Friday, 10:30 AM – 12:00 PM
Chair: Jayant Madhavan

Exploring Web Scale Language Models for Search Query Processing

Jian Huang, Jiangbo Miao, Xiaolong Li, Jianfeng Gao, Kuansan Wang

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 previous 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.

Building Taxonomy of Web Search Intents for Name Entity Queries

Xiaoxin Yin, Sarthak Shah

A significant portion of web search queries are name entity queries. The major search engines have been exploring various ways to provide better user experiences for name entity queries, such as showing “search tasks” (Bing search) and showing direct answers (Yahoo!, Kosmix). In order to provide the search tasks or direct answers that can satisfy most popular user intents, we need to capture these intents, together with relationships between them. In this paper we propose an approach for building a hierarchical taxonomy of the generic search intents for a class of name entities (e.g., musicians or cities). The proposed approach can find phrases representing generic intents from user queries, and organize these phrases into a tree, so that phrases indicating equivalent or similar meanings are on the same node, and the parent-child relationships of tree nodes represent the relationships between search intents and their sub-intents. Three different methods are proposed for tree building, which are based on directed maximum spanning tree, hierarchical agglomerative clustering, and pachinko allocation model. Our approaches are purely based on search logs, and do not utilize any existing taxonomies such as Wikipedia. With the evaluation by human judges (via Mechanical Turk), it is shown that our approaches can build trees of phrases that capture the relationships between important search intents.

Optimal Rare Query Suggestion With Implicit User Feedback

Yang Song, Li-wei He

Query suggestion has been an effective approach to help users narrow down to the information they need. However, most of existing studies focused on only popular/head queries. Since rare queries possess much less information (e.g., clicks) than popular queries in the query logs, it is much more difficult to efficiently suggest relevant queries to a rare query. In this paper, we propose an optimal rare query suggestion framework by leveraging implicit feedbacks from users in the query logs. Our model resembles the principle of pseudo-relevance feedback which assumes that top-returned results by search engines are relevant. However, we argue that the clicked URLs and skipped URLs contain different levels of information and thus should be treated differently. Hence, our framework optimally combines both the click and skip information from users and uses a random walk model to optimize the query correlation. Our model specifically optimizes two parameters: (1) the restarting (jumping) rate of random walk, and (2) the combination ratio of click and skip information. Unlike the Rocchio algorithm, our learning process does not involve the content of the URLs but simply leverages the click and skip counts in the query-URL bipartite graphs. Consequently, our model is capable of scaling up to the need of commercial search engines. Experimental results on one-month query logs from a large commercial search engine with over 40 million rare queries demonstrate the superiority of our framework, with statistical significance, over the traditional random walk models and pseudo-relevance feedback models.


Back to full list of papers