Data Quality Considerations for Petrophysical Machine Learning Models

It is essential that good quality and appropriately selected data is used when building petrophysical machine learning models, otherwise the predictive capability of the selected algorithm could be impacted. In this week's video I present a similar presentation to the one I gave at the SPWLA 2021 Annual Conference, which covers key data quality issues faced when working with well log data.
If you want, you can check the original paper, and the related journal paper at the links below:
Conference Paper: bit.ly/3shSpRB
The journal Paper, which contains more information on validating ML models: bit.ly/3MCfm9t
If you haven't already, make sure you subscribe to the channel: / @andymcdonald42
-----
RECOMMENDED BOOKS
As an Amazon Associate I earn from qualifying purchases. By buying through any of the links below I will earn commission at no extra cost to you.
PYTHON FOR DATA ANALYSIS: Data Wrangling with Pandas, NumPy, and IPython
UK: amzn.to/3HNycJ9
US: amzn.to/3DL7qPv
FUNDAMENTALS OF PETROPHYSICS
UK: amzn.to/3l1PgSf
PETROPHYSICS: Theory and Practice of Measuring Reservoir Rock and Fluid Transport Properties
UK: amzn.to/30UNWZS
US: amzn.to/3DNqBbd
WELL LOGGING FOR EARTH SCIENTISTS
UK: amzn.to/3FHsbfn
US: amzn.to/3CILAuE
GEOLOGICAL INTERPRETATION OF WELL LOGS
UK: amzn.to/3l2v2HV
US: amzn.to/30UOTkU
-----
PYTHON LIBRARY LINKS
Links for Python Libraries Mentioned in the video:
lasio: • Loading and Exploring ...
dlisio: • DLIS Files - Loading a...
welly: • Getting Started with W...
missingno: • Missingno Python Libra...
pandas: • Python Missing Data Fi...
matplotlib: • Matplotlib Histogram |...
Thanks for watching, if you want to connect you can find me at the links below:
/ andymcdonaldgeo
/ geoandymcd
/ andymcdonaldgeo
www.andymcdonald.scot/
Be sure to sign up for my newsletter to be kept updated when I post and share new content on TH-cam and Medium.
www.getrevue.co/profile/andym...
#petrophysics #python #geoscience

[@cardosogeol]

Very good material, thanks for sharing Andy!

[@tanmaysingh1131]

I 'm pursuing my masters in Geophysics and quite interested in Petrophysical Analysis. I have been following your videos for quite a while and I am hooked to the way you explain things. Hoping to get more informative videos in future.
Thankyou-SIR..!

[@mohammadkeshtkar9655]

Excellent 👍 you are number one Andy . thanks for sharing this video

[@AndyMcDonald42]

Thanks 👍

[@anshsharma2060]

I'm working on well log data and need to know some well log acronyms : RSHA , SWE, ZCOR, VOL_UWAT, HAC, HCAL, HRD, KTH, SPD
Thank you

[@AndyMcDonald42]

Sorry for the late reply. Hope you managed to find them out. If not please let me know.
If you are looking for mnemonic descriptions you can get them from the SPWLA website or the service company themselves like this one
www.apps.slb.com/cmd/

[@aliibrahimali3113]

Hi Andy. I‘m learning lot from you. I am keen to apply supervised ML of facies associations interpretation from wireline logs. Wanted to see whether you are for a chat or advises on how to go about it. Cheers!

[@AndyMcDonald42]

Hi Ali. You may want to check out this video that looks at applying Random Forest to well log data in order to predict facies.
th-cam.com/video/VPQsK86Gl5c/w-d-xo.html

[@texasfossilguy]

Normalization is a bad idea because youre seeing siltiness increase in lower gamma ray curves, hence higher porosity and perm! Conversely it could be carbonate and be making it worse! If you force it to look shaly, you could be screwing that up. You need to get the samples and look at the rock BEFORE tweaking gamma ray responses.