Promoting Relevant Results in Time-Ranked Mail Search

Carmel, David; Lewin-Eytan, Liane; Libov, Alex; Maarek, Yoelle; Raviv, Ariel

doi:10.1145/3038912.3052659

Cited by 16 publications

(9 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our work is related to recent work on personal search [2,3,6,22,33] that emphasized relevance-based over time-based ranking [6]. User behaviors in personal search show different traits compared with web search [2].…”

Section: Related Workmentioning

confidence: 98%

See 1 more Smart Citation

Position Bias Estimation for Unbiased Learning to Rank in Personal Search

Wang

Golbandi

Bendersky

et al. 2018

Proceedings of the Eleventh ACM International Conference on Web Search and Data Mining

232

272

View full text Add to dashboard Cite

A well-known challenge in learning from click data is its inherent bias and most notably position bias. Traditional click models aim to extract the ⟨query, document⟩ relevance and the estimated bias is usually discarded after relevance is extracted. In contrast, the most recent work on unbiased learning-to-rank can effectively leverage the bias and thus focuses on estimating bias rather than relevance [20,31]. Existing approaches use search result randomization over a small percentage of production traffic to estimate the position bias. This is not desired because result randomization can negatively impact users' search experience. In this paper, we compare different schemes for result randomization (i.e., RandTopN and RandPair) and show their negative effect in personal search. Then we study how to infer such bias from regular click data without relying on randomization. We propose a regressionbased Expectation-Maximization (EM) algorithm that is based on a position bias click model and that can handle highly sparse clicks in personal search. We evaluate our EM algorithm and the extracted bias in the learning-to-rank setting. Our results show that it is promising to extract position bias from regular clicks without result randomization. The extracted bias can improve the learning-to-rank algorithms significantly. In addition, we compare the pointwise and pairwise learning-to-rank models. Our results show that pairwise models are more effective in leveraging the estimated bias.

show abstract

Section: Related Workmentioning

confidence: 98%

“…With the increasing popularity of cloud-based email and storage services, there is a large amount of personal data stored in the cloud. This large amount of data calls for effective search capabilities to improve its utility [2,5,6,15]. The important difference of personal search from web search is the private content in personal search.…”

Section: Introductionmentioning

confidence: 99%

Position Bias Estimation for Unbiased Learning to Rank in Personal Search

Wang

Golbandi

Bendersky

et al. 2018

Proceedings of the Eleventh ACM International Conference on Web Search and Data Mining

232

272

View full text Add to dashboard Cite

show abstract

“…Carmel et al [12] proposed an email search framework with a learning-to-rank reranking module that combines freshness with relevance signals of emails as well as other features such as user actions. Alternatively, Carmel et al [13] studied to present users with both the relevanceranked results as well as the time-ranked results in two separate lists for better user experience. A number of studies specifically focus on improving the content-based relevance signals in email search.…”

Section: Email Searchmentioning

confidence: 99%

Separate and Attend in Personal Email Search

Karimzadehgan

Zhuang

et al. 2020

Proceedings of the 13th International Conference on Web Search and Data Mining

View full text Add to dashboard Cite

In personal email search, user queries often impose different requirements on different aspects of the retrieved emails. For example, the query "my recent flight to the US" requires emails to be ranked based on both textual contents and recency of the email documents, while other queries such as "medical history" do not impose any constraints on the recency of the email. Recent deep learning-to-rank models for personal email search often directly concatenate dense numerical features 1 (e.g., document age) with embedded sparse features (e.g., n-gram embeddings). In this paper, we first show with a set of experiments on synthetic datasets that direct concatenation of dense and sparse features does not lead to the optimal search performance of deep neural ranking models. To effectively incorporate both sparse and dense email features into personal email search ranking, we propose a novel neural model, SepAttn. SepAttn first builds two separate neural models to learn from sparse and dense features respectively, and then applies an attention mechanism at the prediction level to derive the final prediction from these two models. We conduct a comprehensive set of experiments on a large-scale email search dataset, and demonstrate that our SepAttn model consistently improves the search quality over the baseline models.

show abstract

“…They are including the name of their current location. In typical email search, like real-time search, one of the key features used to present results or to rank is recency [1,9]. If we consider the example of a user submitting the generic query "coupon", a typical system would probably retrieve poor results, given the frequency of promotional email -the most recent "coupon" you've received is not necessarily correlated with your immediate desire for a coupon.…”

Section: Query-log Analysismentioning

confidence: 99%

Semantic Location in Email Query Suggestion

Foley

Zhang

Bendersky

et al. 2018

The 41st International ACM SIGIR Conference on Research &Amp; Development in Information Retrieval

View full text Add to dashboard Cite

Mobile devices are pervasive, which means that users have access to web content and their personal documents at all locations, not just their home or office. Existing work has studied how locations can influence information needs, focusing on web queries. We explore whether or not location information can be helpful to users who are searching their own personal documents. We wish to study whether a users' location can predict their queries over their own personal data, so we focus on the task of query suggestion. While we find that using location directly can be helpful, it does not generalize well to novel locations. To improve this situation, we explore using semantic location: that is, rather than memorizing location-query associations, we generalize our location information to names of the closest point of interest. By using short, semantic descriptions of locations, we find that we can more robustly improve query completion and observe that users are already using locations to extend their own queries in this domain. We present a simple but effective model that can use location to predict queries for a user even before they type anything into a search box, and which learns effectively even when not all queries have location information.

show abstract

Promoting Relevant Results in Time-Ranked Mail Search

Cited by 16 publications

References 17 publications

Position Bias Estimation for Unbiased Learning to Rank in Personal Search

Position Bias Estimation for Unbiased Learning to Rank in Personal Search

Separate and Attend in Personal Email Search

Semantic Location in Email Query Suggestion

Contact Info

Product

Resources

About