A feature-free search query classification approach using semantic distance

When classifying search queries into a set of target categories, machine learning based conventional approaches usually make use of external sources of information to obtain additional features for search queries and training data for target categories. Unfortunately, these approaches rely on large amount of training data for high classification precision. Moreover, they are known to suffer from inability to adapt to different target categories which may be caused by the dynamic changes observed in both Web topic taxonomy and Web content. In this paper, we propose a feature-free classification approach using semantic distance. We analyze queries and categories themselves and utilizes the number of Web pages containing both a query and a category as a semantic distance to determine their similarity. The most attractive feature of our approach is that it only utilizes the Web page counts estimated by a search engine to provide the search query classification with respectable accuracy. In addition, it can be easily adaptive to the changes in the target categories, since machine learning based approaches require extensive updating process, e.g., re-labeling outdated training data, re-training classifiers, to name a few, which is time consuming and high-cost. We conduct experimental study on the effectiveness of our approach using a set of rank measures and show that our approach performs competitively to some popular state-of-the-art solutions which, however, frequently use external sources and are inherently insufficient in flexibility.     

QUBiC: An adaptive approach to query-based recommendation

Search engine users often encounter the difficulty of phrasing the precise query that could lead to satisfactory search results. Query recommendation is considered an effective assistant in enhancing keyword-based queries in search engines and Web search software. In this paper, we present a Query-URL Bipartite based query reCommendation approach, called QUBiC. It utilizes the connectivity of a query-URL bipartite graph to recommend related queries and can significantly improve the accuracy and effectiveness of personalized query recommendation systems comparing with the conventional pairwise similarity based approach. The main contribution of the QUBiC approach is its three-phase framework for personalized query recommendations. The first phase is the preparation of queries and their search results returned by a search engine, which generates a historical query-URL bipartite collection. The second phase is the discovery of similar queries by extracting a query affinity graph from the bipartite graph, instead of operating on the original bipartite graph directly using biclique-based approach or graph clustering. The query affinity graph consists of only queries as its vertices and its edges are weighted according to a query-URL vector based similarity (dissimilarity) measure. The third phase is the ranking of similar queries. We devise a novel rank mechanism for ordering the related queries based on the merging distances of a hierarchical agglomerative clustering (HAC). By utilizing the query affinity graph and the HAC-based ranking, we are able to capture the propagation of similarity from query to query by inducing an implicit topical relatedness between queries. Furthermore, the flexibility of the HAC strategy makes it possible for users to interactively participate in the query recommendation process, and helps to bridge the gap between the determinacy of actual similarity values and the indeterminacy of users’ information needs, allowing the lists of related queries to be changed from user to user and query to query, thus adaptively recommending related queries on demand. Our experimental evaluation results show that the QUBiC approach is highly efficient and more effective compared to the conventional query recommendation systems, yielding about 13.3 % as the most improvement in terms of precision.     

Measuring semantic similarity between words by removing noise and redundancy in web snippets

Semantic similarity measures play important roles in many Web‐related tasks such as Web browsing and query suggestion. Because taxonomy‐based methods can not deal with continually emerging words, recently Web‐based methods have been proposed to solve this problem. Because of the noise and redundancy hidden in the Web data, robustness and accuracy are still challenges. In this paper, we propose a method integrating page counts and snippets returned by Web search engines. Then, the semantic snippets and the number of search results are used to remove noise and redundancy in the Web snippets (‘Web‐snippet’ includes the title, summary, and URL of a Web page returned by a search engine). After that, a method integrating page counts, semantics snippets, and the number of already displayed search results are proposed. The proposed method does not need any human annotated knowledge (e.g., ontologies), and can be applied Web‐related tasks (e.g., query suggestion) easily. A correlation coefficient of 0.851 against Rubenstein–Goodenough benchmark dataset shows that the proposed method outperforms the existing Web‐based methods by a wide margin. Moreover, the proposed semantic similarity measure significantly improves the quality of query suggestion against some page counts based methods. Copyright © 2011 John Wiley & Sons, Ltd.     

一种基于查询特性的查询结果缓存与预取方法

针对搜索引擎查询结果缓存与预取问题,该文提出了一种基于查询特性的搜索引擎查询结果缓存与预取方法,该方法包括用来指导预取的查询结果页码预测模型和缓存与预取算法框架,用于提高搜索引擎系统性能。通过对国内某著名中文商业搜索引擎的某段时间的用户查询日志分析得出,用户对不同查询返回的查询结果所浏览的页数具有显著的非均衡性,结合该特性设计查询结果页码预测模型来进行预取和分区缓存。在该搜索引擎两个月的大规模真实用户查询日志上的实验结果表明,与传统的方法相比,该方法可以获得3.5%~8.45%的缓存命中率提升。     

DeepWeb信息集成系统中查询转换

随着Internet信息的迅速增长,许多Web信息已经被各种各样的可搜索在线数据库所深化,并被隐藏在Web查询接口下面。传统的搜索引擎由于技术原因不能索引这些信息——DeepWeb信息。由于DeepWeb惟一“入口点”是查询接口,为使查询接口自动产生有意义有查询,给出了DeepWeb信息集成系统框架,提出了基于数据类型的搜索驱动的用户查询转换方法,基于此设计并实现了一个针对中文DeepWeb信息集成原型系统。通过在实际DeepWeb站点上的实验证明了此方法是非常有效的。     

基于用户查询日志的网络搜索主题分析

网络搜索分析在优化搜索引擎方面具有举足轻重的作用,而且对用户个人搜索特性进行分析能够提高搜索引擎的精准度。目前,大多数已有模型（比如点击图模型及其变体）,注重研究用户群体的共同特点。然而,关于如何做到既可以获取用户群体共同特点又可以获取用户个人特点方面的研究却非常少。本文研究了基于个人用户网络搜索分析新问题,即通过研究用户搜索的突发性现象,获取个人用户搜索查询的主题分布情况。提出了两个搜索主题模型,即搜索突发性模型（SBM）和耦合敏感搜索突发性模型（CS-SBM）。SBM假设查询词和URL主题是无关的,CS-SBM假设查询词和URL之间是有主题关联的,得到的主题分布信息存储在偏Dirichlet先验中,采用Beta分布刻画用户搜索的时间特性。实验结果表明,每一个用户的网络搜索轨迹都有多种基于用户的独有特点。同时,在使用大量真实用户查询日志数据情况下,与LDA、DCMLDA、TOT相比,本文提出的模型具有明显的泛化性能优势,并且有效地描绘了用户搜索查询主题在时间上的变化过程。     

AN INTERACTIVE SEARCH ASSISTANT ARCHITECTURE BASED ON INTRINSIC QUERY STREAM CHARACTERISTICS

Search engine query log mining has evolved over time to more like data stream mining due to the endless and continuous sequence of queries known as query stream. In this paper, we propose an online frequent sequence discovery (OFSD) algorithm to extract frequent phrases from within query streams, based on a new frequency rate metric, which is suitable for query stream mining. OFSD is an online, single pass, and real-time frequent sequence miner appropriate for data streams. The frequent phrases extracted by the OFSD algorithm are used to guide novice Web search engine users to complete their search queries more efficiently. YourEye, our online phrase recommender is then introduced. The advantages of YourEye compared with Google Suggest, a service powered by Google for phrase suggestion, is also described. Various characteristics of two specific Web search engine query logs are analyzed and then the query logs are used to evaluate YourEye. The experimental results confirm the significant benefit of monitoring frequent phrases within the queries instead of the whole queries because none-separable items. The number of the monitored elements substantially decreases, which results in smaller memory consumption as well as better performance. Re-ranking the retrieved pages based on past users clicks for each frequent phrase extracted by OFSD is also introduced. The preliminary results show the advantages of the proposed method compared to the similar work reported in Smyth et al.     

Using Local Popularity of Web Resources for Geo-Ranking of Search Engine Results

Search engines retrieve and rank Web pages which are not only relevant to a query but also important or popular for the users. This popularity has been studied by analysis of the links between Web resources. Link-based page ranking models such as PageRank and HITS assign a global weight to each page regardless of its location. This popularity measurement has shown successful on general search engines. However unlike general search engines, location-based search engines should retrieve and rank higher the pages which are more popular locally. The best results for a location-based query are those which are not only relevant to the topic but also popular with or cited by local users. Current ranking models are often less effective for these queries since they are unable to estimate the local popularity. We offer a model for calculating the local popularity of Web resources using back link locations. Our model automatically assigns correct locations to the links and content and uses them to calculate new geo-rank scores for each page. The experiments show more accurate geo-ranking of search engine results when this model is used for processing location-based queries.     

Topic-sensitive PageRank: a context-sensitive ranking algorithm for Web search

The original PageRank algorithm for improving the ranking of search-query results computes a single vector, using the link structure of the Web, to capture the relative "importance" of Web pages, independent of any particular search query. To yield more accurate search results, we propose computing a set of PageRank vectors, biased using a set of representative topics, to capture more accurately the notion of importance with respect to a particular topic. For ordinary keyword search queries, we compute the topic-sensitive PageRank scores for pages satisfying the query using the topic of the query keywords. For searches done in context (e.g., when the search query is performed by highlighting words in a Web page), we compute the topic-sensitive PageRank scores using the topic of the context in which the query appeared. By using linear combinations of these (precomputed) biased PageRank vectors to generate context-specific importance scores for pages at query time, we show that we can generate more accurate rankings than with a single, generic PageRank vector. We describe techniques for efficiently implementing a large-scale search system based on the topic-sensitive PageRank scheme.     

Personalized query suggestion diversification in information retrieval

Query suggestions help users refine their queries after they input an initial query.Previous work on query suggestion has mainly concentrated on approaches that are similarity-based or context-based,developing models that either focus on adapting to a specific user(personalization)or on diversifying query aspects in order to maximize the probability of the user being satisfied(diversification).We consider the task of generating query suggestions that are both personalized and diversified.We propose a personalized query suggestion diversification(PQSD)model,where a user's long-term search behavior is injected into a basic greedy query suggestion diversification model that considers a user's search context in their current session.Query aspects are identified through clicked documents based on the open directory project(ODP)with a latent dirichlet allocation(LDA)topic model.We quantify the improvement of our proposed PQSD model against a state-of-the-art baseline using the public america online(AOL)query log and show that it beats the baseline in terms of metrics used in query suggestion ranking and diversification.The experimental results show that PQSD achieves its best performance when only queries with clicked documents are taken as search context rather than all queries,especially when more query suggestions are returned in the list.     

Finding Related Search Engine Queries by Web Community Based Query Enrichment

The conventional approaches of finding related search engine queries rely on the common terms shared by two queries to measure their relatedness. However, search engine queries are usually short and the term overlap between two queries is very small. Using query terms as a feature space cannot accurately estimate relatedness. Alternative feature spaces are needed to enrich the term based search queries. In this paper, given a search query, first we extract the Web pages accessed by users from Japanese Web access logs which store the users individual and collective behavior. From these accessed Web pages we usually can get two kinds of feature spaces, i.e, content-sensitive (e.g., nouns) and content-ignorant (e.g., URLs), to enrich the expressions of search queries. Then, the relatedness between search queries can be estimated on their enriched expressions. Our experimental results show that the URL feature space produces much lower precision scores than the noun feature space which, however, is not applicable in non-text pages, dynamic pages and so on. It is crucial to improve the quality of the URL (content-ignorant) feature space since it is generally available in all types of Web pages. We propose a novel content-ignorant feature space, called Web community which is created from a Japanese Web page archive by exploiting link analysis. Experimental results show that the proposed Web community feature space generates much better results than the URL feature space.     

基于加权SimRank的中文查询推荐研究

查询推荐是搜索引擎系统中的一项重要技术,其通过推荐更合适的查询以提高用户的搜索体验。现有方法能够找到直接通过某种属性关联的相似查询,却忽略了具有间接关联的语义相关查询。该文将用户查询及查询间直接联系建模为查询关系图,并在图结构相似度算法SimRank的基础上提出了加权SimRank (简称WSimRank)用于查询推荐。WSimRank综合考虑了查询关系图的全局信息,因而能挖掘出查询间的间接关联和语义关系。然而,WSimRank复杂度太高而难以实用,该文将WSimRank转换为一个状态层次图的遍历和计算过程,进而采用动态规划、剪枝等策略对其进行优化从而可以实际应用。在大规模真实Web搜索日志上的实验表明, WSimRank在各项评价指标上均优于SimRank和传统查询推荐方法,其MAP指标接近0.9。     

Search personalization through query and page topical analysis

Thousands of users issue keyword queries to the Web search engines to find information on a number of topics. Since the users may have diverse backgrounds and may have different expectations for a given query, some search engines try to personalize their results to better match the overall interests of an individual user. This task involves two great challenges. First the search engines need to be able to effectively identify the user interests and build a profile for every individual user. Second, once such a profile is available, the search engines need to rank the results in a way that matches the interests of a given user. In this article, we present our work towards a personalized Web search engine and we discuss how we addressed each of these challenges. Since users are typically not willing to provide information on their personal preferences, for the first challenge, we attempt to determine such preferences by examining the click history of each user. In particular, we leverage a topical ontology for estimating a user’s topic preferences based on her past searches, i.e. previously issued queries and pages visited for those queries. We then explore the semantic similarity between the user’s current query and the query-matching pages, in order to identify the user’s current topic preference. For the second challenge, we have developed a ranking function that uses the learned past and current topic preferences in order to rank the search results to better match the preferences of a given user. Our experimental evaluation on the Google query-stream of human subjects over a period of 1 month shows that user preferences can be learned accurately through the use of our topical ontology and that our ranking function which takes into account the learned user preferences yields significant improvements in the quality of the search results.     

Regularized query classification using search click information

Hundreds of millions of users each day submit queries to the Web search engine. The user queries are typically very short which makes query understanding a challenging problem. In this paper, we propose a novel approach for query representation and classification. By submitting the query to a web search engine, the query can be represented as a set of terms found on the web pages returned by search engine. In this way, each query can be considered as a point in high-dimensional space and standard classification algorithms such as regression can be applied. However, traditional regression is too flexible in situations with large numbers of highly correlated predictor variables. It may suffer from the overfitting problem. By using search click information, the semantic relationship between queries can be incorporated into the learning system as a regularizer. Specifically, from all the functions which minimize the empirical loss on the labeled queries, we select the one which best preserves the semantic relationship between queries. We present experimental evidence suggesting that the regularized regression algorithm is able to use search click information effectively for query classification.     

Web search results caching service for structured P2P networks

This paper proposes a two-level P2P caching strategy for Web search queries. The design is suitable for a fully distributed service platform based on managed peer boxes (set-top-box or DSL/cable modem) located at the edge of the network, where both boxes and access bandwidth to those boxes are controlled and managed by an ISP provider. Our solution significantly reduces user query traffic going outside of the ISP provider to get query results from the respective Web search engine. Web users are usually very reactive to worldwide events which cause highly dynamic query traffic patterns leading to load imbalance across peers. Our solution contains a strategy to quickly ease imbalance on peers and spread communication flow among participating peers. Each peer maintains a local result cache used to keep the answers for queries originated in the peer itself and queries for which the peer is responsible for by contacting the Web search engine on-demand. When query traffic is predominantly routed to a few responsible peers our strategy replicates the role of “being responsible for” to neighboring peers so that they can absorb query traffic. This is a fairly slow and adaptive process that we call mid-term load balancing. To achieve a short-term fair distribution of queries we introduce a location cache in each peer which keeps pointers to peers that have already requested the same queries in the recent past. This lets these peers share their query answers with newly requesting peers. This process is fast as these popular queries are usually cached in the first DHT hop of a requesting peer which quickly tends to redistribute load among more and more peers.     

An analysis of web proxy logs with query distribution pattern approach for search engines

This study presents an analysis of users' queries directed at different search engines to investigate trends and suggest better search engine capabilities. The query distribution among search engines that includes spawning of queries, number of terms per query and query lengths is discussed to highlight the principal factors affecting a user's choice of search engines and evaluate the reasons of varying the length of queries. The results could be used to develop long to short term business plans for search engine service providers to determine whether or not to opt for more focused topic specific search offerings to gain better market share.     

Mining search engine query logs for social filtering-based query recommendation

This paper presents a simple and intuitive method for mining search engine query logs for fast social filtering, where searchers are provided with dynamic query recommendations on a large-scale industrial-strength search engine. We adopt a dynamic approach that is able to absorb new and recent trends in web usage trends on search engines, while forgetting outdated trends, thus adapting to dynamic changes in web user’s interests. In order to get well-rounded recommendations, we combine two methods: first, we model search engine users’ sequential search behavior, and interpret this consecutive search behavior as client-side query refinement, that should form the basis for the search engine’s own query refinement process. This query refinement process is exploited to learn useful information that helps generate related queries. Second, we combine this method with a traditional text or content based similarity method to compensate for the shortness of query sessions and sparsity of real query log data.     

Further Experiments on Collaborative Ranking in Community-Based Web Search

As the search engine arms-race continues, search engines are constantly looking for ways to improve the manner in which they respond to user queries. Given the vagueness of Web search queries, recent research has focused on ways to introduce context into the search process as a means of clarifying vague, under-specified or ambiguous query terms. In this paper we describe a novel approach to using context in Web search that seeks to personalize the results of a generic search engine for the needs of a specialist community of users. In particular we describe two separate evaluations in detail that demonstrate how the collaborative search method has the potential to deliver significant search performance benefits to end-users while avoiding many of the privacy and security concerns that are commonly associated with related personalization research.     

Reducing Query Latencies in Web Search Using Fine-Grained Parallelism

Semantic Web search is a new application of recent advances in information retrieval (IR), natural language processing, artificial intelligence, and other fields. The Powerset group in Microsoft develops a semantic search engine that aims to answer queries not only by matching keywords, but by actually matching meaning in queries to meaning in Web documents. Compared to typical keyword search, semantic search can pose additional engineering challenges for the back-end and infrastructure designs. Of these, the main challenge addressed in this paper is how to lower query latencies to acceptable, interactive levels. Index-based semantic search requires more data processing, such as numerous synonyms, hypernyms, multiple linguistic readings, and other semantic information, both on queries and in the index. In addition, some of the algorithms can be super-linear, such as matching co-references across a document. Consequently, many semantic queries can run significantly slower than the same keyword query. Users, however, have grown to expect Web search engines to provide near-instantaneous results, and a slow search engine could be deemed unusable even if it provides highly relevant results. It is therefore imperative for any search engine to meet its users’ interactivity expectations, or risk losing them. Our approach to tackle this challenge is to exploit data parallelism in slow search queries to reduce their latency in multi-core systems. Although all search engines are designed to exploit parallelism, at the single-node level this usually translates to throughput-oriented task parallelism. This paper focuses on the engineering of two latency-oriented approaches (coarse- and fine-grained) and compares them to the task-parallel approach. We use Powerset’s deployed search engine to evaluate the various factors that affect parallel performance: workload, overhead, load balancing, and resource contention. We also discuss heuristics to selectively control the degree of parallelism and consequent overhead on a query-by-query level. Our experimental results show that using fine-grained parallelism with these dynamic heuristics can significantly reduce query latencies compared to fixed, coarse-granularity parallelization schemes. Although these results were obtained on, and optimized for, Powerset’s semantic search, they can be readily generalized to a wide class of inverted-index search engines.