"Exploring Machine Learning: Applications and Basics for Beginners".
Table of Contents :-
what is machine learning?
Applications of machine learning.
Types of machine learning.
what is train , test and validation?
what is overfitting, underfitting, bias and Variance?
What is machine learning?
- Machine learning is a field of computer science that uses statistical techniques to give computer systems the ability to “learn” with data, without being explicitly programmed.
- It is an subfield of AI that provides stats tools to analyze, visualize, forecasting and prediction.
Applications of machine learning :-
- email spam
- medical diagnosis
- Fraud detection
- Image recognition
- recommendation systems
- Self driving car
- Robots
Types of machine learning :-
- Supervised Machine Learning.
- Unsupervised Machine Learning.
- Semi-Supervised Machine Leaning.
- Reinforcement Machine Learning.
Supervised Machine Learning :-
- In supervised machine learning the dataset is labelled.
- In this you have data which contains input features and output feature and the main task is to understanding the relationship between this input features and output feature and then predicting the output for new input.
- There are two types of problems in supervised machine learning:-
1) Classification
2) Regression
- Classification :-
- when ae are solving the classification problem then the output is categorical.
- example :- suppose we have input feature like no. of study hour, no. of play hour and the no. of sleep hours. based on this features we will predict the output feature pass/fail.
2. Regression :-
- when we are solving the Regression problem then the output is continuous(Numerical).
- example :- predicting the output feature house price using the input features like size of house, no. of rooms.
Unsupervised Machine Learning :-
Unsupervised machine learning is a type of machine learning where the algorithm learns patterns and structures from unlabeled data.
In unsupervised machine learning only input features are given. for example, A data of 5000 students contains IQ and C.G.P.A of students.
Approaches of unsupervised machine learning :-
- Clustering
- Dimensionality Reduction
- Anomaly Detection
- Association Rule Based Learning
Clustering :- Clustering is the process of grouping similar data points together based on their inherent characteristics or similarities.
Dimensionality Reduction :- Dimensionality reduction refers to the techniques and methods used to reduce the number of input variables or features in a dataset while preserving the most important or relevant information.
Anomaly Detection :- Anomaly detection is the process of identifying patterns or instances in data that deviate significantly from the expected or normal behavior.
Association :- Association refers to the discovery of relationships and patterns between items or variables in a dataset.
Semi-supervised Machine Learning :-
Semi-supervised machine learning is a learning technique that lies between supervised learning and unsupervised learning. In supervised learning, we have a labeled dataset where each data point is associated with a corresponding target or label. In unsupervised learning, we work with unlabeled data and aim to discover meaningful patterns or structures within the data.
In the case of semi-supervised learning, we have a combination of labeled and unlabeled data. The availability of both labeled and unlabeled data makes semi-supervised learning useful in scenarios where acquiring labeled data is expensive, time-consuming.
Example :- One famous example of semi-supervised learning is the Google Photos application. When users upload their photos to Google Photos, the application automatically organizes and categorizes the images. This involves tasks such as recognizing objects, people, and locations in the photos.
Reinforcement Machine Learning :-
- Reinforcement machine learning is a type of machine learning where an agent learns to make decisions based on feedback received from its environment.
- example :- Self Driving Car, Robots
Train, Test, Validation :-
Training dataset :- it is used to train the model using training the dataset
Validation dataset :- in case of validation dataset we will specifically perform hyper-parameter tuning
Test dataset :- In this we test the model and see how the model is perform with respect to accuracy.
What is Overfitting, Underfitting, Bias and Variance?
Overfitting :- when the training accuracy is high but testing accuracy is low then it is referred as overfitting. Overfitting indicates low bias and high variance.
Underfitting :- when the training accuracy is low and testing accuracy is also low then it is referred as Underfitting. Underfitting indicates high bias and high variance.
what we want ??
Generalized Model :- Training accuracy is very high and testing accuracy is also very high. it indicates “low bias and low variance”.
Bias & Variance :-
Bias :- In bias we talk about the training accuracy
Low Bias :- if the accuracy is high then we call it as low bias.
High Bias :- if the accuracy is low then we call it as high bias.
Variance :- In variance we talk about the testing accuracy
low variance :- If the accuracy is high then we call it as low variance.
high variance :- If the accuracy is low then we call it as high variance.
Thank you for joining me on this exciting journey into the world of Machine Learning. Together, we’ve explored its basics, applications, and important concepts. I appreciate your participation and hope to continue our data-driven adventures in the future. Let’s keep exploring and unraveling the potential of Machine Learning together!
