Supervised and unsupervised difference
Question
What is the difference between supervised and unsupervised machine learning?
in progress
0
Machine Learning
4 years
6 Answers
1138 views
Contributor 0
Answers ( 6 )
In supervised machine learning, a model makes predictions or decisions based on past or labeled data. Labeled data refers to sets of data that are given tags or labels, and thus made more meaningful.
In unsupervised learning, we don’t have labeled data. A model can identify patterns, anomalies, and relationships in the input data.
supervised learning is based on labelled dataset and in this we need to predict the output class or a continuous value. examples :- knn, decision tree, logistic regression
Unsupervised learning is a type of machine learning algorithm used to draw inferences from datasets consisting of input data without labeled responses. The most common unsupervised learning method is cluster analysis, in which algos used are kmeans , dbscan , hdbscan etc
When you have a dataset that has been labeled based on your objective classification or regression, and you have been told that if some similar user or sample data comes in the future what will be the output then it is called supervised learning. For example, insurance fraudulent, sales price prediction of a product.
When you don’t know about their historical labeled but you find out similar samples based on som distance matric calculation and assign them a tag that is called unsupervised learning.
In supervised learning, we have label for our target variables. Suppose, we are
trying to solve a credit default problem, we will have a feature in our dataset which
will indicate if a person has defaulted in the past or not. By keeping this feature as our target
variable, we will train our model to predict credit default in the future.
Examples of supervise learning are Linear Regression, Logistic Regression etc.
In Unsupervised learning, we do not have any target variable and our dataset consists of
all independent features. In such learning, the task of the ML algorithm is to learn from the
features of the data and provide insights.
One of the example is the k-means algorithm which is used in clustering. This technique is
widely used in marketing by many companies to find the different types of customers they are catering to.
It helps them in fine tuning their marketing campaigns to cater to different set of customers.
In Unsupervised learning, we do not have any target variable and our dataset consists of
all independent features.
dataset that has been labeled based on your objective classification or regression, and you have been told that if some similar user or sample data comes in the future what will be the output then it is called supervised learning.
1) In Supervised learning, you train the machine using data which is well “labeled.” It means some data is already tagged with the correct answer. It can be compared to learning which takes place in the presence of a supervisor or a teacher. They are categorized as either classification or regression models.
Examples: knn, decision tree, logistic regression, linear regression
Applications: Face detection, Text Classification, Predicting housing prices
2) Unsupervised learning is a machine learning technique, where you allow the model to work on its own to discover information. It mainly deals with the unlabelled data. The idea is to expose the machines to large volumes of varying data and allow it to learn from that data to provide insights that were previously unknown and to identify hidden patterns. As such, there aren’t necessarily defined outcomes from unsupervised learning algorithms. Rather, it determines what is different or interesting from the given dataset. Clustering and Association are two types of Unsupervised learning.
Examples: kmeans , Apriori algorithm
Applications: Anamoly detection – Frauds, Malware detection, Market Basket analysis
There is one major difference between supervised and unsupervised machine learning algorithms is that in supervised model make predictions on labeled data previously known to algorithm For eg classification of emails spam or not spam provided labeled training data of previously classified Spam and not spam emails. And in unsupervised machine learning algorithms model groups similar things together in defined number of groups provided by the user for eg given a dataset of vehicles features model groups them in different groups based on similarities between them.