Page 195 - AI Ver 3.0 class 10_Flipbook
P. 195
onion
ginger
For example, if we want to train a model to identify if an image is of an onion or a ginger, we need to train it with
multiple images of both onion and ginger along with their labels. The machine will then classify images on the
basis of the labels and predict the correct label for testing data. Classification works on discrete dataset. Following
are some examples of classification:
• Image Classification: In this scenario the model would be trained on different categories of images. The model
would learn the pattern as fed by the training test and would be able to categorise images into predefined
categories based on the pixel values of the images. For example, when differenet image of raw, ripe and rotten
bananas are fed into the model, the trained classification model would analyse the images and predict the most
likely category raw banana, ripe banana or rotten banana.
Ripe Banana Ripe Banana Raw Banana Rotten Banana Ripe Banana
Classification
Model
Raw Banana
Raw Banana Ripe Banana Rotten Banana
Rotten Banana Ripe Banana Rotten Banana Rotten Banana
Input Output
• Email Spam Detection: In this scenario, the model is trained with the words in the email, information of the
sender, precedence of links to classify the emails under “Spam” or “Not Spam”. The trained model would then
analyse the emails and predict them as “Spam” or “Not Spam”. Similar models are being used to categorise calls
and messages as “Spam”.
Inbox
Spam
Not Spam
Classification
Model
Spam
spam
Spam
Not Spam
Input Output
Advanced Concepts of Modeling in AI 193

