Hanselminutes with Scott Hanselman

Hanselminutes is Fresh Air for Developers. A weekly commute-time podcast that promotes fresh technology and fresh voices. Talk and Tech for Developers, Life-long Learners, and Technologists.



episode 705: Is it the Data or the Algorithm? Common pitfalls in Data Science and Deep Learning with Sara Beck

Sara Beck is the Machine Learning Solution Principal at Slalom Build. She thinks about Data Science and Deep Learning and how diagnosing and anticipating common data science pitfalls can help prevent issues before they happen. She and Scott talk about the importance of identifying whether it’s the algorithm or the data and contextualize the importance of having a good sense of the problem you’re trying to solve.

Slalom Build puts interdisciplinary teams to work in close proximity with clients, to build modern technology and software products for enterprises – faster, cleaner and more nimbly than ever before. Learn more at http://slalombuild.com.

  • Favorite Text Book: https://www.goodreads.com/book/show/9003187-doing-bayesian-data-analysis
  • Favorite Data Science Forecasting Blog (hyndsight is such a perfect name for someone who went in to this area of data science) https://robjhyndman.com/hyndsight/
  • Kaggle is a great resource for practice problems and general data science knowledge sharing. https://www.kaggle.com/
  • Deep learning resource: https://adventuresinmachinelearning.com/wp-content/uploads/2017/07/An-introduction-to-neural-networks-for-beginners.pdf
  • Dan Jurafsky does a nice intro to NLP Youtube series: https://www.youtube.com/watch?v=oWsMIW-5xUc


 2019-10-11  30m