Machine-Learning-Dictionary

This file contains curated lists of machine learning terms which the ML enthusiasts find difficult in understanding. The terms are

1. Supervised Learning

It is something like,

You get a bunch of photos with information about what is on them and then you train a model to recognize new photos.
You have a bunch of molecules and information about which are drugs and you train a model to answer whether a new molecule is also a drug.

2. Unsupervised learning

You have a bunch of photos of 6 people but without information about who is on which one and you want to divide this dataset into 6 piles, each with the photos of one individual. You have molecules, part of them are drugs and part are not but you do not know which are which and you want the algorithm to discover the drugs.

3. Reinforcement Learning

is a branch of Machine Learning where the algorithm performs a certain action on its own, evaluates the result of the action, and then decides what to do next, keeping the end goal in mind. Here, we just define a set of rules and leave the decision making to the algorithms, hence, making them "autonomous".

A.) let's take a real-life scenario- When you try to warn a kid to not touch fire because it will hurt, you are "supervising" the learning of the kid. This is supervised machine learning.

B.)When the kid tries to touch the fire anyway and jumps back due to the burning sensation. The kid(agent) is interacting with the fire(environment) and learns that it is not good to touch fire. This is unsuperised learning as the kid himself is learning the action.

C.)Hence, the kid is not going to touch the fire again due to the pain(penalty) experienced and decides the next action "stay away from fire" in order to be safe and happy(reward). This is Reinforcement Machine Learning.

EXAMPLES: Self driving cars, in NLP (Natural Language Processing), news recommendation, Amazon /flipcart product recommendations,

4. Regression

Machine is trained to predict some quantitites like price, weight or height.

predicting house/property price, forecasting stocks, forecasting weather, predicting stock market price etc.

5. Clustering

Clustering can be understood with the help of clusters. So a CLUSTER is a group of objects which are similar in some ways. It is used to find a pattern in Machine Learning. It is a part of unsupervised learning so deals with the unlabelled dataset. In layman it is method of grouping objects (similar in some way) called clusters. It does it by finding some similar patterns in the unlabelled dataset such as shape, size, color, behavior, etc., and divides them as per the presence and absence of those similar patterns.

EXAMPLE: Netflix recommendation systems- With the help of clustering, Netflix data scientists find people who like the series "Lost", "Black Mirror" and "Groundhog Day". Then it is used to refine its knowledge of the tastes of viewers and thus make better decisions in the creation of new original series.

News Recommedation system- Like news is shown in different categroies. Like your searched sports news on google, so clustering method will make cluster /group of sports news. Then it is shown in the search results

6. Classification

Machine is trained to classify something into some class. So in layman classification includes those problems which has a certain answer like YES OR NO. EXAMPLE: classifying whether a patient has disease or not, classifying whether an email is spam or not, classifying whether your pass or fail in your exams. It is a part of supervised learning.

CLUSTERING vs CLASSIFICATION SOURCE: https://blog.bismart.com/en/classification-vs.-clustering-a-practical-explanation

7. Support Vector Machine

8. Baysian Network

9. Neural Networks

Neural Networks or Artificial Neural Networks(ANN) are information processing model inspired by the biological nervous system. Neural Networks are formed by highly interconnected processing elements called neurons. Interconnections between the neurons are provided by the weights. ANN like people, learn by example. Neural Networks could be used to extract patterns and detect trends that are too complex to be noticed by either humans or other computer techniques. A simple neural network is shown in figure. Inputs x1, x2, ...xn are fed to the input unit, they are connected to the output units through weights w1,w2, ..wn. Net input is calculated as sum of product of inputs and their weights yin = x1w1 + x2w2 +...xnwn. Activation function applied over net input will produce the output . More layers are added between input layer and output layer to increase accuracy and they are called hidden layers.

sruthy-j-mallya / The-Layman-Machine-Learning-dictionary