DLB Final

1. Instances of class A are separated from instances of class B by drawing a straight line
5. It is an algorithm used to train neural network with hidden layers
6. Nodes with constant input value associated with hidden and output layer nodes
8. It describes only the layer-wise training of the autoencoders
10. A matrix used to summarize the results of a supervised classification
12. A simple feed-forward neural network consisting of an input layer and a single output layer
15. Each node in a given layer connects to all nodes in the next layer

2. The neural network is composed to two or more layers of nodes and typically includes an input layer, one or more hidden layers, and an output layer
3. Associated with the fact that the training time becoming longer withs more layers added to that network
4. One of several commonly used neural network evaluation functions which is continuous and outputs a value between 0 and 1
7. Problem that examines why and how a customer terminates a service or subscription with a company
9. One complete pass of the training data through a neural network
11. A supervised learning strategy designed to determine future outcome
13. Data instances used to test supervised learning models
14. This characteristic implies that information flows in a single direction and does not loop or cycle