Chapter 4. More-sensitive query suggestions

We’ve covered the fundamentals of neural networks and looked at the construction of both shallow and deep architectures for these networks. In practical terms, you now know how to integrate neural networks into a search engine to boost the search engine with two key features: synonym expansion and alternative query generation. Both of these features work on the search engine to make it smarter and return better results to the user. But can you do anything to improve the wording of the query itself? In particular, can you do anything to help the user write better queries—queries that deliver the results that come closest to what the user is looking for?

4.1. Generating query suggestions

4.2. Lucene Lookup APIs

4.3. Analyzed suggesters

4.4. Using language models

4.5. Content-based suggesters

4.6. Neural language models

4.7. Character-based neural language model for suggestions

4.8. Tuning the LSTM language model

4.9. Diversifying suggestions using word embeddings

Summary