Machine Learning

 

Machine Learning (ML) use statistical tool to explore data and is making the computer (MACHINE) learn from studying data. Machine Learning is a program that analyses data and learns to predict the results According to Tom M. Mitchell, ML is    a computer program is said to learn from experience E with respect to some class of tasks T and performance measure P, if its performance at tasks in T, as measured by P, improves with experience E. 

    According to Arthur Samuel, Machine learning enables a machine to automatically learn from data,  improve performance from experiences,   and predict things without being explicitly programmed. Machine Learning is a subset of artificial intelligence  that is mainly concerned with the development of algorithms which allow a computer to learn from the data and past experiences on their own.    

Machine learning uses various algorithms for building mathematical models and making predictions using historical data. Machine learning can be classified into three types:

1. Supervised learning
2. Unsupervised learning
3. Reinforcement learning

1. Supervised Learning

Goal of Supervised Learning is to map input data with the output data and to Predict class Label Supervised learning provide sample labeled data to the machine learning system in order to train it, and on that basis, it predicts the output and The system creates a model using labeled data learn about each data, If the training and processing are Completed then we test the model by providing a sample data to check whether it is predicting the exact output or not. The example of supervised learning is spam filtering. Supervised learning can be grouped further in two categories of algorithms:

•       Classification

•       Prediction

Classification is a supervised learning i.e. we can predict input and out values and classification is divided into groups but not necessarily similar properties is called Classification. It predicts categorical class labels (discrete or nominal). Classification predicts categorical labels (classes). It classifies data (constructs a model) based on the training set and the values (class labels) in a classifying attribute and uses it in classifying new data. Classification and prediction are two forms of data analysis that can be used to extract models describing important data classes or to predict future data trends.

Regression algorithms are

•       Linear Regression

•       Polynomial Regression

•       Exponential Regression

•       Logistic Regression

•       Logarithmic Regression

Classification algorithms are

•       Decision Trees

•       Random Forest

•       Naive Bayes

•       Support Vector Machine

•       Rule-based classification

•       K-Nearest Neighbors

2 Unsupervised Learning

Goal of Unsupervised Learning (UL) is to  a group of objects with similar patterns and determine data patterns/groupings. Machine learns without any supervision is called UL. The training is provided to the machine with no labeled data, classified, or categorized.  It can be further classifieds into two categories of algorithms:

•       Clustering

•       Association

Clustering is a Unsupervised learning i.e. no predefined classes, Group of similar objects that differ significantly from other objects. The process of grouping a set of physical or abstract objects into classes of similar objects is called clustering. Clustering is “the process of organizing objects into groups whose members are similar in some way”. The cluster property is Intra-cluster distances are minimized and Inter-cluster distances are maximized. Cluster is a collection of data objects and Similar to one another within the same cluster, Dissimilar to the objects in other clusters

Clustering algorithms are

1. Partitioning Methods (k-means, k-medoids)
2. Hierarchical Methods (DIvisive ANAlysis, AGglomerative NESting)
3. Density-Based Methods (DBSCAN (Density-Based Spatial Clustering of Applications with Noise))
4. Grid-Based Method(STING (STatistical INformation Grid))

Association algorithms:

•       Market Basket Analysis

•       Apriori Algorithm

•       FP Growth Algorithm

•       Vertical data format 

3 Reinforcement learning

Reinforcement learning is a feedback-based learning method, in which a learning agent gets a reward for each right action and gets a penalty for each wrong action.  The agent learns automatically with these feedbacks and improves its performance.  In reinforcement learning, the agent interacts with the environment and explores it. The goal of an agent is to get the most reward points, and hence, it improves its performance.

References:

Jiawei Han, Micheline Kamber and Jian Pei, Data Mining: Concepts and Techniques, Elsevier, 3rd edition, 2013.

•      https://en.wikipedia.org/wiki/Weka_(machine_learning)

•       https://www.javatpoint.com/machine-learning