共查询到20条相似文献，搜索用时 31 毫秒
1.
Travis Gagie Aleksi Hartikainen Kalle Karhu Juha Kärkkäinen Gonzalo Navarro Simon J. Puglisi Jouni Sirén 《Information Retrieval》2017,20(3):253-291
Most of the fastest-growing string collections today are repetitive, that is, most of the constituent documents are similar to many others. As these collections keep growing, a key approach to handling them is to exploit their repetitiveness, which can reduce their space usage by orders of magnitude. We study the problem of indexing repetitive string collections in order to perform efficient document retrieval operations on them. Document retrieval problems are routinely solved by search engines on large natural language collections, but the techniques are less developed on generic string collections. The case of repetitive string collections is even less understood, and there are very few existing solutions. We develop two novel ideas, interleaved LCPs and precomputed document lists, that yield highly compressed indexes solving the problem of document listing (find all the documents where a string appears), top-k document retrieval (find the k documents where a string appears most often), and document counting (count the number of documents where a string appears). We also show that a classical data structure supporting the latter query becomes highly compressible on repetitive data. Finally, we show how the tools we developed can be combined to solve ranked conjunctive and disjunctive multi-term queries under the simple \({\textsf{tf}}{\textsf{-}}{\textsf{idf}}\) model of relevance. We thoroughly evaluate the resulting techniques in various real-life repetitiveness scenarios, and recommend the best choices for each case. 相似文献
2.
Caio Moura Daoud Edleno Silva de Moura David Fernandes Altigran Soares da Silva Cristian Rossi Andre Carvalho 《Information Retrieval》2017,20(3):292-316
In this paper, we present Waves, a novel document-at-a-time algorithm for fast computing of top-k query results in search systems. The Waves algorithm uses multi-tier indexes for processing queries. It performs successive tentative evaluations of results which we call waves. Each wave traverses the index, starting from a specific tier level i. Each wave i may insert only those documents that occur in that tier level into the answer. After processing a wave, the algorithm checks whether the answer achieved might be changed by successive waves or not. A new wave is started only if it has a chance of changing the top-k scores. We show through experiments that such lazy query processing strategy results in smaller query processing times when compared to previous approaches proposed in the literature. We present experiments to compare Waves’ performance to the state-of-the-art document-at-a-time query processing methods that preserve top-k results and show scenarios where the method can be a good alternative algorithm for computing top-k results. 相似文献
4.
The rapid growth of the Web has increased the difficulty of finding the information that can address the users’ information needs. A number of recommendation approaches have been developed to tackle this problem. The increase in the number of data providers has necessitated the development of multi-publisher recommender systems; systems that include more than one item/data provider. In such environments, preserving the privacy of both publishers and subscribers is a key and challenging point. In this paper, we propose a multi-publisher framework for recommender systems based on a client–server architecture, which preserves the privacy of both data providers and subscribers. We develop our framework as a content-based filtering system using the statistical language modeling framework. We also introduce AUTO, a simple yet effective threshold optimization algorithm, to find a dissemination threshold for making acceptance and rejection decisions for new published documents. We further propose a language model sketching technique to reduce the network traffic between servers and clients in the proposed framework. Extensive experiments using the TREC-9 Filtering Track and the CLEF 2008-09 INFILE Track collections indicate the effectiveness of the proposed models in both single- and multi-publisher settings. 相似文献
5.
Iain Stevenson 《Publishing Research Quarterly》2008,24(4):277-284
This paper reviews the current status of the Anglophone (Anglo-American) publishing business and draws some comparisons with publishing in other languages. It then critically reviews the impact of the Harry Potter phenomenon and the questionable progress of e-books in the trade sector, using the example of Stephen King’s Riding the Bullet. It also comments on Amazon’s introduction of the Kindle e-book reader. 相似文献
6.
This article introduces a new language-independent approach for creating a large-scale high-quality test collection of tweets that supports multiple information retrieval (IR) tasks without running a shared-task campaign. The adopted approach (demonstrated over Arabic tweets) designs the collection around significant (i.e., popular) events, which enables the development of topics that represent frequent information needs of Twitter users for which rich content exists. That inherently facilitates the support of multiple tasks that generally revolve around events, namely event detection, ad-hoc search, timeline generation, and real-time summarization. The key highlights of the approach include diversifying the judgment pool via interactive search and multiple manually-crafted queries per topic, collecting high-quality annotations via crowd-workers for relevancy and in-house annotators for novelty, filtering out low-agreement topics and inaccessible tweets, and providing multiple subsets of the collection for better availability. Applying our methodology on Arabic tweets resulted in EveTAR, the first freely-available tweet test collection for multiple IR tasks. EveTAR includes a crawl of 355M Arabic tweets and covers 50 significant events for which about 62K tweets were judged with substantial average inter-annotator agreement (Kappa value of 0.71). We demonstrate the usability of EveTAR by evaluating existing algorithms in the respective tasks. Results indicate that the new collection can support reliable ranking of IR systems that is comparable to similar TREC collections, while providing strong baseline results for future studies over Arabic tweets. 相似文献
7.
This study is devoted to detection of the lexical environment and demonstration of the thematic medium of the words MEMORY and MEMORIES in the social sciences on the basis of the bibliographic database Social Science Citation Index (SSCI) of the Institute for Scientific Information (USA). The amount of studied material is over 3000 documents in English. Corresponding corpora and subcorpora of summary texts are formed, general frequency dictionaries and frequency dictionaries of binary combinations for each corpus and subcorpus are constructed, words and combinations specific for each subcorpus are found, and corresponding factors (lexical markers) are calculated for them. The general statistical information on the usage of the words under study is given, the obtained results of lexical analysis are represented in a tabulated form, and the corresponding semantic maps are discussed. 相似文献
8.
Tetsuya Sakai 《Information Retrieval》2016,19(3):256-283
Traditional pooling-based information retrieval (IR) test collections typically have \(n= 50\)–100 topics, but it is difficult for an IR researcher to say why the topic set size should really be n. The present study provides details on principled ways to determine the number of topics for a test collection to be built, based on a specific set of statistical requirements. We employ Nagata’s three sample size design techniques, which are based on the paired t test, one-way ANOVA, and confidence intervals, respectively. These topic set size design methods require topic-by-run score matrices from past test collections for the purpose of estimating the within-system population variance for a particular evaluation measure. While the previous work of Sakai incorrectly used estimates of the total variances, here we use the correct estimates of the within-system variances, which yield slightly smaller topic set sizes than those reported previously by Sakai. Moreover, this study provides a comparison across the three methods. Our conclusions nevertheless echo those of Sakai: as different evaluation measures can have vastly different within-system variances, they require substantially different topic set sizes under the same set of statistical requirements; by analysing the tradeoff between the topic set size and the pool depth for a particular evaluation measure in advance, researchers can build statistically reliable yet highly economical test collections. 相似文献
9.
Xu Zhuang Yan Zhu Chin-Chen Chang Qiang Peng Faisal Khurshid 《Information Retrieval》2017,20(6):547-574
Web spam pages exploit the biases of search engine algorithms to get higher than their deserved rankings in search results by using several types of spamming techniques. Many web spam demotion algorithms have been developed to combat spam via the use of the web link structure, from which the goodness or badness score of each web page is evaluated. Those scores are then used to identify spam pages or punish their rankings in search engine results. However, most of the published spam demotion algorithms differ from their base models by only very limited improvements and still suffer from some common score manipulation methods. The lack of a general framework for this field makes the task of designing high-performance spam demotion algorithms very inefficient. In this paper, we propose a unified score propagation model for web spam demotion algorithms by abstracting the score propagation process of relevant models with a forward score propagation function and a backward score propagation function, each of which can further be expressed as three sub-functions: a splitting function, an accepting function and a combination function. On the basis of the proposed model, we develop two new web spam demotion algorithms named Supervised Forward and Backward score Ranking (SFBR) and Unsupervised Forward and Backward score Ranking (UFBR). Our experiments, conducted on three large-scale public datasets, show that (1) SFBR is very robust and apparently outperforms other algorithms and (2) UFBR can obtain results comparable to some well-known supervised algorithms in the spam demotion task even if the UFBR is unsupervised. 相似文献
10.
Gaurav Pandey Zhaochun Ren Shuaiqiang Wang Jari Veijalainen Maarten de Rijke 《Information Retrieval》2018,21(6):481-506
We address the feature extraction problem for document ranking in information retrieval. We then propose LifeRank, a Linear feature extraction algorithm for Ranking. In LifeRank, we regard each document collection for ranking as a matrix, referred to as the original matrix. We try to optimize a transformation matrix, so that a new matrix (dataset) can be generated as the product of the original matrix and a transformation matrix. The transformation matrix projects high-dimensional document vectors into lower dimensions. Theoretically, there could be very large transformation matrices, each leading to a new generated matrix. In LifeRank, we produce a transformation matrix so that the generated new matrix can match the learning to rank problem. Extensive experiments on benchmark datasets show the performance gains of LifeRank in comparison with state-of-the-art feature selection algorithms. 相似文献
11.
In 2004, the Scottish Parliament commissioned an independent review of abuse in children’s residential establishments between 1950 and 1995. In 2007, the review’s findings were published in a report entitled Historical Abuse Systemic Review: Residential Schools and Children’s Homes in Scotland 1950 to 1995, also known as the Shaw Report. In this article, the Shaw Report provides the jumping off point for a case study of the social justice impact of records. Drawing on secondary literature, interviews, and care-related records, the study identifies narratives that speak to the social justice impact of care records on care-leavers seeking access to them; it also assesses the potential of the surviving administrative records to serve as a foundation on which to construct historical narratives that speak more generally to the experience of children in residential care. 相似文献
12.
Qinglei Wang Yanan Qian Ruihua Song Zhicheng Dou Fan Zhang Tetsuya Sakai Qinghua Zheng 《Information Retrieval》2013,16(4):484-503
Web search queries are often ambiguous or faceted, and the task of identifying the major underlying senses and facets of queries has received much attention in recent years. We refer to this task as query subtopic mining. In this paper, we propose to use surrounding text of query terms in top retrieved documents to mine subtopics and rank them. We first extract text fragments containing query terms from different parts of documents. Then we group similar text fragments into clusters and generate a readable subtopic for each cluster. Based on the cluster and the language model trained from a query log, we calculate three features and combine them into a relevance score for each subtopic. Subtopics are finally ranked by balancing relevance and novelty. Our evaluation experiments with the NTCIR-9 INTENT Chinese Subtopic Mining test collection show that our method significantly outperforms a query log based method proposed by Radlinski et al. (2010) and a search result clustering based method proposed by Zeng et al. (2004) in terms of precision, I-rec, D-nDCG and D#-nDCG, the official evaluation metrics used at the NTCIR-9 INTENT task. Moreover, our generated subtopics are significantly more readable than those generated by the search result clustering method. 相似文献
13.
Automatic detection of source code plagiarism is an important research field for both the commercial software industry and within the research community. Existing methods of plagiarism detection primarily involve exhaustive pairwise document comparison, which does not scale well for large software collections. To achieve scalability, we approach the problem from an information retrieval (IR) perspective. We retrieve a ranked list of candidate documents in response to a pseudo-query representation constructed from each source code document in the collection. The challenge in source code document retrieval is that the standard bag-of-words (BoW) representation model for such documents is likely to result in many false positives being retrieved, because of the use of identical programming language specific constructs and keywords. To address this problem, we make use of an abstract syntax tree (AST) representation of the source code documents. While the IR approach is efficient, it is essentially unsupervised in nature. To further improve its effectiveness, we apply a supervised classifier (pre-trained with features extracted from sample plagiarized source code pairs) on the top ranked retrieved documents. We report experiments on the SOCO-2014 dataset comprising 12K Java source files with almost 1M lines of code. Our experiments confirm that the AST based approach produces significantly better retrieval effectiveness than a standard BoW representation, i.e., the AST based approach is able to identify a higher number of plagiarized source code documents at top ranks in response to a query source code document. The supervised classifier, trained on features extracted from sample plagiarized source code pairs, is shown to effectively filter and thus further improve the ranked list of retrieved candidate plagiarized documents. 相似文献
14.
T. N. Domnina 《Scientific and Technical Information Processing》2016,43(4):241-250
The megajournal is a new form of an academic open-access journal that is peer reviewed for scientific and methodological soundness. Thus, the perceived importance is not assessed, leaving it to the readers to decide whether an article is of interest and importance to them. After the highly successful launch of the PLOS ONE megajournal, other publishers followed this path. The current paper undertakes a review of the articles published internationally on the megajournal criteria defining its concept. It also traces how the new type of scientific publication has been developing and expanding since the PLoS ONE launch in 2006. 相似文献
15.
Richard J. Matthews 《Archival Science》2016,16(3):213-260
Hägglund’s “radical atheism”—innovative thinking within the philosophical current of “speculative materialism”—revitalizes deconstruction and provides an important basis to define parameters for the archivist’s role as activist for social justice. This paper argues postmodern archival theory gets deconstruction wrong by misreading Derrida’s “Archive fever” as a theory of “archontic power”; this misleads archivists on the call for justice. Properly understanding that justice is undecidable, radical atheism explodes the tension between postmodernists’ appreciation of all views and perspectives and their commitment to right unjust relations of power. This paper first advances the negative argument that “Archive fever” is not about power and injustice. It then advances the positive argument that “Archive fever” is Derrida’s effort to look at actual archives to resolve Freud’s problematic theorizing of a “death drive.” In a close and comprehensive reading of “Archive fever,” this paper explores the notion of “archive fever” as a death drive and suggests Derrida’s efforts are inconclusive. Viewed through the lens of radical atheism, the archive’s “traces”—the material of actual archives writ large in the manner of Derrida’s thinking about a universal archive—serve to mark the flow of time. Understanding the structure of the trace reveals the source of internal contradictions, discontinuities, and instabilities in the meaning of all things. It explains why justice is undecidable. In face of the unconditional condition of this undecidability, we as archivists and humans are compelled to make decisions and to act. Deconstruction politicizes our actions and evokes a responsibility that cannot be absolved. 相似文献
16.
Ilaria Bordino Mounia Lalmas Yelena Mejova Olivier Van Laere 《Information Retrieval》2016,19(5):447-486
Search engines are increasingly going beyond the pure relevance of search results to entertain users with information items that are interesting and even surprising, albeit sometimes not fully related to their search intent. In this paper, we study this serendipitous search space in the context of entity search, which has recently emerged as a powerful paradigm for building semantically rich answers. Specifically, our work proposes to enhance an explorative search system that represents a large sample of Yahoo Answers as an entity network, with a result structuring that goes beyond ranked lists, using composite entity retrieval, which requires a bundling of the results. We propose and compare six bundling methods, which exploit topical categories, entity specializations, and sentiment, and go beyond simple entity clustering. Two large-scale crowd-sourced studies show that users find a bundled organization—especially based on the topical categories of the query entity—to be better at revealing the most useful results, as well as at organizing the results, helping to discover novel and interesting information, and promoting exploration. Finally, a third study of 30 simulated search tasks reveals the bundled search experience to be less frustrating and more rewarding, with more users willing to recommend it to others. 相似文献
17.
Casey Brienza 《Publishing Research Quarterly》2017,33(4):373-379
In this paper, I will argue for key role played by the global culture of comics, of which the French publication Charlie Hebdo is but one small part, in the development and aftermath of the Paris terrorist attacks of 2015. I begin by exploring and elucidating this culture of comics that exists across France and other comic book producing nations and its associations with youthful rebellion, anarchy, and, more recently, misrecognized privilege and bigotry. A sociological perspective on the cultural history and practice of comics publishing through the twentieth and twenty-first centuries adds important context to the Je Suis Charlie movement and its place in the public discourse—while also further problematizing it. 相似文献
18.
Word embeddings and convolutional neural networks (CNN) have attracted extensive attention in various classification tasks for Twitter, e.g. sentiment classification. However, the effect of the configuration used to generate the word embeddings on the classification performance has not been studied in the existing literature. In this paper, using a Twitter election classification task that aims to detect election-related tweets, we investigate the impact of the background dataset used to train the embedding models, as well as the parameters of the word embedding training process, namely the context window size, the dimensionality and the number of negative samples, on the attained classification performance. By comparing the classification results of word embedding models that have been trained using different background corpora (e.g. Wikipedia articles and Twitter microposts), we show that the background data should align with the Twitter classification dataset both in data type and time period to achieve significantly better performance compared to baselines such as SVM with TF-IDF. Moreover, by evaluating the results of word embedding models trained using various context window sizes and dimensionalities, we find that large context window and dimension sizes are preferable to improve the performance. However, the number of negative samples parameter does not significantly affect the performance of the CNN classifiers. Our experimental results also show that choosing the correct word embedding model for use with CNN leads to statistically significant improvements over various baselines such as random, SVM with TF-IDF and SVM with word embeddings. Finally, for out-of-vocabulary (OOV) words that are not available in the learned word embedding models, we show that a simple OOV strategy to randomly initialise the OOV words without any prior knowledge is sufficient to attain a good classification performance among the current OOV strategies (e.g. a random initialisation using statistics of the pre-trained word embedding models). 相似文献
19.
Kezban Dilek Onal Ye Zhang Ismail Sengor Altingovde Md Mustafizur Rahman Pinar Karagoz Alex Braylan Brandon Dang Heng-Lu Chang Henna Kim Quinten McNamara Aaron Angert Edward Banner Vivek Khetan Tyler McDonnell An Thanh Nguyen Dan Xu Byron C. Wallace Maarten de Rijke Matthew Lease 《Information Retrieval》2018,21(2-3):111-182
A recent “third wave” of neural network (NN) approaches now delivers state-of-the-art performance in many machine learning tasks, spanning speech recognition, computer vision, and natural language processing. Because these modern NNs often comprise multiple interconnected layers, work in this area is often referred to as deep learning. Recent years have witnessed an explosive growth of research into NN-based approaches to information retrieval (IR). A significant body of work has now been created. In this paper, we survey the current landscape of Neural IR research, paying special attention to the use of learned distributed representations of textual units. We highlight the successes of neural IR thus far, catalog obstacles to its wider adoption, and suggest potentially promising directions for future research. 相似文献
20.
Ying Peng 《Publishing Research Quarterly》2016,32(3):247-260
This paper provides an overview of the e-book publishing industry in China and its changing business models. As the conventional publishing industry has continued to contract, digitization is becoming increasingly popular. Although China is an influential developing country undergoing rapid change, a dearth of research on its digital publishing industry exists at the present time. This article therefore explores two main research questions: (1) What are the recent trends and business models of digital publishing in China? (2) How might we evaluate the effectiveness and efficiency of such business models? To this end, multiple case studies are employed, as is the paradigm of structure-conduct-performance (SCP). It is proposed that Content plus App and Content plus Device are the basic business models for Chinese digital content producers and distributors. 相似文献