BEGIN:VCALENDAR VERSION:2.0 PRODID:-//hacksw/handcal//NONSGML v1.0//EN METHOD:PUBLISH BEGIN:VEVENT DTSTAMP:20240328T224354Z DESCRIPTION:Click for Latest Location Information: http://edw2018.dataversi ty.net/sessionPop.cfm?confid=121&proposalid=9764\n
NOTE: This seminar continues from Thursday afternoon.
\nNo organization should hesitate to deploy Artificial Intellige nce (AI) and Machine Learning (ML) throughout their company and products. O n the contrary, a failure to do so will certainly imply obsolescence. In fa ct, today’s current open-source software environment makes it easier than ever to quickly prototype products and experiments to easily see and i terate on the potential productivity gains and benefits that ML can have on your company’s work and impact, whether internal to the organization 's efficiency or external through your company’s core product val ue.
\nThis two-part workshop will bring participants up to speed on t he complete Data Science spectrum: from data munging and cleaning to data e xploration and visualizations, to building machine learning models and pred ictive analytics, with a focus on ML models and their applications.
\nWithin the Jupyter notebook universe, using both Python and R programming languages and packages, we will build and explore working code on varying d atasets that cover a breadth of topics and contexts, motivating the introdu ction of clustering techniques, classification problems, and regression: th e overarching classes of problems into which all data problems fit. We will go over the basic ideas of the various techniques as well as the motivatio ns for their use: when to use them and why.
\nWe will utilize K-Neare st Neighbors algorithms (KNN) and Principle Components Analysis (PCA) for d imensionality reduction and clustering, Logistic Regression and Random Fore sts for classification, and Neural Networks and extreme Gradient Boosted De cision Trees (xgBoost) for regression and predictive modeling. We will also briefly extend the models to include modern Deep Learning motivations and uses.
\nParticipants will leave with a solid cursory understanding of the types of problems that Artificial Intelligence and Machine Learning ca n tackle, as well as the practical skills and know-how to bring the approac hes back to their respective industries and problems. Each participant will leave with meaningful, working code. More importantly, however, each parti cipant will leave with the mindset and optimism of using machine learning t o approach and transform any data problem across their individual industrie s and settings.
\n