What Is Machine Learning?
We know that humans learn from their past experiences and machines follow instructions given by humans. But, what if humans can train the machines to learn from the past? Well, that’s called machine learning. But, it’s a lot more than just learning. It’s also about understanding and reasoning.
BASICS OF MACHINE LEARNING
There are basically three ways in which a machine can learn:
1. SUPERVISED LEARNING
Suppose your friend gives you 1 million coins of three different currencies and each coin has different weights. For example, a coin of one rupee weighs three grams, one euro weighs seven grams and one dirham weighs four grams.
When you feed this data to the machine learning model, it will learn that if a coin is of three grams, it will be a one rupee coin. Hence, supervised learning uses labels(currency) and features(weight) of data to train the model.
Supervised learning is a learning in which we teach or train the machine using data which is well labeled after which the machine is provided with a new set of examples(data) so that supervised learning algorithm analyses the training data(set of training examples) and produces a correct outcome from labeled data.
There are two important categories:
Regression problems: It is a predictive analysis used to predict a continuous variable. Continuous variable is a variable that has infinite number of possibilities. For eg. predicting the prices of a stock.
Classification problems: It is used to predict a label or a class. Let’s say we have to classify our mails as spam or non-spam. Here, to assign our input to different classes, we use classification algorithms.
2. UNSUPERVISED LEARNING
Suppose we have cricket data set of various players with their respective scores and wickets taken. When we feed this data set to the machine, the machine identifies the pattern of player performance. While looking at the data, there will be two clusters, one cluster are the players who scored high runs and took less wickets i.e. batsman while the other cluster is of the players who scored less runs but took many wickets i.e. bowlers. There were no labels of batsmen and bowlers.
The learning with unlabeled data is unsupervised learning. The task of machine is to group unsorted information according to similarities, patterns and differences without any prior training of data.
It mainly has two types:
Association problems: These involve discovering patterns in data, find co-occurrences and so on. For example, people who buy bread also tend to buy jam. It’s all about finding associations between items that frequently co-occur or items that are similar to each other.
Clustering problems: The method of identifying similar groups of data in a dataset is called clustering.
3. REINFORCEMENT LEARNING
Reinforcement learning is a reward based learning. It on the principle of feedback.
Input is given to a machine learning model which then gives the output according to the algorithm applied. If it’s right, we take the output as a final result, else we provide feedback to the trained model and ask it to predict until it learns.
Reinforcement Learning(RL) is a type of machine learning technique that enables an agent to learn in an interactive environment by trial and error using feedback from its own actions and experiences.
For example, in robotics, we might start in a situation where the robot doesn’t know anything about the surroundings. So, after performing actions, it finds out more about the world based on whether it decides to move left, right, backward or forward. here, the robot is the agent and the surrounding is the environment. For each action it takes, it might receive a reward or a punishment.
APPLICATIONS OF ML
- Facebook recognizes our friend in a picture from an album of tagged photographs.
- Netflix recommends new movies based on someone’s past movie choices.
- Analyzing bank data for suspicious transactions and flagging fraud transactions.
- Used in healthcare where diagnostics are predicted for doctors.
- Review the sentiment analysis that the tech giants are doing on social media.
- Fraud detection in the finance sector.