wars@15

Hello, I'm trying to follow this work : An Empirical Study of Learning to Rank for Entity Search https://dl.acm.org/citation.cfm?id=2914725 They are basically applying FSDM , Coordinate match and Cosine similarity etc. to entity search task. 1- The FSDM : Fielded Sequential Dependence Model for Ad-Hoc Entity Retrieval in the Web of Data https://dl.acm.org/citation.cfm?id=2767756 , they mention that it's provided by Galago , but I couldn't find any information about it in documentation. So is it just...

Thank you so much.

Hi all, This is my first time to use indri. I got exactly the same results for both language model and okapi model. runid all indri num_q all 485 num_ret all 482154 num_rel all 13090 num_rel_ret all 5380 map all 0.1351 gm_map all 0.0131 The syntaxes I have used for the queries are : 1- #combine(query) for language model 2- # combine(#bm25(term1) #bm25(term2)) for bm25 (seen here https://sourceforge.net/p/lemur/discussion/galago/thread/10ff53c2/?limit=25#444d ) suspecting those results I changed bm25...

Hi all, This is my first time to use indri. I got exactly the same results for both language model and okapi model. runid all indri num_q all 485 num_ret all 482154 num_rel all 13090 num_rel_ret all 5380 map all 0.1351 gm_map all 0.0131 The syntaxes I have used for the queries are : 1- #combine(query) for language model 2- # combine(#bm25(term1) #bm25(term2)) for bm25 (seen here https://sourceforge.net/p/lemur/discussion/galago/thread/10ff53c2/?limit=25#444d ) suspecting those results I changed bm25...

Hi there, How can I use language model with different smoothings ? what is the syntaxe in indri ?