Projects | Rasik Kane

Projects

Dublin Bus Journey time Prediction

IoT Cattle Monitoring System

Daily analysis of dairy animal health using On-Body wireless activity tracker and on-Campus image processing system

Mortality prediction at Texas animal Shelter

solves a prediction problem: Whether enrolled animal in Texas Animal Shelter shall live or not.

Comparision of Feature selection techniques

Performance analysis of Sequential Backward floating selection mlxtend and Recurssive feature eliminator with cross validation by scikit-learn

Clustering - kmeans and DBscan

k-means clustering is a partition based on centroids. DBSCAN clustering, in contrast does not define a central tendency which defines range of cluster. DBSCAN clusters areas of high density separated by areas of low density. Even in abscense of ground truth; Quantitive performance evaluation for unsupervised learning algorithms can be done using Silhoutte score and calinski harabasz score

Gender detection using nltk's Naive Bayes classifier

Check the link to know how Naive Bays classifier mitigates a potential challenge with Bays classifier that is – predicating conditional probability of all feature vector tuples given a label. Instead, Naïve Bayes Classifier considers all features conditionally independent.

Unified IoT Ecosystem

TADHACK’17– India winner