Answering General Time-Sensitive Queries

Abstract

Time is an important dimension of relevance for a large number of searches, such as over blogs and news archives. So far, research on searching over such collections has largely focused on locating topically similar documents for a query. Unfortunately, topic similarity alone is not always sufficient for document ranking. In this paper, we observe that, for an important class of queries that we call time-sensitive queries, the publication time of the documents in a news archive is important and should be considered in conjunction with the topic similarity to derive the final document ranking. Earlier work has focused on improving retrieval for "recency" queries that target recent documents. We propose a more general framework for handling time-sensitive queries and we automatically identify the important time intervals that are likely to be of interest for a query. Then, we build scoring techniques that seamlessly integrate the temporal aspect into the overall ranking mechanism. We present an extensive experimental evaluation using a variety of news article data sets, including TREC data as well as real web data analyzed using the Amazon Mechanical Turk. We examine several techniques for detecting the important time intervals for a query over a news archive and for incorporating this information in the retrieval process. We show that our techniques are robust and significantly improve result quality for time-sensitive queries compared to state-of-the-art retrieval techniques.

Question temporality

“…It is these 17 strongly temporal questions that we attempt to identify automatically. 2 We only showed the questions and not the answers provided on those questions. …”

“…Zhang et al [7] analyzed query volume at Yahoo to detect time sensitive queries, based on which queries are accompanied by year, and proposed a ranking method for web results for the time sensitive queries. Dakka et al [2] presented a framework for handling time sensitive queries and identifying the important time intervals that might be interesting for a given query. They ran their experiments on news articles to show that temporality aspect of a document is an important aspect during ranking of web results.…”

Supervised language modeling for temporal resolution of texts

“…IR research has investigated time-sensitivity query interpretation and document ranking [17,9], time-based organization and presentation of search results [2], how queries and documents change over time [16], etc. One of the first LM-based temporal approaches by Li and Croft [17] used explicit document dates to estimate a more informative document prior.…”

“…One of the first LM-based temporal approaches by Li and Croft [17] used explicit document dates to estimate a more informative document prior. More recent work by Dakka et al [9] automatically identify important time intervals likely to be of interest for a query and similarly integrate knowledge of document publication date into the ranking function. The most relevant work to ours is that by Alonso et al [2], who provide valuable background on motivation, overview and discussion of temporal analysis in IR.…”

“…In computational linguistics, the primary focus has been on the finegrained level of temporal ordering of individual events necessary for deep natural language understanding [1,24]. In Information Retrieval (IR), research has investigated time-sensitivity document ranking [9,17], time-based organization and presentation of search results [2], how queries and documents change over time [16], etc.…”