Query Logs Research Articles

A framework for corroborating answers from multiple web sources

Search engines are increasingly efficient at identifying the best sources for any given keyword query, and are often able to identify the answer within the sources. Unfortunately, many web sources are not trustworthy, because of erroneous, misleading, biased, or outdated information. In many cases, users are not satisfied with the results from any single source. In this paper, we propose a framework to aggregate query results from different sources in order to save users the hassle of individually checking query-related web sites to corroborate answers. To return the best answers to the users, we assign a score to each individual answer by taking into account the number, relevance and originality of the sources reporting the answer, as well as the prominence of the answer within the sources, and aggregate the scores of similar answers. We conducted extensive qualitative and quantitative experiments of our corroboration techniques on queries extracted from the TREC Question Answering track and from a log of real web search engine queries. Our results show that taking into account the quality of web pages and answers extracted from the pages in a corroborative way results in the identification of a correct answer for a majority of queries.

Query Rewriting Using Monolingual Statistical Machine Translation

Long queries often suffer from low recall in Web search due to conjunctive term matching. The chances of matching words in relevant documents can be increased by rewriting query terms into new terms with similar statistical properties. We present a comparison of approaches that deploy user query logs to learn rewrites of query terms into terms from the document space. We show that the best results are achieved by adopting the perspective of bridging the “lexical chasm” between queries and documents by translating from a source language of user queries into a target language of Web documents. We train a state-of-the-art statistical machine translation model on query-snippet pairs from user query logs, and extract expansion terms from the query rewrites produced by the monolingual translation system. We show in an extrinsic evaluation in a real-world Web search task that the combination of a query-to-snippet translation model with a query language model achieves improved contextual query expansion compared to a state-of-the-art query expansion model that is trained on the same query log data.

ρ-uncertainty

The publication of transaction data, such as market basket data, medical records, and query logs, serves the public benefit. Mining such data allows for the derivation of association rules that connect certain items to others with measurable confidence. Still, this type of data analysis poses a privacy threat; an adversary having partial information on a person's behavior may confidently associate that person to an item deemed to be sensitive . Ideally, an anonymization of such data should lead to an inference-proof version that prevents the association of individuals to sensitive items, while otherwise allowing for truthful associations to be derived. Original approaches to this problem were based on value perturbation , damaging data integrity. Recently, value generalization has been proposed as an alternative; still, approaches based on it have assumed either that all items are equally sensitive, or that some are sensitive and can be known to an adversary only by association, while others are non-sensitive and can be known directly. Yet in reality there is a distinction between sensitive and non-sensitive items, but an adversary may possess information on any of them. Most critically, no antecedent method aims at a clear inference-proof privacy guarantee. In this paper, we propose ρ-uncertainty, the first , to our knowledge, privacy concept that inherently safeguards against sensitive associations without constraining the nature of an adversary's knowledge and without falsifying data. The problem of achieving ρ-uncertainty with low information loss is challenging because it is natural . A trivial solution is to suppress all sensitive items. We develop more sophisticated schemes. In a broad experimental study, we show that the problem is solved non-trivially by a technique that combines generalization and suppression, which also achieves favorable results compared to a baseline perturbation-based scheme.

Towards the web of concepts

Concepts are sequences of words that represent real or imaginary entities or ideas that users are interested in. As a first step towards building a web of concepts that will form the backbone of the next generation of search technology, we develop a novel technique to extract concepts from large datasets. We approach the problem of concept extraction from corpora as a market-basket problem, adapting statistical measures of support and confidence. We evaluate our concept extraction algorithm on datasets containing data from a large number of users (e.g., the AOL query log data set), and we show that a high-precision concept set can be extracted.

SQL QueRIE recommendations

This demonstration presents QueRIE, a recommender system that supports interactive database exploration. This system aims at assisting non-expert users of scientific databases by tracking their querying behavior and generating personalized query recommendations. The system is supported by two recommendation engines and the underlying recommendation algorithms. The first identifies potentially "interesting" parts of the database related to the corresponding data analysis task by locating those database parts that were accessed by similar users in the past. The second identifies structurally similar queries to the ones posted by the current user. Both approaches result in a recommendation set of SQL queries that is provided to the user to modify, or directly post to the database. The demonstrated system will enable users to query and get real-time recommendations from the SkyServer database, using user traces collected from the SkyServer query log.

Learning query ambiguity models by using search logs

Identifying ambiguous queries is crucial to research on personalized Web search and search result diversity. Intuitively, query logs contain valuable information on how many intentions users have when issuing a query. However, previous work showed user clicks alone are misleading in judging a query as being ambiguous or not. In this paper, we address the problem of learning a query ambiguity model by using search logs. First, we propose enriching a query by mining the documents clicked by users and the relevant follow up queries in a session. Second, we use a text classifier to map the documents and the queries into predefined categories. Third, we propose extracting features from the processed data. Finally, we apply a state-of-the-art algorithm, Support Vector Machine (SVM), to learn a query ambiguity classifier. Experimental results verify that the sole use of click based features or session based features perform worse than the previous work based on top retrieved documents. When we combine the two sets of features, our proposed approach achieves the best effectiveness, specifically 86% in terms of accuracy. It significantly improves the click based method by 5.6% and the session based method by 4.6%.

Privacy-preserving query log mining for business confidentiality protection

We introduce the concern of confidentiality protection of business information for the publication of search engine query logs and derived data. We study business confidentiality, as the protection of nonpublic data from institutions, such as companies and people in the public eye. In particular, we relate this concern to the involuntary exposure of confidential Web site information, and we transfer this problem into the field of privacy-preserving data mining. We characterize the possible adversaries interested in disclosing Web site confidential data and the attack strategies that they could use. These attacks are based on different vulnerabilities found in query log for which we present several anonymization heuristics to prevent them. We perform an experimental evaluation to estimate the remaining utility of the log after the application of our anonymization techniques. Our experimental results show that a query log can be anonymized against these specific attacks while retaining a significant volume of useful data.

Similarity joins as stronger metric operations

Similarity joins between two sets of records return pairs of records whose similarity is no less than a given threshold. More specifically, consider two sets of records, R and S , a similarity function sim (.,.) and a threshold t , a similarity join between R and S is defined as { ( r, s ) | ( r, s ) ∈ R x S, sim(r, s) ≥ t }. A similarity join is a generalization of the traditional equality join commonly found in database systems. A variant of the similarity join is to use a distance threshold to replace the similarity threshold. It is generally expected that the similarity threshold is close to the maximum possible value (usually 1.0), and the distance threshold is close to the minimum possible value (usually 0). For example, we may find near-duplicate documents in a document repository using a cosine similarity threshold of 0.9, or we may find pairs of incorrectly spelt queries and their correct versions in a query log with an edit distance threshold of 2.

TSS: Efficient Term Set Search in Large Peer-to-Peer Textual Collections

Previous multikeyword search in DHT-based P2P systems often relies on multiple single keyword search operations, suffering from unacceptable traffic cost and poor accuracy. Precomputing term-set-based index can significantly reduce the cost but needs exponentially growing index size. Based on our observations that 1) queries are typically short and 2) users usually have limited interests, we propose a novel index pruning method, called TSS. By solely publishing the most relevant term sets from documents on the peers, TSS provides comparable search performance with a centralized solution, while the index size is reduced from exponential to the scale of O(nlog(n)). We evaluate this design through comprehensive trace-driven simulations using the TREC WT10G data collection and the query log of a major commercial search engine.

Query Logs as Folksonomies

Query logs provide a valuable resource for preference information in search. A user clicking on a specific resource after submitting a query indicates that the resource has some relevance with respect to the query. To leverage the information of query logs, one can relate submitted queries from specific users to their clicked resources and build a tripartite graph of users, resources and queries. This graph resembles the folksonomy structure of social bookmarking systems, where users add tags to resources. In this article, we summarize our work on building folksonomies from query log files. The focus is on three comparative studies of the system’s content, structure and semantics. Our results show that query logs incorporate typical folksonomy properties and that approaches to leverage the inherent semantics of folksonomies can be applied to query logs as well.

Exploiting query logs for cross-lingual query suggestions

Query suggestion aims to suggest relevant queries for a given query, which helps users better specify their information needs. Previous work on query suggestion has been limited to the same language. In this article, we extend it to cross-lingual query suggestion (CLQS): for a query in one language, we suggest similar or relevant queries in other languages. This is very important to the scenarios of cross-language information retrieval (CLIR) and other related cross-lingual applications. Instead of relying on existing query translation technologies for CLQS, we present an effective means to map the input query of one language to queries of the other language in the query log. Important monolingual and cross-lingual information such as word translation relations and word co-occurrence statistics, and so on, are used to estimate the cross-lingual query similarity with a discriminative model. Benchmarks show that the resulting CLQS system significantly outperforms a baseline system that uses dictionary-based query translation. Besides, we evaluate CLQS with French-English and Chinese-English CLIR tasks on TREC-6 and NTCIR-4 collections, respectively. The CLIR experiments using typical retrieval models demonstrate that the CLQS-based approach has significantly higher effectiveness than several traditional query translation methods. We find that when combined with pseudo-relevance feedback, the effectiveness of CLIR using CLQS is enhanced for different pairs of languages.

Optimal distance bounds for fast search on compressed time-series query logs

Consider a database of time-series, where each datapoint in the series records the total number of users who asked for a specific query at an internet search engine. Storage and analysis of such logs can be very beneficial for a search company from multiple perspectives. First, from a data organization perspective, because query Weblogs capture important trends and statistics, they can help enhance and optimize the search experience (keyword recommendation, discovery of news events). Second, Weblog data can provide an important polling mechanism for the microeconomic aspects of a search engine, since they can facilitate and promote the advertising facet of the search engine (understand what users request and when they request it). Due to the sheer amount of time-series Weblogs, manipulation of the logs in a compressed form is an impeding necessity for fast data processing and compact storage requirements. Here, we explicate how to compute the lower and upper distance bounds on the time-series logs when working directly on their compressed form. Optimal distance estimation means tighter bounds, leading to better candidate selection/elimination and ultimately faster search performance. Our derivation of the optimal distance bounds is based on the careful analysis of the problem using optimization principles. The experimental evaluation suggests a clear performance advantage of the proposed method, compared to previous compression/search techniques. The presented method results in a 10--30% improvement on distance estimations, which in turn leads to 25--80% improvement on the search performance.

Ontology-Assisted Analysis of Web Queries to Determine the Knowledge Radiologists Seek

Radiologists frequently search the Web to find information they need to improve their practice, and knowing the types of information they seek could be useful for evaluating Web resources. Our goal was to develop an automated method to categorize unstructured user queries using a controlled terminology and to infer the type of information users seek. We obtained the query logs from two commonly used Web resources for radiology. We created a computer algorithm to associate RadLex-controlled vocabulary terms with the user queries. Using the RadLex hierarchy, we determined the high-level category associated with each RadLex term to infer the type of information users were seeking. To test the hypothesis that the term category assignments to user queries are non-random, we compared the distributions of the term categories in RadLex with those in user queries using the chi square test. Of the 29,669 unique search terms found in user queries, 15,445 (52%) could be mapped to one or more RadLex terms by our algorithm. Each query contained an average of one to two RadLex terms, and the dominant categories of RadLex terms in user queries were diseases and anatomy. While the same types of RadLex terms were predominant in both RadLex itself and user queries, the distribution of types of terms in user queries and RadLex were significantly different (p < 0.0001). We conclude that RadLex can enable processing and categorization of user queries of Web resources and enable understanding the types of information users seek from radiology knowledge resources on the Web.

Efficient successor retrieval operations for aggregate query processing on clustered road networks

Get-Successors (GS) which retrieves all successors of a junction is a kernel operation used to facilitate aggregate computations in road network queries. Efficient implementation of the GS operation is crucial since the disk access cost of this operation constitutes a considerable portion of the total query processing cost. Firstly, we propose a new successor retrieval operation Get-Unevaluated-Successors (GUS), which retrieves only the unevaluated successors of a given junction. The GUS operation is an efficient implementation of the GS operation, where the candidate successors to be retrieved are pruned according to the properties and state of the algorithm. Secondly, we propose a hypergraph-based model for clustering successively retrieved junctions by the GUS operations to the same pages. The proposed model utilizes query logs to correctly capture the disk access cost of GUS operations. The proposed GUS operation and associated clustering model are evaluated for two different instances of GUS operations which typically arise in Dijkstra's single source shortest path algorithm and incremental network expansion framework. Our simulation results show that the proposed successor retrieval operation together with the proposed clustering hypergraph model is quite effective in reducing the number of disk accesses in query processing.

국내 포털 이용자들의 멀티미디어 검색 행태 분석

멀티미디어 검색은 전통적인 검색과 웹 검색을 차별화시키는 중요한 요소 중 하나이다. 본 연구에서는 국내 주요 검색 포털인 네이버 이용자들의 멀티미디어 검색 행태를 분석, 평가하였다. 이를 위하여 이용자들이 입력한 통합 검색 질의들로 구성된 질의 로그와 질의에 대한 검색 결과에서 이용자들이 조회한 문서를 기록한 클릭 로그를 분석하였다. 연구 결과, 이용자가 입력하는 멀티미디어 질의 중 음악 질의가 차지하는 비중이 가장 높았으며, 이미지 질의와 동영상 질의의 비중은 비슷한 것으로 나타났다. 한편 멀티미디어 관련 컬렉션들 중 이용자가 가장 많이 조회하는 컬렉션은 동영상 컬렉션이었고, 이어서 이미지 컬렉션, 음악 컬렉션 순으로 나타났다. 이용자가 입력하는 멀티미디어 질의나 멀티미디어 컬렉션에서 클릭하는 문서 모두 엔터테인먼트와 관련된 주제가 가장 많았다. 본 연구의 결과는 향후 포털의 멀티미디어 컨텐츠 구축 및 검색 알고리즘 개발에 활용될 수 있을 것으로 기대된다.

Modeling temporal text streams using the local multinomial model

Temporal text data such as news feeds cannot be adequately modeled by standard n-grams which correspond to multinomial or Markov chain models. Instead, we examine the application of local n-grams to modeling time stamped documents. We derive the asymptotic bias and variance and consider the bandwidth selection problem. Experimental results are presented on news feeds and web search query logs.

Report on the TrebleCLEF query log analysis workshop 2009

This report summarises a workshop organised as a part of the EU-funded TrebleCLEF project entitled "Query Log Analysis: From Research to Best Practice" held on 27-28th May 2009 at the British Computer Science Offices in London, UK. The event involved 12 invited speakers from various academic and commercial institutions from around the world who are all involved, in some way, with query log analysis. A number of other people attended the event including local businesses and academic institutions. The workshop provided a forum in which to discuss and share experiences and best practices regarding query log analysis. This report describes the event and the presentations given by the invited speakers, and it summarises discussions held.

An experimental investigation of set intersection algorithms for text searching

The intersection of large ordered sets is a common problem in the context of the evaluation of boolean queries to a search engine. In this article, we propose several improved algorithms for computing the intersection of sorted arrays, and in particular for searching sorted arrays in the intersection context. We perform an experimental comparison with the algorithms from the previous studies from Demaine, López-Ortiz, and Munro [ALENEX 2001] and from Baeza-Yates and Salinger [SPIRE 2005]; in addition, we implement and test the intersection algorithm from Barbay and Kenyon [SODA 2002] and its randomized variant [SAGA 2003]. We consider both the random data set from Baeza-Yates and Salinger, the Google queries used by Demaine et al., a corpus provided by Google, and a larger corpus from the TREC Terabyte 2006 efficiency query stream, along with its own query log. We measure the performance both in terms of the number of comparisons and searches performed, and in terms of the CPU time on two different architectures. Our results confirm or improve the results from both previous studies in their respective context (comparison model on real data, and CPU measures on random data) and extend them to new contexts. In particular, we show that value-based search algorithms perform well in posting lists in terms of the number of comparisons performed.

Exploring features for the automatic identification of user goals in web search

Queries submitted to search engines can be classified according to the user goals into three distinct categories: navigational, informational, and transactional. Such classification may be useful, for instance, as additional information for advertisement selection algorithms and for search engine ranking functions, among other possible applications. This paper presents a study about the impact of using several features extracted from the document collection and query logs on the task of automatically identifying the users’ goals behind their queries. We propose the use of new features not previously reported in literature and study their impact on the quality of the query classification task. Further, we study the impact of each feature on different web collections, showing that the choice of the best set of features may change according to the target collection. The results obtained indicate the new proposed set of features improves the quality of the classification task when compared to previous proposals. We report experiments with two web collections where we were able to obtain 82.5% and 77.67% of overall accuracy when classifying queries according to the three distinct user goals studied.

Emergence of consensus and shared vocabularies in collaborative tagging systems

This article uses data from the social bookmarking site del.icio.us to empirically examine the dynamics of collaborative tagging systems and to study how coherent categorization schemes emerge from unsupervised tagging by individual users. First, we study the formation of stable distributions in tagging systems, seen as an implicit form of “consensus” reached by the users of the system around the tags that best describe a resource. We show that final tag frequencies for most resources converge to power law distributions and we propose an empirical method to examine the dynamics of the convergence process, based on the Kullback-Leibler divergence measure. The convergence analysis is performed for both the most utilized tags at the top of tag distributions and the so-called long tail. Second, we study the information structures that emerge from collaborative tagging, namely tag correlation (or folksonomy) graphs. We show how community-based network techniques can be used to extract simple tag vocabularies from the tag correlation graphs by partitioning them into subsets of related tags. Furthermore, we also show, for a specialized domain, that shared vocabularies produced by collaborative tagging are richer than the vocabularies which can be extracted from large-scale query logs provided by a major search engine. Although the empirical analysis presented in this article is based on a set of tagging data obtained from del.icio.us, the methods developed are general, and the conclusions should be applicable across other websites that employ tagging.