Question

please don't copy and paste
please no handwriting
USING YOUR OWN WORDS
COURSE Data Mining and Data Warehousing
thanks
Q- Explain in your own words, what is Semi-Supervised Classification?
Why we use Semi-Supervised Learning?

Answer 1

Machine Learning is a concept that allows machines to learn, adapt and understand from the examples and experiences without being explicitly programmed, i.e., instead of writing the program, what we do is feed data to the algorithm/machine, and the algorithm/machine builds the logic based on the given data.

Machine learning examples in day to day life are:-

• Product Recommendation
• Videos Surveillance
• Social Media Services
• Online Customer Support, etc.

Machine learning is classified into sub-categories on the basis of different types of learning;-

Learning Problems

1)Supervised Learning

2)Unsupervised Learning

3)Reinforcement Learning

Hybrid Learning Problems

4)Semi-Supervised Learning

5)Self-Supervised Learning

6)Multi-Instance Learning

Statistical Inference

7)Inductive Learning

8)Deductive Inference

9)Transductive Learning

Learning Techniques

10)Multi-Task Learning

11)Active Learning

12)Online Learning

13)Transfer Learning

14)Ensemble Learning

Various hybrid approaches can be drawn from each field of study incase of surpervised and unsupervised learning. Supervised and Unsupervised learning techniques does not have clear distinction and incase of practical application and implementation both of them are used together to attain more precise and accurate results. As a result in order to improve the resultant accuracy and to solve the problems which involves the use of the concepts of both supervised and unsupervised learning, Semi-Supervised classification was made.

Semi-Supervised learning is a learning approach in which the training data contains very few or little labeled examples and large number of unlabeled examples. Semi-supervised learning lies between unsupervised(with no labeled training data) and supervised(with only labeled training data). The primary objective of a semi-supervised learning model is to effectively and optimally use the available data to considerably improve the accuracy.

The available unlabeled data is used effectively by the use of unsupervised methods such as clustering and density estimation. Once groups or patterns are discovered, supervised methods maybe used to label the unlabeled examples or unlabeled representation that are used for predictions.

Various problem examples that uses Semi-supervised learning are:-

• Many problems from the field of computer vision(image data)
• Natural language processing(text data)
• Automatic speech recognition(audio data), etc.