Document replication strategies for geographically distributed web search engines

Large-scale web search engines are composed of multiple data centers that are geographically distant to each other. Typically, a user query is processed in a data center that is geographically close to the origin of the query, over a replica of the entire web index. Compared to a centralized, single-center search engine, this architecture offers lower query response times as the network latencies between the users and data centers are reduced. However, it does not scale well with increasing index sizes and query traffic volumes because queries are evaluated on the entire web index, which has to be replicated and maintained in all data centers. As a remedy to this scalability problem, we propose a document replication framework in which documents are selectively replicated on data centers based on regional user interests. Within this framework, we propose three different document replication strategies, each optimizing a different objective: reducing the potential search quality loss, the average query response time, or the total query workload of the search system. For all three strategies, we consider two alternative types of capacity constraints on index sizes of data centers. Moreover, we investigate the performance impact of query forwarding and result caching. We evaluate our strategies via detailed simulations, using a large query log and a document collection obtained from the Yahoo! web search engine.     

Mimicking Web search engines for expert search

Many enterprise employees may publish content outside their corporate intranet, making the Web a valuable source for identifying company experts. In this article, we thoroughly investigate the usefulness of Web search engines (WSEs) for expert search. In particular, we claim that the ranking of documentary expertise evidence provided by a WSE should also give an indication of the importance of such evidence. To investigate this, we mimic the rankings of seven different WSEs by trying to reproduce their underlying ranking mechanisms in order to search for candidate experts in the TREC CERC collection. Experimental results show that our approach is effective for expert search, and can significantly improve an intranet-based expert search engine. Moreover, when the mimicking of WSEs is further improved by training, expert search performance is also generally enhanced. Finally, we show that WSEs can be mimicked as effectively using only titles and snippets instead of the full content of WSEs’ results, while drastically reducing network costs.     

Distributed search based on self-indexed compressed text

Query response times within a fraction of a second in Web search engines are feasible due to the use of indexing and caching techniques, which are devised for large text collections partitioned and replicated into a set of distributed-memory processors. This paper proposes an alternative query processing method for this setting, which is based on a combination of self-indexed compressed text and posting lists caching. We show that a text self-index (i.e., an index that compresses the text and is able to extract arbitrary parts of it) can be competitive with an inverted index if we consider the whole query process, which includes index decompression, ranking and snippet extraction time. The advantage is that within the space of the compressed document collection, one can carry out the posting lists generation, document ranking and snippet extraction. This significantly reduces the total number of processors involved in the solution of queries. Alternatively, for the same amount of hardware, the performance of the proposed strategy is better than that of the classical approach based on treating inverted indexes and corresponding documents as two separate entities in terms of processors and memory space.     

The effectiveness of Web search engines for retrieving relevant ecommerce links

Ecommerce is developing into a fast-growing channel for new business, so a strong presence in this domain could prove essential to the success of numerous commercial organizations. However, there is little research examining ecommerce at the individual customer level, particularly on the success of everyday ecommerce searches. This is critical for the continued success of online commerce. The purpose of this research is to evaluate the effectiveness of search engines in the retrieval of relevant ecommerce links. The study examines the effectiveness of five different types of search engines in response to ecommerce queries by comparing the engines’ quality of ecommerce links using topical relevancy ratings. This research employs 100 ecommerce queries, five major search engines, and more than 3540 Web links. The findings indicate that links retrieved using an ecommerce search engine are significantly better than those obtained from most other engines types but do not significantly differ from links obtained from a Web directory service. We discuss the implications for Web system design and ecommerce marketing campaigns.     

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.     

Using a new relational concept to improve the clustering performance of search engines

In this paper, we present a novel clustering algorithm to generate a number of candidate clusters from other web search results. The candidate clusters generate a connective relation among the clusters and the relation is semantic. Moreover, the algorithm also contains the following attractive properties: (1) it can be applied to multilingual web documents, (2) it improves the clustering performance of any search engine, (3) its unsupervised learning can automatically identify potentially relevant knowledge without using any corpus, and (4) clustering results are generated on the fly and fitted into search engines.     

Social summarization in collaborative web search

A critical challenge for Web search engines concerns how they present relevant results to searchers. The traditional approach is to produce a ranked list of results with title and summary (snippet) information, and these snippets are usually chosen based on the current query. Snippets play a vital sensemaking role, helping searchers to efficiently make sense of a collection of search results, as well as determine the likely relevance of individual results. Recently researchers have begun to explore how snippets might also be adapted based on searcher preferences as a way to better highlight relevant results to the searcher. In this paper we focus on the role of snippets in collaborative web search and describe a technique for summarizing search results that harnesses the collaborative search behaviour of communities of like-minded searchers to produce snippets that are more focused on the preferences of the searchers. We go on to show how this so-called social summarization technique can generate summaries that are significantly better adapted to searcher preferences and describe a novel personalized search interface that combines result recommendation with social summarization.     

How are we searching the World Wide Web? A comparison of nine search engine transaction logs

The Web and especially major Web search engines are essential tools in the quest to locate online information for many people. This paper reports results from research that examines characteristics and changes in Web searching from nine studies of five Web search engines based in the US and Europe. We compare interactions occurring between users and Web search engines from the perspectives of session length, query length, query complexity, and content viewed among the Web search engines. The results of our research shows (1) users are viewing fewer result pages, (2) searchers on US-based Web search engines use more query operators than searchers on European-based search engines, (3) there are statistically significant differences in the use of Boolean operators and result pages viewed, and (4) one cannot necessary apply results from studies of one particular Web search engine to another Web search engine. The wide spread use of Web search engines, employment of simple queries, and decreased viewing of result pages may have resulted from algorithmic enhancements by Web search engine companies. We discuss the implications of the findings for the development of Web search engines and design of online content.     

A generic construct based workload model for web search

Benchmarks are vital tools in the performance measurement, evaluation, and comparison of computer hardware and software systems. Standard benchmarks such as the TREC, TPC, SPEC, SAP, Oracle, Microsoft, IBM, Wisconsin, AS³AP, OO1, OO7, XOO7 benchmarks have been used to assess the system performance. These benchmarks are domain-specific and domain-dependent in that they model typical applications and tie to a problem domain. Test results from these benchmarks are estimates of possible system performance for certain pre-determined problem types. When the user domain differs from the standard problem domain or when the application workload is divergent from the standard workload, they do not provide an accurate way to measure the system performance of the user problem domain. System performance of the actual problem domain in terms of data and transactions may vary significantly from the standard benchmarks.In this research, we address the issue of generalization and precision of benchmark workload model for web search technology. The current performance measurement and evaluation method suffers from the rough estimate of system performance which varies widely when the problem domain changes. The performance results provided by the vendors cannot be reproduced nor reused in the real users’ environment. Hence, in this research, we tackle the issue of domain boundness and workload boundness which represents the root of the problem of imprecise, ir-representative, and ir-reproducible performance results. We address the issue by presenting a domain-independent and workload-independent workload model benchmark method which is developed from the perspective of the user requirements and generic constructs. We present a user-driven workload model to develop a benchmark in a process of workload requirements representation, transformation, and generation via the common carrier of generic constructs. We aim to create a more generalized and precise evaluation method which derives test suites from the actual user domain and application setting.The workload model benchmark method comprises three main components. They are a high-level workload specification scheme, a translator of the scheme, and a set of generators to generate the test database and the test suite. They are based on the generic constructs. The specification scheme is used to formalize the workload requirements. The translator is used to transform the specification. The generator is used to produce the test database and the test workload. We determine the generic constructs via the analysis of search methods. The generic constructs form a page model, a query model, and a control model in the workload model development. The page model describes the web page structure. The query model defines the logics to query the web. The control model defines the control variables to set up the experiments.In this study, we have conducted ten baseline research experiments to validate the feasibility and validity of the benchmark method. An experimental prototype is built to execute these experiments. Experimental results demonstrate that the method based on generic constructs and driven by the perspective of user requirements is capable of modeling the standard benchmarks as well as more general benchmark requirements.     

Analysis of multiple query reformulations on the web: The interactive information retrieval context

This study examines the facets and patterns of multiple Web query reformulations with a focus on reformulation sequences. Based on IR interaction models, it was presumed that query reformulation is the product of the interaction between the user and the IR system. Query reformulation also reflects the interplay between the surface and deeper levels of user interaction. Query logs were collected from a Web search engine through the selection of search sessions in which users submitted six or more unique queries per session. The final data set was composed of 313 search sessions. Three facets of query reformulation (content, format, and resource) as well as nine sub-facets were derived from the data. In addition, analysis of modification sequences identified eight distinct patterns: specified, generalized, parallel, building-block, dynamic, multitasking, recurrent, and format reformulation. Adapting Saracevic’s stratified model, the authors develop a model of Web query reformulation based on the results of the study. The implications for Web search engine design are finally discussed and the functions of an interactive reformulation tool are suggested.     

Many are Better than One: Algorithm Selection for Faster Top-K Retrieval

Large-scale search engines have become a fundamental tool to efficiently access information on the Web. Typically, users expect answers in sub-second time frames, which demands highly efficient algorithms to traverse the data structures to return the top-k results. Despite different top-k algorithms that avoid processing all postings for all query terms, finding one algorithm that performs the fastest on any query is not always possible. The fastest average algorithm does not necessarily perform the best on all queries when evaluated on a per-query basis. To overcome this challenge, we propose to combine different state-of-the-art disjunctive top-k query processing algorithms to minimize the execution time by selecting the most promising one for each query. We model the selection step as a classification problem in a machine-learning setup. We conduct extensive experimentation and compare the results against state-of-the-art baselines using standard document collections and query sets. On ClueWeb12, our proposal shows a speed-up of up to 1.20x for non-blocked index organizations and 1.19x for block-based ones. Moreover, tail latencies are reduced showing proportional improvements on average, but a resulting dramatic decrease in latency variance. Given these findings, the proposed approach can be easily applied to existing search infrastructures to speed up query processing and reduce resource consumption, positively impacting providers’ operative costs.     

Music search engines: Specifications and challenges

Nowadays we have a proliferation of music data available over the Web. One of the imperative challenges is how to search these vast, global-scale musical resources to find preferred music. Recent research has envisaged the notion of music search engines (MSEs) that allow for searching preferred music over the Web. In this paper, we examine the growing research topic of MSEs, and provide potential specifications to follow and challenges to face.     

网络检索工具性能评价标准浅议

本文根据当前网络检索工具的新变化、新特点，提出了一套对网络检索工具进行性能评价与比较的标准，该标准由5个方面组成，分别是:索引 数据库构成（如，标引深度、更新频率）、检索功能（基本检索功能与高级检索功能）、检索效果（响应时间、查全率R、查准率P、重复率R、死链接率Rd）、检索结果显示（结果显示格式的种类与内容、相关性排序依据等）以及用户负担（用户界面、相关文献、信息过滤功能等），其中重复率Rr、互链接率Rd是作者根据需要提出的两个新的定量评价指标。     

SSTS: A syntactic tool for pattern search on time series

Nowadays, data scientists are capable of manipulating and extracting complex information from time series data, given the current diversity of tools at their disposal. However, the plethora of tools that target data exploration and pattern search may require an extensive amount of time to develop methods that correspond to the data scientist's reasoning, in order to solve their queries. The development of new methods, tightly related with the reasoning and visual analysis of time series data, is of great relevance to improving complexity and productivity of pattern and query search tasks. In this work, we propose a novel tool, capable of exploring time series data for pattern and query search tasks in a set of 3 symbolic steps: Pre-Processing, Symbolic Connotation and Search. The framework is called SSTS (Symbolic Search in Time Series) and uses regular expression queries to search the desired patterns in a symbolic representation of the signal. By adopting a set of symbolic methods, this approach has the purpose of increasing the expressiveness in solving standard pattern and query tasks, enabling the creation of queries more closely related to the reasoning and visual analysis of the signal. We demonstrate the tool's effectiveness by presenting 9 examples with several types of queries on time series. The SSTS queries were compared with standard code developed in Python, in terms of cognitive effort, vocabulary required, code length, volume, interpretation and difficulty metrics based on the Halstead complexity measures. The results demonstrate that this methodology is a valid approach and delivers a new abstraction layer on data analysis of time series.     

Analyzing imbalance among homogeneous index servers in a web search system

The performance of parallel query processing in a cluster of index servers is crucial for modern web search systems. In such a scenario, the response time basically depends on the execution time of the slowest server to generate a partial ranked answer. Previous approaches investigate performance issues in this context using simulation, analytical modeling, experimentation, or a combination of them. Nevertheless, these approaches simply assume balanced execution times among homogeneous servers (by uniformly distributing the document collection among them, for instance)—a scenario that we did not observe in our experimentation. On the contrary, we found that even with a balanced distribution of the document collection among index servers, correlations between the frequency of a term in the query log and the size of its corresponding inverted list lead to imbalances in query execution times at these same servers, because these correlations affect disk caching behavior. Further, the relative sizes of the main memory at each server (with regard to disk space usage) and the number of servers participating in the parallel query processing also affect imbalance of local query execution times. These are relevant findings that have not been reported before and that, we understand, are of interest to the research community.     

An alternative approach to natural language query expansion in search engines: Text analysis of non-topical terms in Web documents

This paper presents a new approach to query expansion in search engines through the use of general non-topical terms (NTTs) and domain-specific semi-topical terms (STTs). NTTs and STTs can be used in conjunction with topical terms (TTs) to improve precision in retrieval results. In Phase I, 20 topical queries in two domains (Health and the Social Sciences) were carried out in Google and from the results of the queries, 800 pages were textually analysed. Of 1442 NTTs and STTs identified, 15% were shared between the two domains; 62% were NTTs and 38% were STTs; and approximately 64% occurred before while 36% occurred after their respective topical terms (TTs). Findings of Phase II showed that query expansion through NTTs (or STTs) particularly in the ‘exact title’ and URL search options resulted in more precise and manageable results. Statistically significant differences were found between Health and the Social Sciences vis-à-vis keyword and ‘exact phrase’ search results; however there were no significant differences in exact title and URL search results. The ratio of exact phrase, exact title, and URL search result frequencies to keyword search result frequencies also showed statistically significant differences between the two domains. Our findings suggest that web searching could be greatly enhanced combining NTTs (and STTs) with TTs in an initial query. Additionally, search results would improve if queries are restricted to the exact title or URL search options. Finally, we suggest the development and implementation of knowledge-based lists of NTTs (and STTs) by both general and specialized search engines to aid query expansion.     

A study of results overlap and uniqueness among major Web search engines

The performance and capabilities of Web search engines is an important and significant area of research. Millions of people world wide use Web search engines very day. This paper reports the results of a major study examining the overlap among results retrieved by multiple Web search engines for a large set of more than 10,000 queries. Previous smaller studies have discussed a lack of overlap in results returned by Web search engines for the same queries. The goal of the current study was to conduct a large-scale study to measure the overlap of search results on the first result page (both non-sponsored and sponsored) across the four most popular Web search engines, at specific points in time using a large number of queries. The Web search engines included in the study were MSN Search, Google, Yahoo! and Ask Jeeves. Our study then compares these results with the first page results retrieved for the same queries by the metasearch engine Dogpile.com. Two sets of randomly selected user-entered queries, one set was 10,316 queries and the other 12,570 queries, from Infospace’s Dogpile.com search engine (the first set was from Dogpile, the second was from across the Infospace Network of search properties were submitted to the four single Web search engines). Findings show that the percent of total results unique to only one of the four Web search engines was 84.9%, shared by two of the three Web search engines was 11.4%, shared by three of the Web search engines was 2.6%, and shared by all four Web search engines was 1.1%. This small degree of overlap shows the significant difference in the way major Web search engines retrieve and rank results in response to given queries. Results point to the value of metasearch engines in Web retrieval to overcome the biases of individual search engines.     

Users can change their web search tactics: Design guidelines for categorized overviews

Categorized overviews of web search results are a promising way to support user exploration, understanding, and discovery. These search interfaces combine a metadata-based overview with the list of search results to enable a rich form of interaction. A study of 24 sophisticated users carrying out complex tasks suggests how searchers may adapt their search tactics when using categorized overviews. This mixed methods study evaluated categorized overviews of web search results organized into thematic, geographic, and government categories. Participants conducted four exploratory searches during a 2-hour session to generate ideas for newspaper articles about specified topics such as “human smuggling.” Results showed that subjects explored deeper while feeling more organized, and that the categorized overview helped subjects better assess their results, although no significant differences were detected in the quality of the article ideas. A qualitative analysis of searcher comments identified seven tactics that participants reported adopting when using categorized overviews. This paper concludes by proposing a set of guidelines for the design of exploratory search interfaces. An understanding of the impact of categorized overviews on search tactics will be useful to web search researchers, search interface designers, information architects and web developers.     

Automatic suggestion of phrasal-concept queries for literature search

Both general and domain-specific search engines have adopted query suggestion techniques to help users formulate effective queries. In the specific domain of literature search (e.g., finding academic papers), the initial queries are usually based on a draft paper or abstract, rather than short lists of keywords. In this paper, we investigate phrasal-concept query suggestions for literature search. These suggestions explicitly specify important phrasal concepts related to an initial detailed query. The merits of phrasal-concept query suggestions for this domain are their readability and retrieval effectiveness: (1) phrasal concepts are natural for academic authors because of their frequent use of terminology and subject-specific phrases and (2) academic papers describe their key ideas via these subject-specific phrases, and thus phrasal concepts can be used effectively to find those papers. We propose a novel phrasal-concept query suggestion technique that generates queries by identifying key phrasal-concepts from pseudo-labeled documents and combines them with related phrases. Our proposed technique is evaluated in terms of both user preference and retrieval effectiveness. We conduct user experiments to verify a preference for our approach, in comparison to baseline query suggestion methods, and demonstrate the effectiveness of the technique with retrieval experiments.     

Evaluating WordBars in exploratory Web search scenarios

Web searchers commonly have difficulties crafting queries to fulfill their information needs; even after they are able to craft a query, they often find it challenging to evaluate the results of their Web searches. Sources of these problems include the lack of support for constructing and refining queries, and the static nature of the list-based representations of Web search results. WordBars has been developed to assist users in their Web search and exploration tasks. This system provides a visual representation of the frequencies of the terms found in the first 100 document surrogates returned from an initial query, in the form of a histogram. Exploration of the search results is supported through term selection in the histogram, resulting in a re-sorting of the search results based on the use of the selected terms in the document surrogates. Terms from the histogram can be easily added or removed from the query, generating a new set of search results. Examples illustrate how WordBars can provide valuable support for query refinement and search results exploration, both when vague and specific initial queries are provided. User evaluations with both expert and intermediate Web searchers illustrate the benefits of the interactive exploration features of WordBars in terms of effectiveness as well as subjective measures. Although differences were found in the demographics of these two user groups, both were able to benefit from the features of WordBars.