Supervised Vs Unsupervised ML

Supervised learning

Supervised learning as the name suggests is the presence of a supervisor who acts as a mentor. In simple words, supervised learning is a way in which we teach or train the machine using data that is well labeled. A well-labeled data is one that has some portion of data that already has the correct answer.

The machine is then provided with a new set of data in a way that a supervised learning algorithm analyses the set of existing data and produces a correct outcome from labeled data.

For instance, suppose you are given a basket filled with different kinds of vegetables. Now the first step is to train the machine with all different vegetables one by one like this:

• If the shape of the object is rounded and depressed at top having the color Red with some green attached then it will be labeled as – Tomato.
• If the shape of the object is a long curving cylinder having the color Light-Green then it will be labeled as –Bottle Gourd.

After training the data, you are given a new separate vegetable say Bottle Gourd, and asked to identify it. Since the machine has already learned the things from previous data and hence used it accordingly. It will first classify the fruit with its shape and color and would confirm the fruit name as BOTTLE GOURD and put it in that same place. Thus the machine learns the things from the existing labeled data of vegetables and then applies the knowledge to the new data entering the system.

Supervised learning can be of two types: Classification and Regression. A classification problem is the one when the output variable is a category, such as “Orange” and “Yellow” or “Vaccine” and “No Vaccine”. Meanwhile, a Regression Problem is the one when the output variable is a real value such as “Rupee” or “Lbs”.

The other types of Supervised Learning are Logistic Regression, K-NN, Decision Tree, and Support Vector Machine.

Supervised and Undervised ML is considered to be one of the frequently asked Machine Learning Interview Questions.

Advantages

• Using previous experience output can be generated for the new data that is being collected
• With the help of sequential learning by gaining experience the performance can be optimized.
• Some real-life problems and challenges can be easily solved using this approach of supervised learning.

Disadvantages

• The segregation of big data can be challenging and might lead to error.
• Teaching the system for supervised learning is a time taking process.

Unsupervised Learning

Unsupervised learning is the training of machines using data points that do not have classified or labeled data. It allows the algorithm to act on its own without giving any specific direction. The machine is expected to group unorganized and unsorted data based on certain common traits, similarities, patterns, and differences without providing any teaching on the data set. It allows the machine to design the use case on its own and thus find hidden connections between data on its own.

For instance, an image of a tiger and a lion is given to the machine. The machine has no idea what is a tiger or a lion. about the features, patterns, color, similarities, or differences of lion and tiger. Hence, it will not be able to categorize it in tiger and lion. But it can categorize them according to the features, patterns, color, similarities, or differences of lion and tiger. In this system we were not taught anything before i.e no data was shown or given to us at the beginning of the process.

It allows the technique to work on its own by discovering patterns and information that was not known to the system before. It mainly deals with unlabelled data.

Unsupervised learning can be of two types:

• Clustering
• Association

The clustering problem is the one where we check for inherent grouping of data such as grouping cardholders based on their spending.
Whereas Association is the one where we discover a set of rules. That describes large portions of the data and brings an X on Y dependency.

The other types of Unsupervised Learning are Agglomerative, Overlapping, and Probabilistic. There are some clustering types like K-map and others that help in the same.

Supervised Vs Unsupervised

Conclusion

Machine Learning has come a long way and advanced in such a manner that is always on the move. Supervised and Unsupervised Learning are two realms of this world that differ in their approaches and complexity making each other complete.