-
Book Overview & Buying
-
Table Of Contents
Machine Learning for Emotion Analysis in Python
By :
In this chapter, we looked at using neural networks for our task of identifying the emotions expressed in informal communications such as tweets. We examined the way that the lexicon for the datasets is used as the nodes in the input layer and looked at how the weights associated with individual words reflect the emotional significance of those words. We considered simple neural networks with no hidden layers and also slightly deeper ones with a single hidden layer with slightly more nodes than the set of output nodes – the performance of the neural network flattened out once the hidden layer contained 1.5 to 2 times as many nodes as the output layer, so there seemed little point.
The highest-scoring algorithms for the various datasets are now as follows:
|
LEX (unstemmed) |
LEX (stemmed) |
CP (stemmed) |
NB (multi... |