Supervised-labelled does classification
unsupervised-not labeled finds similarity
reinforcement-continous learning by reward and penalty
the K-means algorithm differs in the method used for calculating the Euclidean distance while calculating the distance between each of two data items; and EM uses statistical methods.
A Perceptron is a neural network unit that does certain computations to detect features or business intelligence in the input data.
Collection of data-identifying sources,data quality,new features
Data preprocessing- data cleaning,filtering,removing outliers
Validation set is used for determining the parameters of the model,
and test set is used for evaluate the performance of the model
Centroid- imaginary center of cluster
By averaging the data points feature wise
Forward Propagation is the way to move from the Input layer (left) to the Output layer (right) in the neural network. The process of moving from the right to left i.e backward from the Output to the Input layer is called the Backward Propagation.
12.If for P(C1/X) and P(C2/X) is found using naive bayes classifier which gives higher accuracy or which one would the model choose
Greatest among two classes gives higher accuracy
KNN is a supervised learning algorithm used for classification.
A* : pathfinding problem in applications such as video games
AO*:
Candidate Elimination:
ID3:
ANN:
Naive Bayes:
KMeans:
KNN:
Regression:
