Learning to Attend, Copy, and Generate for Session-Based Query Suggestion

Abstract

Users try to articulate their complex information needs during search sessions by reformulating their queries. To make this process more e ective, search engines provide related queries to help users in specifying the information need in their search process. In this paper we propose a customized sequence-to-sequence model for sessionbased query suggestion. In our model, we employ a query-aware a ention mechanism to capture the structure of the session context. is enables us to control the scope of the session from which we infer the suggested next query, which helps not only handle the noisy data but also automatically detect session boundaries. Furthermore we observe that, based on the user query reformulation behavior, within a single session a large portion of query terms is retained from the previously submi ed queries and consists of mostly infrequent or unseen terms that are usually not included in the vocabulary. We therefore empower the decoder of our model to access the source words from the session context during decoding by incorporating a copy mechanism. Moreover, we propose evaluation metrics to assess the quality of the generative models for query suggestion. We conduct an extensive set of experiments and analysis. e results suggest that our model outperforms the baselines both in terms of the generating queries and scoring candidate queries for the task of query suggestion.

Context Attentive Document Ranking and Query Suggestion

“…As we filtered out queries that do not have any associated clicks when constructing the experiment dataset, we lost some longer tasks; otherwise our test data distribution is similar to [8].…”

Using Image Captions and Multitask Learning for Recommending Query Reformulations

“…In this section, we evaluate the performance of the aforementioned models using multiple metrics for each of the two tasks: query reformulation and ranking. The metrics used here are largely inspired from [11], and we discuss these below briefly. Towards the end of the section we also provide some qualitative results.…”

“…, w lc }. We use LSTMs [15] to model the sequences, owing to their demonstrated capabilities in modeling various natural language tasks, ranging from machine translation [27] to query suggestion [11].…”

Stuck? No worries!

“…One such class of approaches is referred to as sequence-to-sequence (seq2seq) modeling. These techniques have been successfully applied to tasks ranging from machine translation [22] to query suggestion [8]. Motivated by their wide-ranging applicability, we use variants of seq2seq models to recommend future commands.…”