Upgrade to Pro — share decks privately, control downloads, hide ads and more …

Data Science : Theory

Data Science : Theory

Reasoning under uncertainty is the central task of data science.

0f80748bac600948e3d3a233a85ac16c?s=128

Ronojoy Adhikari

April 25, 2015
Tweet

Transcript

  1. Data Science: Theory Ronojoy Adhikari The Institute of Mathematical Sciences

  2. Axiom : your organization will benefit from data

  3. None
  4. Lots of data - where is the science ? Science

    : observation - hypothesis - experiment - theory What are we observing ? What is our hypothesis ? Can we experiment ? Will there be a theory ?
  5. None
  6. None
  7. None
  8. Need

  9. Need Collection

  10. Need Collection Inference

  11. Need Collection Inference

  12. Need Collection Inference Automated Inference ~ Machine Learning

  13. None
  14. Uncertainty - how much will the Nile flood ? -

    when will this equipment fail ? - is this email spam ? - is this applicant a good hire ? Decisions - should we invest in dams ? - should we build redundancy ? - should i delete without reading? - should we look more ? We need to make reasoned decisions in the face of uncertainty Reasoning : Logic - Boolean algebra Uncertainty : Chance, probability Combine : Bayesian probability
  15. 1. Identify the problem 2. Find relevant data sources 3.

    Preprocess the data 4. Apply the algorithm (ML) 5. Visualize the process 6.Tell your story and maintain Six steps to value from data