String Processing and Information Retrieval

Abstract

e Unbiased Learning-to-Rank framework [19] has been recently proposed as a general approach to systematically remove biases, such as position bias, from learning-to-rank models. e method takes two steps -estimating click propensities and using them to train unbiased models. Most common methods proposed in the literature for estimating propensities involve some degree of intervention in the live search engine. An alternative approach proposed recently uses an Expectation Maximization (EM) algorithm to estimate propensities by using ranking features for estimating relevances [23]. In this work we propose a novel method to directly estimate propensities which does not use any intervention in live search or rely on modeling relevance. Rather, we take advantage of the fact that the same query-document pair may naturally change ranks over time.is typically occurs for eCommerce search because of change of popularity of items 1 over time, existence of time dependent ranking features, or addition or removal of items to the index (an item ge ing sold or a new item being listed). However, our method is general and can be applied to any search engine for which the rank of the same document may naturally change over time for the same query. We derive a simple likelihood function that depends on propensities only, and by maximizing the likelihood we are able to get estimates of the propensities. We apply this method to eBay search data to estimate click propensities for web and mobile search and compare these with estimates using the EM method [23]. We also use simulated data to show that the method gives reliable estimates of the "true" simulated propensities. Finally, we train an unbiased learning-to-rank model for eBay search using the estimated propensities and show that it outperforms both baselines -one without position bias correction and one with position bias correction using the EM method.

Palidis: fast discovery of novel insertion sequences

“…These ISs can then be pooled and clustered to create a non-redundant catalogue of ISs. Palidis can also predict the origins of these ISs by querying them against ISfinder or a compact bit-sliced signature (COBS) index [ 11 ] of 661 405 microbial genomes [ 12 ]. Here, we present the theory and implementation of this tool on 264 short-read metagenomes to generate 879 unique ISs included in the first release of the insertion sequence Catalogue (ISC).…”