Machine Learning I

Table of contents

No heading

No headings in the article.

What is Machine Learning?

Machine Learning was grown out of the advance work in Artificial Intelligence to help find new capabilities for computers.

According to Arthur Samuel (1959), machine learning is a field of study that gives computers the ability to learn without being explicitly programmed.

Also Tom Mitchell (1998) in his book > Well-posed Learning Problem stated that, a computer program is said to learn from experience E with respect to some task T and some performance measure P, if its performance on T is measured by P, improves with experience E.

Some Advance fields and their examples in machine learning:

  • Database Mining: process of finding patterns, anomalies, correlations within large set of data from growth of automation or web. examples, web click data, medical records etc.

  • Self-customizing programs: programs like amazon, netflix, hulu etc. use your data to predict the type of products or movies you would like and those products are recommended to you.

Machine Learning algorithms

  1. Supervised Learning: In supervised learning, labeled datasets are used to train algorithms to classify and predicts outcome accurately. There are two categories of supervised learning, they are:

  2. Classification problem that either produce discrete values like zero and one.

  3. Regression Problem that produces a continuous value.

Applications of supervised Learning:

  • Weather predictions

  • Fraud detection

  • Algorithmic trading and many more.

  1. Unsupervised Learning: It allows us to approach problems with little or no idea what type of results we are getting. Structures can also be derived from data where we do not necessarily know the effects of the variables.

Categories of unsupervised learning

  • Clustering Algorithm.

  • Cocktail party problem algorithm.

Applications of unsupervised learning:

  • Organize computing clusters.

  • Marketing segmentation

  • Social network analysis

  • Astronomical data analysis.