Data Science Austria

Clustering Using OPTICS

A seemingly parameter-less algorithm See What I Did There? Clustering is a powerful unsupervised knowledge discovery tool used today, which aims to segment your data points into groups of similar features. However, each algorithm is pretty sensitive to the parameters. Similarity based techniques (K-means, etc) are tasked with designating how many … Read moreClustering Using OPTICS

Nimble tweak to use specific python version or virtual environment in RStudio

Reticulate made switch between R & Python easy, and doing its best to facilitate both worlds of data science. Meanwhile, I noticed that most of my followers or students raised the issues of uneasy switch between Here is my nimble tweak, hope, RStudio will come up with a better solution … Read moreNimble tweak to use specific python version or virtual environment in RStudio

Know your enemy: the fascinating implications of adversarial examples

Fascinating properties of adversarial examples Adversarial examples transfer to physical world After reading this far you might be thinking that adversarial examples are just an academic curiosity, or at least limited to digital environments such as evasion of spam filters. It’s easy to believe that the imperceptible distortions that you’ve seen … Read moreKnow your enemy: the fascinating implications of adversarial examples

Spectral graph clustering and optimal number of clusters estimation

This post explains the functioning of the spectral graph clustering algorithm, then it looks at a variant named self tuned graph clustering. This adaptation has the advantage of providing an estimation for the optimal number of clusters and also for the similarity measure between data points. Next, we will provide … Read moreSpectral graph clustering and optimal number of clusters estimation

Using Artificial Intelligence for Diabetic Readmission Prediction

Hospital readmission prediction continues to be a highly encouraged area of investigation mainly because of the readmissions reduction program by the Centers for Medicare and Medicaid services (CMS). The overall goal is to reduce the number of early hospital readmissions by identifying the key risk factors that cause hospital readmissions. … Read moreUsing Artificial Intelligence for Diabetic Readmission Prediction