Machine Learning

All About Machine Learning from Zero to Infinity…

Machine learning means Machine that learns Automatically.

Machine Learning is Branch of Artificial Intelligence which deals with Every kind of Data as well as it provides systems the ability to automatically learn and improve from experience without being explicitly programmed.

In simply word to understand the terminology Machine Learning we can say that Machine learning means Machine that learns Automatically and return useful information about data or prediction, classification on the basis of Given Data.

Humans who can learn everything from their experiences with their learning capability, and we have computers or machines which work on our instructions as well as machine learning we trained a machine in this way machine also learn from experiences or past data like a human does? and give us result on the basis of data or past data.

With the help of sample historical data, which is known as training data, machine learning algorithms build a mathematical model that helps in making predictions or decisions without being explicitly programmed. Machine learning brings computer science and statistics together for creating predictive models. Machine learning constructs or uses the algorithms that learn from historical data. The more we will provide the information, the higher will be the performance.

We read everywhere ML is part of AI but why ML is part of AI ?… Because ML trained a Machine that how to learn as well as ML models look for data and give us a conclusion.

Difference in AI and Machine Learning :

While artificial intelligence (AI) is the broad science of mimicking human abilities, machine learning is a specific subset of AI that trains a machine how to learn.

What is Data ?

Data is a collection of facts, such as numbers, words, measurements, observations or just descriptions of things, images, video, emails, etc. those things that we can store is data or we can say everything on Internet is data.

Type of Machine Learning :

1. Supervised Machine Learning
2. Unsupervised Machine Learning
3. Semi Supervised Machine Learning
4. Reinforcement Learning

Supervised Machine Learning :

In Supervised Machine Learning the machine learning algorithm is trained on labelled data. data needs to be labelled accurately for this method to work, supervised learning is extremely powerful when used in the right circumstances.

Supervised learning can used to solve two types of problems:

1. Regression
2. Classification.

Supervised Machine Learning have data as input and where desire output is known or we can say that Machine Learning use Historical data and on the basis of historical data predict future Data or on the basis of previous create new data and model will check the accuracy with the new data if new data is not accurate then we do some changes or modified in model for find accuracy in data or for a correct result.

Regression :

Regression or regression problem is use for Continuous Values or we can say predict Continuous Values. For example, if a company’s sales have increased steadily every month for the past few years, by conducting a linear analysis on the sales data with monthly sales, the company could forecast sales in future months. Another Example predict the cost of a house or the weather outside in degrees.

Classification :

Classification is use to predict the categorical values for example (categorical values = ([He,she],[Male,Female],[Good,Better,Best]) etc. and in Classification we convert this categorical value to binary value because machine understand binary values.

classification problem because you’re trying to classify the answer into two specific categories: yes or no ([He,she],[Male,Female],[Good,Better,Best]) . This is also called a, binary classification problem.

Steps for Machine Learning :

1. Data Collection
2. Data Cleaning or Data preparing
3. EDA (Exploratory Data Analysis)
4. Split Data for train and test
5. Choosing a model or Build a model
6. Evaluation of Model on the basis of Test data
7. Hyper parameter Tuning
8. Prediction On real Data
9. Deploy the Model

Supervised Machine-learning

Example -Suppose you have data set of two companies let’s assume Company A and Company B, so in both of this Company they have total amount of expense for advertisement of there product but the difference is, Company B Data is properly in labelled form or we can say Data are properly manage in Excel according to expense date and month and amount of expense everything so via supervised ML algorithm we can trained our model in that way it will give us a prediction with accuracy that in which business segment how much amount we have to spend or we don’t have to waste money in which area but Company A Data is not in label form or not store properly.

Type of split Data in Machine Learning :

1. Training Data [Training data is use to train model parameter]
2. Validation Data[which hyper parameter we have to use for data validation]
3. Test Data [finally with this get metric of result]

Unsupervised Machine Learning :

In Unsupervised Machine Learning data are not labelled, classified, or categorized, or machine learns without any supervision. The main goal of unsupervised learning is to restructure the input data into new features or a group of objects or find out new pattern with similarities.

Unsupervised Machine Learning is divided in 2 parts :

• Clustering
• Association

Semi Supervised Machine Learning:

Semi-supervised learning is a bit of both supervised and unsupervised learning and uses both labelled and unlabelled data for training. the algorithm would use a small amount of labelled data with a large amount of unlabelled data.

Semi-Supervised Machine Learning use for :

1. Classification
2. Regression
3. Prediction.

Semi-supervised learning is supervised learning where the training data contains very few labelled examples and a large number of unlabelled examples. The goal of a semi-supervised learning model is to make effective use of all of the available data, not just the labelled data like in supervised learning. Making effective use of unlabelled data may require the use of or inspiration from unsupervised methods such as clustering and density estimation. Once groups or patterns are discovered, supervised methods or ideas from supervised learning may be used to label the unlabelled examples or apply labels to unlabelled representations later used for prediction. It is common for many real-world supervised learning problems to be examples of semi-supervised learning problems given the expense or computational cost for labelling examples. For example, classifying photographs requires a datasets of photographs that have already been labelled by human operators. Many problems from the fields of computer vision (image data), natural language processing (text data), and automatic speech recognition (audio data) fall into this category and cannot be easily addressed using standard supervised learning methods.

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.

The robotic dog, which automatically learns the movement of his arms, is an example of Reinforcement learning.