Search (94 results, page 5 of 5)

0.0011898974 = product of:
  0.0071393843 = sum of:
    0.0071393843 = weight(_text_:in in 6519) [ClassicSimilarity], result of:
      0.0071393843 = score(doc=6519,freq=8.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.120230645 = fieldWeight in 6519, product of:
          2.828427 = tf(freq=8.0), with freq of:
            8.0 = termFreq=8.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.03125 = fieldNorm(doc=6519)
  0.16666667 = coord(1/6)

Abstract

Relevance has been a difficult concept to define, let alone measure. In this paper, a simple operational definition of relevance is proposed for a Web-based library catalog: whether or not during a search session the user saves, prints, mails, or downloads a citation. If one of those actions is performed, the session is considered relevant to the user. An analysis is presented illustrating the advantages and disadvantages of this definition. With this definition and good transaction logging, it is possible to ascertain the relevance of a session. This was done for 905,970 sessions conducted with the University of California's Melvyl online catalog. Next, a methodology was developed to try to predict the relevance of a session. A number of variables were defined that characterize a session, none of which used any demographic information about the user. The values of the variables were computed for the sessions. Principal components analysis was used to extract a new set of variables out of the original set. A stratified random sampling technique was used to form ten strata such that each new strata of 90,570 sessions contained the same proportion of relevant to nonrelevant sessions. Logistic regression was used to ascertain the regression coefficients for nine of the ten strata. Then, the coefficients were used to predict the relevance of the sessions in the missing strata. Overall, 17.85% of the sessions were determined to be relevant. The predicted number of relevant sessions for all ten strata was 11 %, a 6.85% difference. The authors believe that the methodology can be further refined and the prediction improved. This methodology could also have significant application in improving user searching and also in predicting electronic commerce buying decisions without the use of personal demographic data

0.0011898974 = product of:
  0.0071393843 = sum of:
    0.0071393843 = weight(_text_:in in 2583) [ClassicSimilarity], result of:
      0.0071393843 = score(doc=2583,freq=2.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.120230645 = fieldWeight in 2583, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.0625 = fieldNorm(doc=2583)
  0.16666667 = coord(1/6)

Abstract

The periodic TRECs - Text REtrieval Conferences - have reported the results of a variety of recall studies in large-scale document retrieval. While the efforts of TREC are noteworthy and laudable, there are reasons why its results, especially the recall values which are central to its conclusions, should be accepted with some caution.

0.0011898974 = product of:
  0.0071393843 = sum of:
    0.0071393843 = weight(_text_:in in 2031) [ClassicSimilarity], result of:
      0.0071393843 = score(doc=2031,freq=8.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.120230645 = fieldWeight in 2031, product of:
          2.828427 = tf(freq=8.0), with freq of:
            8.0 = termFreq=8.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.03125 = fieldNorm(doc=2031)
  0.16666667 = coord(1/6)

Abstract

Whereas in language words of high frequency are generally associated with low content [Bookstein, A., & Swanson, D. (1974). Probabilistic models for automatic indexing. Journal of the American Society of Information Science, 25(5), 312-318; Damerau, F. J. (1965). An experiment in automatic indexing. American Documentation, 16, 283-289; Harter, S. P. (1974). A probabilistic approach to automatic keyword indexing. PhD thesis, University of Chicago; Sparck-Jones, K. (1972). A statistical interpretation of term specificity and its application in retrieval. Journal of Documentation, 28, 11-21; Yu, C., & Salton, G. (1976). Precision weighting - an effective automatic indexing method. Journal of the Association for Computer Machinery (ACM), 23(1), 76-88], shallow syntactic fragments of high frequency generally correspond to lexical fragments of high content [Lioma, C., & Ounis, I. (2006). Examining the content load of part of speech blocks for information retrieval. In Proceedings of the international committee on computational linguistics and the association for computational linguistics (COLING/ACL 2006), Sydney, Australia]. We implement this finding to Information Retrieval, as follows. We present a novel automatic query reformulation technique, which is based on shallow syntactic evidence induced from various language samples, and used to enhance the performance of an Information Retrieval system. Firstly, we draw shallow syntactic evidence from language samples of varying size, and compare the effect of language sample size upon retrieval performance, when using our syntactically-based query reformulation (SQR) technique. Secondly, we compare SQR to a state-of-the-art probabilistic pseudo-relevance feedback technique. Additionally, we combine both techniques and evaluate their compatibility. We evaluate our proposed technique across two standard Text REtrieval Conference (TREC) English test collections, and three statistically different weighting models. Experimental results suggest that SQR markedly enhances retrieval performance, and is at least comparable to pseudo-relevance feedback. Notably, the combination of SQR and pseudo-relevance feedback further enhances retrieval performance considerably. These collective experimental results confirm the tenet that high frequency shallow syntactic fragments correspond to content-bearing lexical fragments.

0.0011898974 = product of:
  0.0071393843 = sum of:
    0.0071393843 = weight(_text_:in in 3801) [ClassicSimilarity], result of:
      0.0071393843 = score(doc=3801,freq=2.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.120230645 = fieldWeight in 3801, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.0625 = fieldNorm(doc=3801)
  0.16666667 = coord(1/6)

Abstract

Jean Tague-Sutcliffe was an important figure in information retrieval experimentation. Here, she reviews the history of IR research, and provides a description of the fundamental paradigm of information retrieval experimentation that continues to dominate the field.

0.0011898974 = product of:
  0.0071393843 = sum of:
    0.0071393843 = weight(_text_:in in 3891) [ClassicSimilarity], result of:
      0.0071393843 = score(doc=3891,freq=2.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.120230645 = fieldWeight in 3891, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.0625 = fieldNorm(doc=3891)
  0.16666667 = coord(1/6)

Abstract

Research and development of search engines and other information retrieval (IR) systems proceeds by a cycle of design, implementation, and experimentation, with the results of each experiment influencing design decisions in the next iteration of the cycle. Batch experiments on test collections help ensure that this process goes as smoothly and as quickly as possible. A test collection comprises a collection of documents, a set of information needs, and judgments of the relevance of documents to those needs.

0.0010517307 = product of:
  0.006310384 = sum of:
    0.006310384 = weight(_text_:in in 4549) [ClassicSimilarity], result of:
      0.006310384 = score(doc=4549,freq=4.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.10626988 = fieldWeight in 4549, product of:
          2.0 = tf(freq=4.0), with freq of:
            4.0 = termFreq=4.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.0390625 = fieldNorm(doc=4549)
  0.16666667 = coord(1/6)

Abstract

This paper presents a set of basic components which constitutes the experimental setting intended for the evaluation of interactive information retrieval (IIR) systems, the aim of which is to facilitate evaluation of IIR systems in a way which is as close as possible to realistic IR processes. The experimental settings consists of 3 components: (1) the involvement of potential users as test persons; (2) the application of dynamic and individual information needs; and (3) the use of multidimensionsal and dynamic relevance judgements. Hidden under the information need component is the essential central sub-component, the simulated work task situation, the tool that triggers the (simulated) dynamic information need. This paper also reports on the empirical findings of the meta-evaluation of the application of this sub-component, the purpose of which is to discover whether the application of simulated work task situations to future evaluation of IIR systems can be recommended. Investigations are carried out to dertermine whether any search behavioural differences exist between test persons' treatment of their own real information needs versus simulated information needs. The hypothesis is that if no difference exist one can correctly substitute real information needs with simulated information needs through the application of simulated work task situations. The empirical results of the meta-evaluation provide positive evidence for the application of simulated work task situations to the evaluation of IIR systems. The results also indicate that tailoring work task situations to the group of test persons is important in motivating them. Furthermore, the results of the evaluation show that different versions of semantic openness of the simulated situations make no difference to the test persons' search treatment

0.0010517307 = product of:
  0.006310384 = sum of:
    0.006310384 = weight(_text_:in in 5048) [ClassicSimilarity], result of:
      0.006310384 = score(doc=5048,freq=4.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.10626988 = fieldWeight in 5048, product of:
          2.0 = tf(freq=4.0), with freq of:
            4.0 = termFreq=4.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.0390625 = fieldNorm(doc=5048)
  0.16666667 = coord(1/6)

Abstract

In this article, we introduce a new information system evaluation method and report on its application to a collaborative information seeking system, AntWorld. The key innovation of the new method is to use precisely the same group of users who work with the system as judges, a system we call Cross-Evaluation. In the new method, we also propose to assess the system at the level of task completion. The obvious potential limitation of this method is that individuals may be inclined to think more highly of the materials that they themselves have found and are almost certain to think more highly of their own work product than they do of the products built by others. The keys to neutralizing this problem are careful design and a corresponding analytical model based on analysis of variance. We model the several measures of task completion with a linear model of five effects, describing the users who interact with the system, the system used to finish the task, the task itself, the behavior of individuals as judges, and the selfjudgment bias. Our analytical method successfully isolates the effect of each variable. This approach provides a successful model to make concrete the "threerealities" paradigm, which calls for "real tasks," "real users," and "real systems."

0.0010517307 = product of:
  0.006310384 = sum of:
    0.006310384 = weight(_text_:in in 5154) [ClassicSimilarity], result of:
      0.006310384 = score(doc=5154,freq=4.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.10626988 = fieldWeight in 5154, product of:
          2.0 = tf(freq=4.0), with freq of:
            4.0 = termFreq=4.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.0390625 = fieldNorm(doc=5154)
  0.16666667 = coord(1/6)

Abstract

Lopez-Pujalte and Guerrero-Bote test a relevance feedback genetic algorithm while varying its order based fitness functions and generating a function based upon the Ide dec-hi method as a base line. Using the non-zero weighted term types assigned to the query, and to the initially retrieved set of documents, as genes, a chromosome of equal length is created for each. The algorithm is provided with the chromosomes for judged relevant documents, for judged irrelevant documents, and for the irrelevant documents with their terms negated. The algorithm uses random selection of all possible genes, but gives greater likelihood to those with higher fitness values. When the fittest chromosome of a previous population is eliminated it is restored while the least fittest of the new population is eliminated in its stead. A crossover probability of .8 and a mutation probability of .2 were used with 20 generations. Three fitness functions were utilized; the Horng and Yeh function which takes into account the position of relevant documents, and two new functions, one based on accumulating the cosine similarity for retrieved documents, the other on stored fixed-recall-interval precessions. The Cranfield collection was used with the first 15 documents retrieved from 33 queries chosen to have at least 3 relevant documents in the first 15 and at least 5 relevant documents not initially retrieved. Precision was calculated at fixed recall levels using the residual collection method which removes viewed documents. One of the three functions improved the original retrieval by127 percent, while the Ide dec-hi method provided a 120 percent improvement.

0.0010517307 = product of:
  0.006310384 = sum of:
    0.006310384 = weight(_text_:in in 2094) [ClassicSimilarity], result of:
      0.006310384 = score(doc=2094,freq=4.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.10626988 = fieldWeight in 2094, product of:
          2.0 = tf(freq=4.0), with freq of:
            4.0 = termFreq=4.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.0390625 = fieldNorm(doc=2094)
  0.16666667 = coord(1/6)

Abstract

Some of the most popular measures to evaluate information filtering systems are usually independent of the users because they are based in relevance judgments obtained from experts. On the other hand, the user-centred evaluation allows showing the different impressions that the users have perceived about the system running. This work is focused on discussing the problem of user-centred versus system-centred evaluation of a Web content personalization system where the personalization is based on a user model that stores long term (section, categories and keywords) and short term interests (adapted from user provided feedback). The user-centred evaluation is based on questionnaires filled in by the users before and after using the system and the system-centred evaluation is based on the comparison between ranking of documents, obtained from the application of a multi-tier selection process, and binary relevance judgments collected previously from real users. The user-centred and system-centred evaluations performed with 106 users during 14 working days have provided valuable data concerning the behaviour of the users with respect to issues such as document relevance or the relative importance attributed to different ways of personalization. The results obtained shows general satisfaction on both the personalization processes (selection, adaptation and presentation) and the system as a whole.

0.0010411602 = product of:
  0.006246961 = sum of:
    0.006246961 = weight(_text_:in in 5368) [ClassicSimilarity], result of:
      0.006246961 = score(doc=5368,freq=2.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.10520181 = fieldWeight in 5368, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.0546875 = fieldNorm(doc=5368)
  0.16666667 = coord(1/6)

0.0010304814 = product of:
  0.0061828885 = sum of:
    0.0061828885 = weight(_text_:in in 840) [ClassicSimilarity], result of:
      0.0061828885 = score(doc=840,freq=6.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.1041228 = fieldWeight in 840, product of:
          2.4494898 = tf(freq=6.0), with freq of:
            6.0 = termFreq=6.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.03125 = fieldNorm(doc=840)
  0.16666667 = coord(1/6)

Abstract

Purpose - This paper reports the findings of a study designed to explore web searchers' perceptions of the causes of their search failure and success. In particular, it seeks to discover the extent to which the constructs locus of control and attribution theory might provide useful frameworks for understanding searchers' perceptions. Design/methodology/approach - A combination of inductive and deductive approaches were employed. Perceptions of failed and successful searches were derived from the inductive analysis of using open-ended qualitative interviews with a sample of 37 biologists at the University of Sheffield. These perceptions were classified into "internal" and "external" attributions, and the relationships between these categories and "successful" and "failed" searches were analysed deductively to test the extent to which they might be explainable using locus of control and attribution theory interpretive frameworks. Findings - All searchers were readily able to recall "successful" and "unsuccessful" searches. In a large majority of cases (82.4 per cent), they clearly attributed each search to either internal (e.g. ability or effort) or external (e.g. luck or information not being available) factors. The pattern of such relationships was analysed, and mapped onto those that would be predicted by locus of control and attribution theory. The authors conclude that the potential of these theoretical frameworks to illuminate one's understanding of web searching, and associated training, merits further systematic study. Research limitations/implications - The findings are based on a relatively small sample of academic and research staff in a particular subject area. Importantly, also, the study can at best provide a prima facie case for further systematic study since, although the patterns of attribution behaviour accord with those predictable by locus of control and attribution theory, data relating to the predictive elements of these theories (e.g. levels of confidence and achievement) were not available. This issue is discussed, and recommendations made for further work. Originality/value - The findings provide some empirical support for the notion that locus of control and attribution theory might - subject to the limitations noted above - be potentially useful theoretical frameworks for helping us better understand web-based information seeking. If so, they could have implications particularly for better understanding of searchers' motivations, and for the design and development of more effective search training programmes.

8.9242304E-4 = product of:
  0.005354538 = sum of:
    0.005354538 = weight(_text_:in in 1850) [ClassicSimilarity], result of:
      0.005354538 = score(doc=1850,freq=2.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.09017298 = fieldWeight in 1850, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.046875 = fieldNorm(doc=1850)
  0.16666667 = coord(1/6)

Abstract

Answering open-domain factual questions requires Natural Language processing for refining document selection and answer identification. With our system QALC, we have participated in the Question Answering track of the TREC8, TREC9 and TREC10 evaluations. QALC performs an analysis of documents relying an multiword term searches and their linguistic variation both to minimize the number of documents selected and to provide additional clues when comparing question and sentence representations. This comparison process also makes use of the results of a syntactic parsing of the questions and Named Entity recognition functionalities. Answer extraction relies an the application of syntactic patterns chosen according to the kind of information that is sought, and categorized depending an the syntactic form of the question. These patterns allow QALC to handle nicely linguistic variations at the answer level.

8.9242304E-4 = product of:
  0.005354538 = sum of:
    0.005354538 = weight(_text_:in in 3459) [ClassicSimilarity], result of:
      0.005354538 = score(doc=3459,freq=2.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.09017298 = fieldWeight in 3459, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.046875 = fieldNorm(doc=3459)
  0.16666667 = coord(1/6)

Abstract

Latent Semantic Indexing (LSI), when applied to semantic space built an text collections, improves information retrieval, information filtering, and word sense disambiguation. A new dual probability model based an the similarity concepts is introduced to provide deeper understanding of LSI. Semantic associations can be quantitatively characterized by their statistical significance, the likelihood. Semantic dimensions containing redundant and noisy information can be separated out and should be ignored because their negative contribution to the overall statistical significance. LSI is the optimal solution of the model. The peak in the likelihood curve indicates the existence of an intrinsic semantic dimension. The importance of LSI dimensions follows the Zipf-distribution, indicating that LSI dimensions represent latent concepts. Document frequency of words follows the Zipf distribution, and the number of distinct words follows log-normal distribution. Experiments an five standard document collections confirm and illustrate the analysis.

7.4368593E-4 = product of:
  0.0044621155 = sum of:
    0.0044621155 = weight(_text_:in in 3456) [ClassicSimilarity], result of:
      0.0044621155 = score(doc=3456,freq=2.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.07514416 = fieldWeight in 3456, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.0390625 = fieldNorm(doc=3456)
  0.16666667 = coord(1/6)

Abstract

The existence, public availability, and widespread acceptance of a standard benchmark for a given information retrieval (IR) task are beneficial to research an this task, because they allow different researchers to experimentally compare their own systems by comparing the results they have obtained an this benchmark. The Reuters-21578 test collection, together with its earlier variants, has been such a standard benchmark for the text categorization (TC) task throughout the last 10 years. However, the benefits that this has brought about have somehow been limited by the fact that different researchers have "carved" different subsets out of this collection and tested their systems an one of these subsets only; systems that have been tested an different Reuters-21578 subsets are thus not readily comparable. In this article, we present a systematic, comparative experimental study of the three subsets of Reuters-21578 that have been most popular among TC researchers. The results we obtain allow us to determine the relative hardness of these subsets, thus establishing an indirect means for comparing TC systems that have, or will be, tested an these different subsets.