Supervised and Unsupervised Machine Learning
1 like3,082 views
The document provides an overview of supervised and unsupervised learning in machine learning. Supervised learning involves training a model with labeled data to make predictions, while unsupervised learning allows algorithms to find patterns in unlabeled data. The differences, applications, and various algorithms associated with both learning methods are discussed.
Supervised and Unsupervised Machine Learning
- 2. Spotle.ai Study Material Spotle.ai/Learn What is Supervised learning? You have separate baskets for yellow banana, golden pineapple, black grapes and so on. Now if I give you a golden pineapple you know exactly what it is and in which basket you need to keep it. The labeled fruits help you train your brain about their respective correct baskets. Supervised learning is a learning in which we teach or train the machine using data which are properly or rather correctly labeled.
- 3. Spotle.ai Study Material Spotle.ai/Learn The supervised learning process Let’s say, you want to predict for a group of people their chances of becoming Coronavirus infected. You will need training dataset from previous cases. You will discover new direct relationships via a feedback loop. Your model will evolve and become more and more accurate. Training data Machine learning algorithm Predictive model Model evaluation Feedback loop Labeled
- 4. Spotle.ai Study Material Spotle.ai/Learn I give you 100 items that are not labeled. And then I give you a new item, another puppy, and ask you what it is. Will you be able to tell me? I have not labeled the items for you to understand. You have learnt about them on your own. Unsupervised learning is the learning of machine using information that is neither classified nor labeled and allowing the algorithm to act on that information without guidance. What is Unsupervised learning?
- 5. Spotle.ai Study Material Spotle.ai/Learn The unsupervised learning process There are many animals, snakes, birds and insects that you have never ever seen in your life. But when you see a new bird, that no one has labeled as bird for you to understand, you can still make out that it is a bird because it has feather, it has beak, it can fly etc. This is unsupervised learning. Computationally complex and less accurate. Input data Machine learning algorithm Outputl Unlabeled
- 6. Spotle.ai Study Material Spotle.ai/Learn Supervised learning algorithms When the output variable is categorical or discrete, such as good or bad, yes or no, prone to disease or not prone to disease, the problem that we need to solve is a classification problem. When the output variable is a real value that is continuous, such as age, weight or price, the problem that we need to solve is a regression problem. Please note, logistic regression in not a regression algorithm. Machine Learning Supervised Learning Unsupervised Learning RegressionClassification Decision Tree Discriminant Analysis Naïve Bayes Logistic Regression Support Vector Machine Linear Regression SVR Regression Tree Ensemble Methods GLM
- 7. Spotle.ai Study Material Spotle.ai/Learn Unsupervised learning algorithms We use clustering model when we want to discover the inherent groupings in the data. We use association rule when we want to discover rules that describe large portions of our data. Using association rules we discover relationships between variables in large dataset. Machine Learning Supervised Learning Unsupervised Learning AssociationClustering Hierarchical K-means Hidden Markov Model Gaussian Mixture Fuzzy C-means
- 8. Why do we need supervised learning ❖ Supervised learning allows us to collect data or produce a data output from the previous experience. ❖ It helps us optimize performance criteria using experience. ❖ Supervised machine learning helps us solve various types of real-world computation problems.
- 9. Why do we need unsupervised learning ❖ Unsupervised machine learning finds all kinds of unknown patterns in data. ❖ Unsupervised methods help us find features which can be useful for categorization. ❖ It is takes place in real time, so all the input data to be analyzed and labeled in the presence of the learners. ❖ It is easier to get unlabeled data from a computer than labeled data. Because labeling of data needs manual intervention.
- 10. Spotle.ai Study Material Spotle.ai/Learn Supervised vs Unsupervised learning Supervised Learning Process: input output variables Input data: Labeled data Computational Complexity: Simple Use of Data: Learns from training data Accuracy: Highly accurate Learning: Offline Unsupervised Learning Number of classes: Known Drawback: Classifying big data is a challenge Process: Only input data Input data: Unlabeled data Computational Complexity: Complex Use of Data: Does not learn from output data Accuracy: Less accurate Learning: Realtime Number of classes: Unknown Drawback: No precise data sorting formation