Domain independent keyword identification for question answering

In this paper, we look at domain independent keyword identification for natural language queries using statistical methods. We took queries supplemented by only their dependency tags (Stanford Parser) and part-of-speech tags (Stanford POS tagger) and labeled the keywords. We then delexicalised the training data, and used the Conditional Random Fields algorithm to learn these labels. We used the queries created by [1] in the course management domain for training, and tested our model on the queries of three domains: course management, library and the GeoQueries250 dataset and report fairly high accuracies of 90.65%, 83.19% and 97.13% respectively, making our model a truly domain independent and highly accurate keyword identifier.