You are on page 1of 8

Data Mining(Forecasting)

Muhamad Fathurahman
Fakultas Teknologi Informasi
Universitas YARSI
Outline
• Taxonomy of the Forecasting taks
• Long Short Term Memory Network to perform forecasting task
in python
Forecasting
• Forecasting is the use of time-series data to predict the future
outcome. (E.g sales, currency, weather, population estimate)
Forecasting
• Forecasting can be categorized as a type of supervised learning
Time Dollar to Rupiah X Y
1 16.440 ? 16.440
2 16.280 16.440 16.280
3 16.418 16.418
16.280
4 16.449
16.418 16.449
5 16.644
16.449 16.644
6 16.385
16.644 16.385
7 16.425
16.385 16.425
8 16.369
16.425 16.369
16.369 ?
Forecasting
• Endogenous variables ( e.g, the observation at time t is
dependent upon the observation at t-1; t-1 may depend on t-2,
and so on.
• Single-step vs Multi-step
• Static vs Dynamic
• Classification vs Regression
Forecasting
• Single-step vs Multi-step
Forecasting
• Classification vs Regression
Sport Classification
ECG Classification

Image src: alivercor.com


Forecasting
• Static vs Dynamic model

You might also like