Chapter 1. Neural search
Figure 1.1. Artificial intelligence, machine learning, and deep learning
Figure 1.2. A deep neural network with two hidden layers
Figure 1.3. Searching and getting results
Figure 1.4. Getting the words of “I like search engines” using a simple text analysis pipeline
Figure 1.5. The traversed token graph
Figure 1.6. Query parsing
Figure 1.7. Index, search-time analysis, and term matching
Figure 1.8. Similarities between document and query vectors according to VSM
Figure 1.9. Learning image abstractions incrementally
Figure 1.10. A deep feed-forward neural network with three hidden layers
Figure 1.11. Word vectors derived from the text of research articles on word2vec
Figure 1.12. A neural search application: using word representations generated by a deep neural network to provide more-relevant results
Figure 1.13. Forward step (feeding input) and backward step (backpropagating an error)
Chapter 2. Generating synonyms
Figure 2.1. Synonym expansion at search time, with a neural network
Figure 2.2. Synonym expansion graph
Figure 2.3. Splitting portions of the text depending on the type of data
Figure 2.4. Predicting price with a feed-forward neural network with three inputs, five hidden units, and one output unit
Figure 2.5. Propagating signals through the network
Figure 2.6. Backpropagating a signal from the output layer to the hidden layer
Figure 2.7. Geometric interpretation of backpropagation with gradient descent