988 9,91 1989 22,95 1990 450,53 1991 73,93 1992 40,38 1993 22,03 1994 295,03 1995 115,17 Random Split Year Pineapple Harvest 1986 139,09 1987 175,31 1988 9,91 1989 22,95 1990 450,53 1991 73,93 1992 40,38 1993 22,03 1994 295,03 1995 115,17 Linear Split
  15. 15. BigML Education Program 15Time Series Review • Time series prediction is a distinct type of prediction problem, with data that is assumed to have a certain structure • BigML takes into account a variety of factors when doing time series modeling, including trend and seasonality • You can create a collection of time series models, select the best one, and do forecasting in one click via the BigML dashboard • With linear training/testing splits, you can perform a reasonable out-of-sample evaluation of your time series model

×