Learning goals

Post-process datasets for enhanced information extraction

Take informed decisions by applying exploratory data analysis tecnhiques to dataset

Create a statistically significant supervised machine learning model Optimize an industrial operation based on a (surrogate) supervised machine learning 

Project statement - updated

Dataset Part 1

Dataset Part 2

Glossary
Temp = temperature in ºC
Total_flow = total inlet flowrate in m3/h
Ar_flow = argon inlet flowrate in m3/h
CH4_flow = methane inlet flowrate in m3/h
O2_flow = oxygen inlet flowrate in m3/h
CH4/O2 = inlet molar ratio of methane to oxygen
CT = contact time in h
CH4_conv = methane conversion
C2s = selectivity towards C2 compounds
C2H6s = selectivity towards ethane
C2H4s = selectivity towards ethylene

Teams 

Same as before.

Guidelines for presentation

DosDont's
1 question = 1 slide; make exceptions (2 slides) for most complex answersShow code or discuss Python details
Goal, strategy, key data & conclusion for every questionIntro longer than 15 s - go straight to the questions (we all know each other, we all know why are you presenting)
A conclusion requires a justification based on either evidence (data/statistics) or hypothesis/theory Index slide 
Theoretical arguments require at the least a biblio referenceVarious potential hypothesis to explain data - make that clear by what is said (when you're are relatively confident, but more explanations are possible: "the most likely explanation is"; when various explanations are equally likely: "Among various possible explanations, one is")

Evaluation

The final individual grade (iFG) will be computed from the team average presentation grade (tPG) and individual discussion grade (iDG) 


iFG= 2/3 x tPG + 1/3 x iDG 

Delivery

Presentation doc (pptx, PDF) by March 19th at 18h via Fénix.
Presentation + discussion (8 min + 6 min) - everyone presents and discusses

In view of limited class duration, presentations will be stopped at 8 min sharp.

Grades