Ranking a set of retrieved documents according to their relevance to a query is a popular problem in information retrieval. Methods that learn ranking functions are difficult to o...
We present a machine learning approach for the task of ranking previously answered questions in a question repository with respect to their relevance to a new, unanswered referenc...
Several algorithms have been proposed to learn to rank entities modeled as feature vectors, based on relevance feedback. However, these algorithms do not model network connections...
As with any application of machine learning, web search ranking requires labeled data. The labels usually come in the form of relevance assessments made by editors. Click logs can...
We describe a machine learning approach for predicting sponsored search ad relevance. Our baseline model incorporates basic features of text overlap and we then extend the model t...
Dustin Hillard, Stefan Schroedl, Eren Manavoglu, H...