I am trying to implement a natural language query preprocessing module which would, given a query formulated in natural language, extract the keywords from that query and submit it to an IR (information retrieval system) system.
At first, I thought about using some training set to compute TF-IDF values of terms and use these values for estimating the importance of single words. But on second thought, this does not make any sense in this scenario - I only have a training collection but I dont have access to index the indexed IR data. Would it be reasonable to only use the IDF value for such estimation (is IDF enough to establish weight of a term in general)? Or maybe another weighting approach?
Could you suggest how to tackle this problem? Usually, the articles about NLP processing that I read talks about training and test data sets. But what if I only have the query and training data?