Learning to rank search results [DevCon]


With advanced tools available for search like Solr and Elasticsearch, companies are embedding search in almost all their products and websites. Search is becoming mainstream. Therefore we can focus on teaching the search engine tricks to return more relevant results. One new trick is called “learning to rank”. Learning to rank uses a trained model to come up with a better ranking of the search results. During the presentation you’ll learn what Learning To Rank is. To be able to understand the machine learning part, you get information about machine learning models, feature extraction and the training of models. You will also learn about when to apply learning to rank and of course you’ll get an example to show how it works using elasticsearch and a learning to rank plugin. After this presentation you have learned how and why to combine Machine Learning and Search.