Dave SoteloinTowards Data ScienceExploring Pulsating Variable Stars with Gaussian Process RegressionGaussian processes prove useful for modeling phenomena that aren’t easily described by classic uni- or multi-variate regression. In many…Mar 30, 2018Mar 30, 2018
Dave SoteloinTowards Data ScienceA Visual Comparison of Gaussian Process Regression KernelsA Gaussian process regression is an application of a multivariate Gaussian distribution as a powerful predictive tool for data that is…Mar 2, 2018Mar 2, 2018
Dave SoteloinTowards Data ScienceEffect of Feature Standardization on Linear Support Vector MachinesBecause Support Vector Machine (SVM) optimization occurs by minimizing the decision vector w, the optimal hyperplane is influenced by the…Jul 26, 2017Jul 26, 2017
Dave SoteloinTowards Data ScienceUsing Bagging and Boosting to Improve Classification Tree AccuracyBagging and boosting are two techniques that can be used to improve the accuracy of Classification & Regression Trees (CART). In this post…Jul 9, 2017Jul 9, 2017
Dave SoteloinTowards Data SciencePredicting 90+ Wines with Classification & Regression TreesA Classification & Regression Tree (CART) is a flexible tool for classification problems. It’s popular for its adaptability and ability to…Jul 3, 2017Jul 3, 2017
Dave SoteloClassifying MGS Character Wikis: a Tale of Bayesian PriorsFeeling inspired by my post about Kobe’s Shooting Percentages through the lens of Bayesian Statistics, I decided to dig a little deeper…Jun 27, 2017Jun 27, 2017
Dave SoteloKobe’s Shooting Percentages through the lens of Bayesian StatisticsFor an NBA fan, this may sound familiar, “through the end of November, [Player] is shooting 45% from the paint and 40% from mid-range.”…Jun 21, 2017Jun 21, 2017
Dave SoteloCategorizing a Fast Food Menu using K-Means ClusteringK-Means is an iterative algorithm useful for classifying data that have quantitative characteristics (numbers). Using freely available menu…May 27, 20171May 27, 20171
Dave SoteloUsing Logistic Regression to Understand Speed DatingBinary Logistic Regression is a predictive technique that’s useful when we’re interested in knowing the likelihood of an event occurring…May 13, 2017May 13, 2017
Dave SoteloRelationship between Energy Consumption and Energy PriceThe notion that energy consumption is related to energy price is an intuitive one. All else equal, one would expect higher energy prices to…May 1, 2017May 1, 2017