Python Deep Learning Projects

Chapter 3 – Word Representations Using word2vec

Core to computational linguistics is an effective representation of words and the features they embody. word2vec was used to transform the words into dense vectors (that is, tensors), creating embedding representations for the corpus. We then created a convolutional neural network (CNN) to build a language model for sentiment analysis. To help us frame this task we envisioned the hypothetical use case of our restaurant chain client asking us to make sense of the response texts that they were receiving from their patrons getting the notification that their table was ready. Particularly interesting was the realization that CNNs can be applied to more than just image data! We also took this project ...