Incorporating features extracted from clickthrough data (called clickthrough features) has been demonstrated to significantly improve the performance of ranking models for Web search applications. Such benefits, however, are severely limited by the data sparseness problem, i.e., many queries and documents have no or very few clicks. The ranker thus cannot rely strongly on clickthrough features for document ranking. This paper presents two smoothing methods to expand clickthrough data: query clustering via Random Walk on click graphs and a discounting method inspired by the Good-Turing estimator. Both methods are evaluated on real-world data in three Web search domains. Experimental results show that the ranking models trained on smoothed clickthrough features consistently outperform those trained on unsmoothed features. This study demonstrates both the importance and the benefits of dealing with the sparseness problem in clickthrough data.

Users frequently modify a previous search query in hope of retrieving better results. These modifications are called query reformulations or query refinements. Existing research has studied how web search engines can propose reformulations, but has given less attention to how people perform query reformulations. In this paper, we aim to better understand how web searchers refine queries and form a theoretical foundation for query reformulation. We study users ’ reformulation strategies in the context of the AOL query logs. We create a taxonomy of query refinement strategies and build a high precision rule-based classifier to detect each type of reformulation. Effectiveness of reformulations is measured using user click behavior. Most reformulation strategies result in some benefit to the user. Certain strategies like add/remove words, word substitution, acronym expansion, and spelling correction are more likely to cause clicks, especially on higher ranked results. In contrast, users often click the same result as their previous query or select no results when forming acronyms and reordering words. Perhaps the most surprising finding is that some reformulations are better suited to helping users when the current results are already fruitful, while other reformulations are more effective when the results are lacking. Our findings inform the design of applications that can assist searchers; examples are described in this paper.

The World Wide Web provides an abundant source of medical information. This information can assist people who are not healthcare professionals to better understand health and disease, and to provide them with feasible explanations for symptoms. However, the Web has the potential to increase the anxieties of people who have little or no medical training, especially when Web search is employed as a diagnostic procedure. We use the term cyberchondria to refer to the unfounded escalation of concerns about common symptomatology, based on the review of search results and literature on the Web. We performed a large-scale, longitudinal, log-based study of how people search for medical information online, supported by a large-scale survey of 515 individuals ’ health-related search experiences. We focused on the extent to which common, likely innocuous symptoms can escalate into the review of content on serious, rare conditions that are linked to the common symptoms. Our results show that Web search engines have the potential to escalate medical concerns. We show that escalation is influenced by the amount and distribution of medical content viewed by users, the presence of escalatory terminology in pages visited, and a user’s predisposition to escalate versus to seek more reasonable explanations for ailments. We also demonstrate the persistence of post-session anxiety following escalations and the effect that such anxieties can have on interrupting user’s activities across multiple sessions. Our findings underscore the potential costs and challenges of cyberchondria and suggest actionable design implications that hold opportunity for improving the search and navigation experience for people turning to the Web to interpret common symptoms.

Traditionally, stemming has been applied to Information Retrieval tasks by transforming words in documents to the their root form before indexing, and applying a similar transformation to query terms. Although it increases recall, this naive strategy does not work well for Web Search since it lowers precision and requires a significant amount of additional computation. In this paper, we propose a context sensitive stemming method that addresses these two issues. Two unique properties make our approach feasible for Web Search. First, based on statistical language modeling, we perform context sensitive analysis on the query side. We accurately predict which of its morphological variants is useful to expand a query term with before submitting the query to the search engine. This dramatically reduces the number of bad expansions, which in turn reduces the cost of additional computation and improves the precision at the same time. Second, our approach performs a context sensitive document matching for those expanded variants. This conservative strategy serves as a safeguard against spurious stemming, and it turns out to be very important for improving precision. Using word pluralization handling as an example of our stemming approach, our experiments on a major Web search engine show that stemming only 29 % of the query traffic, we can improve relevance as measured by average Discounted Cumulative Gain (DCG5) by 6.1 % on these queries and 1.8 % over all query traffic.

Web search engines are traditionally evaluated in terms of the relevance of web pages to individual queries. However, relevance of web pages does not tell the complete picture, since an individual query may represent only a piece of the user’s information need and users may have different information needs underlying the same queries. We address the problem of predicting user search goal success by modeling user behavior. We show empirically that user behavior alone can give an accurate picture of the success of the user’s web search goals, without considering the relevance of the documents displayed. In fact, our experiments show that models using user behavior are more predictive of goal success than those using document relevance. We build novel sequence models incorporating time distributions for this task and our experiments show that the sequence and time distribution models are more accurate than static models based on user behavior, or predictions based on document relevance.

This paper explores the use of clickthrough data for query spelling correction. First, large amounts of query-correction pairs are derived by analyzing users ' query reformulation behavior encoded in the clickthrough data. Then, a phrase-based error model that accounts for the transformation probability between multi-term phrases is trained and integrated into a query speller system. Experiments are carried out on a human-labeled data set. Results show that the system using the phrase-based error model outperforms significantly its baseline systems. 1

It has been widely observed that search queries are composed in a very different style from that of the body or the title of a document. Many techniques explicitly accounting for this language style discrepancy have shown promising results for information retrieval, yet a large scale analysis on the extent of the language differences has been lacking. In this paper, we present an extensive study on this issue by examining the language model properties of search queries and the three text streams associated with each web document: the body, the title, and the anchor text. Our information theoretical analysis shows that queries seem to be composed in a way most similar to how authors summarize documents in anchor texts or titles, offering a quantitative explanation to the observations in past work. We apply these web scale n-gram language models to three search query processing (SQP) tasks: query spelling correction, query bracketing and long query segmentation. By controlling the size and the order of different language models, we find that the perplexity metric to be a good accuracy indicator for these query processing tasks. We show that using smoothed language models yields significant accuracy gains for query bracketing for instance, compared to using web counts as in the literature. We also demonstrate that applying web-scale language models can have marked accuracy advantage over smaller ones.

A growing trend in commercial search engines is the display of specialized content such as news, products, etc. interleaved with web search results. Ideally, this content should be displayed only when it is highly relevant to the search query, as it competes for space with “regular ” results and advertisements. One measure of the relevance to the search query is the click-through rate the specialized content achieves when displayed; hence, if we can predict this click-through rate accurately, we can use this as the basis for selecting when to show specialized content. In this paper, we consider the problem of estimating the clickthrough rate for dedicated news search results. For queries for which news results have been displayed repeatedly before, the click-through rate can be tracked online; however, the key challenge for which previously unseen queries to display news results remains. In this paper we propose a supervised model that offers accurate prediction of news click-through rates and satisfies the requirement of adapting quickly to emerging news events.

Many have speculated that classifying web pages can improve a search engine’s ranking of results. Intuitively results should be more relevant when they match the class of a query. We present a simple framework for classification-enhanced ranking that uses clicks in combination with the classification of web pages to derive a class distribution for the query. We then go on to define a variety of features that capture the match between the class distributions of a web page and a query, the ambiguity of a query, and the coverage of a retrieved result relative to a query’s set of classes. Experimental results demonstrate that a ranker learned with these features significantly improves ranking over a competitive baseline. Furthermore, our methodology is agnostic with respect to the classification space and can be used to derive query classes for a variety of different taxonomies.

While Implicit Relevance Feedback (IRF) algorithms exploit users ’ interactions with information to customize support offered to users of search systems, it is unclear how individual and task differences impact the effectiveness of such algorithms. In this paper we describe a study on the effect on retrieval performance of using additional information about the user and their search tasks when developing IRF algorithms. We tested four algorithms that use document display time to estimate relevance, and tailored the threshold times (i.e., the time distinguishing relevance from non-relevance) to the task, the user, a combination of both, or neither. Interaction logs gathered during a longitudinal naturalistic study of online information-seeking behavior are used as stimuli for the algorithms. The findings show that tailoring display time thresholds based on task information improves IRF algorithm performance, but doing so based on user information worsens performance. This has implications for the development of effective IRF algorithms.