What is Machine Learning
Machine learning is an application or sub-branch of Artificial intelligence that gives systems the facility to automatically learn and improve from experience without being explicitly programmed by Programmers. Machine learning focuses on the event of computer programs which may access data and use it learn for themselves.
“ARTHUR SAMUEL” first coined the term machine learning within the Year 1959.
Machine learning applications learn from experience like humans without direct programming. When exposed to new data, these applications will learn, grow, change, and develop by own. In other words, with ML, computers find insightful information without being told where to seem . Instead, they are doing this by leveraging algorithms that learn from data in an iterative process.
At a high levels, Machine Learning is that the power to adapt to new data independently and through iterations. Basically, applications will get from previous computations and transactions and use “pattern recognition” to supply reliable and informed results.
APPLICATIONS OF MACHINE LEARNING
1. Image Recognition
2. Speech Recognition
3. Traffic prediction
4. Product recommendations:
5. Email Spam and Malware Filtering
6. Self-driving cars
7. Virtual Personal Assistant:
8. Stock Market trading
9. Medical Diagnosis
1. ALGORITHM : A Set of rules and statistical techniques used to learn patterns from data. (EX: linear regression,decision tree..)
2.MODEL : A Model is trained by using a machine learning algorithm.
3. PREDICTOR VARIABLE : It is a features of data that can be used to predict the output.
4. TRAINING DATA: The machine learning model is used to built the training data.
5.TESTING DATA: ML Model is evaluated using testing data
STEPS INVOLVED IN ML
1.DATA COLLECTION/GET THE DATA
2.CLEAN, PREPARE AND MANIPULATE DATA
4. TEST DATA
TYPES OF MACHINE LEARNING
In supervised learning, we use known or labeled dataset for the training data. Since the info is understood , the training is, therefore, supervised, i.e., directed into successful execution. The input file goes through the Machine Learning algorithm and is employed to coach the mode
EXAMPLE FOR SUPERVISED LEARNING
Suppose you would like to develop a supervised machine learning model to predict whether a given email is “spam” or “ham.” we’d like to use Emails not marked as “spam” or “not spam” are unlabeled examples.
Because our label consists of the values “spam” and “ham”, any email not yet marked as spam or not spam is an unlabeled example.
In unsupervised learning, the training data is unknown and unlabeled – meaning that nobody has checked out the info before. Without the aspect of known data, the input can’t be guided to the algorithm, which is where the unsupervised term originates from. This data is fed to the Machine Learning algorithm and is employed to coach the model. The trained model tries to look for a pattern and provides the specified response
Reinforcement learning ment learning is a part of machine learning where an agent is put in an environment and he learns to behave in this environment by performing certain actions and observing rewards which it gets from those actions.
VARIOUS TYPES OF PROBLEMS SOLVED USING MACHIME LEARNING MODELS
A regression problem is when the output variable could also be a true or continuous value, like “salary” or “weight”. Many different models are often used, the only is that the rectilinear regression . It tries to suit data with the only hyper-plane which matches through the points.
A classification problem is occurs when the output variable could also be a category, like “red” or “blue” or “disease” and “no disease”. A classification model attempts to find some conclusion results from observed values datasets. Given one or more inputs a classification model will plan to predict the price of 1 or more outcomes.
For example, when filtering emails “spam” or “not spam”, when watching transaction data, “fraudulent”, or “authorized”. In short Classification either predicts categorical class labels or classifies data (construct a model) supported the training set and therefore the values (class labels) in classifying attributes and uses it in classifying new data. There are a number of classification models. Classification models include the following algorithms like logistic regression, decision tree, random forest, gradient-boosted tree, multilayer perceptron, one-vs-rest, and Naive Bayes algorithm.
Cluster analysis, or clustering, is an unsupervised machine learning task that will learn by itself.
It involves automatically discovering natural grouping in data. Unlike supervised learning (like predictive modeling), clustering algorithms only interpret the input file and find natural groups or clusters in feature space
A cluster is usually a neighborhood of density within the feature space where examples from the domain (observations or rows of data) are closer to the cluster than other clusters. The cluster may have a middle (the centroid) that’s a sample or some extent feature space and should have a boundary or extent.