CS513 KDDM Spring 2020 DataMiners - Learning Machines

Team Members

Dishti Dave

Dyuti Dave

Eman Alofi

Vaishnavi Gopalakrishnan

Types of machine learning algorithms

Supervised Learning

It models relationships and dependancies between the target prediction output and the input features such that we can predict the output values for the new data based on those relationships which it learned from previous data sets.

It is a predictive model, we have labeled data, includes regression adn classification.

List of supervised algorithms

Nearest neighbor

Naive Bayes

Decision trees

Linear regression

Support Vector Machines(SVM)

Neural Networks

Unsupervised Learning

It mainly deals with pattern detection adn descriptive modeling. They are trained with unlabelled data. They use techniques on input data to mine for rules, detect patterns, and summerize and group the data points.

It is a descriptive model, includes clustering and association rule.

List of unsupervised learning

k-means clustering

Association rules

Semi supervised learning

It is a combination of supervised and unsupervised learning. It is the best algorithm when the cost and human efforts are considered. It has unlabeled data but data carries important information about the group parameters

Reinforcement learning

Also known as agent, continuously learns from the environment in an iterative fashion. It will either maximize the rward or minimize the risk. It is a branch of Artificial Intelligence. It automatically tends to determine the behavious to increase the performance. The repitation of decision making is known as Markov Decision Process.

List of algorithms

Q-Learning

Temporal Difference(TD)

Deep Adversarial Networks