Search (100 results, page 1 of 5)

0.082231134 = product of:
  0.2466934 = sum of:
    0.1497742 = weight(_text_:suchmaschine in 5234) [ClassicSimilarity], result of:
      0.1497742 = score(doc=5234,freq=4.0), product of:
        0.21191008 = queryWeight, product of:
          5.6542544 = idf(docFreq=420, maxDocs=44218)
          0.03747799 = queryNorm
        0.7067818 = fieldWeight in 5234, product of:
          2.0 = tf(freq=4.0), with freq of:
            4.0 = termFreq=4.0
          5.6542544 = idf(docFreq=420, maxDocs=44218)
          0.0625 = fieldNorm(doc=5234)
    0.0969192 = weight(_text_:ranking in 5234) [ClassicSimilarity], result of:
      0.0969192 = score(doc=5234,freq=2.0), product of:
        0.20271951 = queryWeight, product of:
          5.4090285 = idf(docFreq=537, maxDocs=44218)
          0.03747799 = queryNorm
        0.47809508 = fieldWeight in 5234, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          5.4090285 = idf(docFreq=537, maxDocs=44218)
          0.0625 = fieldNorm(doc=5234)
  0.33333334 = coord(2/6)

Abstract

Anhand eines einfachen Modells wird geprüft, ob der Nutzen den Aufbau einer Spezialsuchmaschine trägt. Zum Vergleich werden fünf große Suchmaschinen herangezogen und deren Ergebnisse bei gleichen Fragestellungen analysiert. Es ergeben sich gravierende Abweichungen sowohl in der Überdeckung der Ergebnismengen als auch im Ranking. Es erscheint also sinnvoll, eine eigene Suchmaschine zu betreiben

0.037682008 = product of:
  0.11304602 = sum of:
    0.102798335 = weight(_text_:ranking in 1454) [ClassicSimilarity], result of:
      0.102798335 = score(doc=1454,freq=4.0), product of:
        0.20271951 = queryWeight, product of:
          5.4090285 = idf(docFreq=537, maxDocs=44218)
          0.03747799 = queryNorm
        0.5070964 = fieldWeight in 1454, product of:
          2.0 = tf(freq=4.0), with freq of:
            4.0 = termFreq=4.0
          5.4090285 = idf(docFreq=537, maxDocs=44218)
          0.046875 = fieldNorm(doc=1454)
    0.010247685 = product of:
      0.030743055 = sum of:
        0.030743055 = weight(_text_:29 in 1454) [ClassicSimilarity], result of:
          0.030743055 = score(doc=1454,freq=2.0), product of:
            0.13183585 = queryWeight, product of:
              3.5176873 = idf(docFreq=3565, maxDocs=44218)
              0.03747799 = queryNorm
            0.23319192 = fieldWeight in 1454, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              3.5176873 = idf(docFreq=3565, maxDocs=44218)
              0.046875 = fieldNorm(doc=1454)
      0.33333334 = coord(1/3)
  0.33333334 = coord(2/6)

Abstract

Information retrieval performance evaluation is commonly made based an the classical recall and precision based figures or graphs. However, important information indicating causes for variation may remain hidden under the average recall and precision figures. Identifying significant causes for variation can help researchers and developers to focus an opportunities for improvement that underlay the averages. This article presents a case study showing the potential of a statistical repeated measures analysis of variance for testing the significance of factors in retrieval performance variation. The TREC-9 Query Track performance data is used as a case study and the factors studied are retrieval method, topic, and their interaction. The results show that retrieval method, topic, and their interaction are all significant. A topic level analysis is also made to see the nature of variation in the performance of retrieval methods across topics. The observed retrieval performances of expansion runs are truly significant improvements for most of the topics. Analyses of the effect of query expansion an document ranking confirm that expansion affects ranking positively.

Date

29. 3.2003 19:28:33

0.031401675 = product of:
  0.09420502 = sum of:
    0.085665286 = weight(_text_:ranking in 4152) [ClassicSimilarity], result of:
      0.085665286 = score(doc=4152,freq=4.0), product of:
        0.20271951 = queryWeight, product of:
          5.4090285 = idf(docFreq=537, maxDocs=44218)
          0.03747799 = queryNorm
        0.42258036 = fieldWeight in 4152, product of:
          2.0 = tf(freq=4.0), with freq of:
            4.0 = termFreq=4.0
          5.4090285 = idf(docFreq=537, maxDocs=44218)
          0.0390625 = fieldNorm(doc=4152)
    0.008539738 = product of:
      0.025619213 = sum of:
        0.025619213 = weight(_text_:29 in 4152) [ClassicSimilarity], result of:
          0.025619213 = score(doc=4152,freq=2.0), product of:
            0.13183585 = queryWeight, product of:
              3.5176873 = idf(docFreq=3565, maxDocs=44218)
              0.03747799 = queryNorm
            0.19432661 = fieldWeight in 4152, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              3.5176873 = idf(docFreq=3565, maxDocs=44218)
              0.0390625 = fieldNorm(doc=4152)
      0.33333334 = coord(1/3)
  0.33333334 = coord(2/6)

Abstract

A software system has been developed which orders citations retrieved from an online database in terms of relevancy. The system resulted from an effort generated by NASA's Technology Utilization Program to create new advanced software tools to largely automate the process of determining relevancy of database citations retrieved to support large technology transfer studies. The ranking is based on the generation of an enriched vocabulary using lexical association methods, a user assessment of the vocabulary and a combination of the user assessment and the lexical metric. One of the key elements in relevancy ranking is the enriched vocabulary -the terms mst be both unique and descriptive. This paper examines term uniqueness. Six lexical association methods were employed to generate characteristic word indices. A limited subset of the terms - the highest 20,40,60 and 7,5% of the uniquess words - we compared and uniquess factors developed. Computational times were also measured. It was found that methods based on occurrences and signal produced virtually the same terms. The limited subset of terms producedby the exact and centroid discrimination value were also nearly identical. Unique terms sets were produced by teh occurrence, variance and discrimination value (centroid), An end-user evaluation showed that the generated terms were largely distinct and had values of word precision which were consistent with values of the search precision.

Date

23.11.1995 11:29:46

0.028268103 = product of:
  0.16960861 = sum of:
    0.16960861 = weight(_text_:ranking in 3858) [ClassicSimilarity], result of:
      0.16960861 = score(doc=3858,freq=2.0), product of:
        0.20271951 = queryWeight, product of:
          5.4090285 = idf(docFreq=537, maxDocs=44218)
          0.03747799 = queryNorm
        0.8366664 = fieldWeight in 3858, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          5.4090285 = idf(docFreq=537, maxDocs=44218)
          0.109375 = fieldNorm(doc=3858)
  0.16666667 = coord(1/6)

0.027614966 = product of:
  0.0828449 = sum of:
    0.0726894 = weight(_text_:ranking in 328) [ClassicSimilarity], result of:
      0.0726894 = score(doc=328,freq=2.0), product of:
        0.20271951 = queryWeight, product of:
          5.4090285 = idf(docFreq=537, maxDocs=44218)
          0.03747799 = queryNorm
        0.35857132 = fieldWeight in 328, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          5.4090285 = idf(docFreq=537, maxDocs=44218)
          0.046875 = fieldNorm(doc=328)
    0.0101555 = product of:
      0.030466499 = sum of:
        0.030466499 = weight(_text_:22 in 328) [ClassicSimilarity], result of:
          0.030466499 = score(doc=328,freq=2.0), product of:
            0.13124153 = queryWeight, product of:
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.03747799 = queryNorm
            0.23214069 = fieldWeight in 328, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.046875 = fieldNorm(doc=328)
      0.33333334 = coord(1/3)
  0.33333334 = coord(2/6)

Abstract

In diesem Artikel wird ein Eyetracking-Experiment beschrieben, bei dem untersucht wurde, wann und auf Basis welcher Informationen Relevanzentscheidungen bei der themenbezogenen Dokumentenbewertung fallen und welche Faktoren auf die Relevanzentscheidung einwirken. Nach einer kurzen Einführung werden relevante Studien aufgeführt, in denen Eyetracking als Untersuchungsmethode für Interaktionsverhalten mit Ergebnislisten (Information Seeking Behavior) verwendet wurde. Nutzerverhalten wird hierbei vor allem durch unterschiedliche Aufgaben-Typen, dargestellte Informationen und durch das Ranking eines Ergebnisses beeinflusst. Durch EyetrackingUntersuchungen lassen sich Nutzer außerdem in verschiedene Klassen von Bewertungs- und Lesetypen einordnen. Diese Informationen können als implizites Feedback genutzt werden, um so die Suche zu personalisieren und um die Relevanz von Suchergebnissen ohne aktives Zutun des Users zu erhöhen. In einem explorativen Eyetracking-Experiment mit 12 Studenten der Hochschule Darmstadt werden anhand der Länge der Gesamtbewertung, Anzahl der Fixationen, Anzahl der besuchten Metadatenelemente und Länge des Scanpfades zwei typische Bewertungstypen identifiziert. Das Metadatenfeld Abstract wird im Experiment zuverlässig als wichtigste Dokumenteigenschaft für die Zuordnung von Relevanz ermittelt.

Date

22. 7.2012 19:25:54

0.022844076 = product of:
  0.13706446 = sum of:
    0.13706446 = weight(_text_:ranking in 3802) [ClassicSimilarity], result of:
      0.13706446 = score(doc=3802,freq=4.0), product of:
        0.20271951 = queryWeight, product of:
          5.4090285 = idf(docFreq=537, maxDocs=44218)
          0.03747799 = queryNorm
        0.67612857 = fieldWeight in 3802, product of:
          2.0 = tf(freq=4.0), with freq of:
            4.0 = termFreq=4.0
          5.4090285 = idf(docFreq=537, maxDocs=44218)
          0.0625 = fieldNorm(doc=3802)
  0.16666667 = coord(1/6)

Abstract

Recall, the proortion of the relevant documents retrieved, is a key indicator of the performance of an information retrieval system. With large information systems, like the WWW, recal is almost impossible to measure or estimate by all standard techniques. Proposes an 'needle hiding' technique for measuring recall under these circumstances. Shows that ranking by relative recall does not have to be isomorphic to ranking by recall and hence the use of relative recall for comparative evaluation might not be entirely sound

0.020983625 = product of:
  0.12590174 = sum of:
    0.12590174 = weight(_text_:ranking in 3855) [ClassicSimilarity], result of:
      0.12590174 = score(doc=3855,freq=6.0), product of:
        0.20271951 = queryWeight, product of:
          5.4090285 = idf(docFreq=537, maxDocs=44218)
          0.03747799 = queryNorm
        0.62106377 = fieldWeight in 3855, product of:
          2.4494898 = tf(freq=6.0), with freq of:
            6.0 = termFreq=6.0
          5.4090285 = idf(docFreq=537, maxDocs=44218)
          0.046875 = fieldNorm(doc=3855)
  0.16666667 = coord(1/6)

Abstract

The University of Huddersfield, UK, is researching ways of incorporating ranking and relevance feedback techniques into a thesaurus based searching system. The INSPEC database on STN International was searched using the VUSE (View-based Search Engine) interface. Thesaurus terms from documents judged to be relevant by users were used to query INSPEC and create a ranking of documents based on probabilistic methods. An evaluation was carried out to establish whether or not it would be better for the user to continue searching with the thesaurus based front end or to use relevance feedback, looking at the ranked list of documents it would produce. Also looks at the amount of effort the user had to expend to get relevant documents in terms of the number of non relevant documents seen between relevant documents

0.020983625 = product of:
  0.12590174 = sum of:
    0.12590174 = weight(_text_:ranking in 2060) [ClassicSimilarity], result of:
      0.12590174 = score(doc=2060,freq=6.0), product of:
        0.20271951 = queryWeight, product of:
          5.4090285 = idf(docFreq=537, maxDocs=44218)
          0.03747799 = queryNorm
        0.62106377 = fieldWeight in 2060, product of:
          2.4494898 = tf(freq=6.0), with freq of:
            6.0 = termFreq=6.0
          5.4090285 = idf(docFreq=537, maxDocs=44218)
          0.046875 = fieldNorm(doc=2060)
  0.16666667 = coord(1/6)

Abstract

Information retrieval involves finding some desired information in a store of information or a database. In this paper, Co-word analysis will be used to achieve a ranking of a selected sample of FA terms. Based on this ranking a better arranging of search results can be achieved. Experimental results achieved using 41 MB of data (7660 documents) in the field of sports. The corpus was collected from CNN newspaper, sports field. This corpus was chosen to be distributed over 11 sub-fields of the field sports from the experimental results, the average precision increased by 18.3% after applying the proposed arranging scheme depending on the absolute frequency to count the terms weights, and the average precision increased by 17.2% after applying the proposed arranging scheme depending on a formula based on "TF*IDF" to count the terms weights.

0.020191502 = product of:
  0.121149 = sum of:
    0.121149 = weight(_text_:ranking in 4052) [ClassicSimilarity], result of:
      0.121149 = score(doc=4052,freq=2.0), product of:
        0.20271951 = queryWeight, product of:
          5.4090285 = idf(docFreq=537, maxDocs=44218)
          0.03747799 = queryNorm
        0.5976189 = fieldWeight in 4052, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          5.4090285 = idf(docFreq=537, maxDocs=44218)
          0.078125 = fieldNorm(doc=4052)
  0.16666667 = coord(1/6)

0.018430464 = product of:
  0.055291392 = sum of:
    0.0484596 = weight(_text_:ranking in 3960) [ClassicSimilarity], result of:
      0.0484596 = score(doc=3960,freq=2.0), product of:
        0.20271951 = queryWeight, product of:
          5.4090285 = idf(docFreq=537, maxDocs=44218)
          0.03747799 = queryNorm
        0.23904754 = fieldWeight in 3960, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          5.4090285 = idf(docFreq=537, maxDocs=44218)
          0.03125 = fieldNorm(doc=3960)
    0.0068317903 = product of:
      0.02049537 = sum of:
        0.02049537 = weight(_text_:29 in 3960) [ClassicSimilarity], result of:
          0.02049537 = score(doc=3960,freq=2.0), product of:
            0.13183585 = queryWeight, product of:
              3.5176873 = idf(docFreq=3565, maxDocs=44218)
              0.03747799 = queryNorm
            0.15546128 = fieldWeight in 3960, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              3.5176873 = idf(docFreq=3565, maxDocs=44218)
              0.03125 = fieldNorm(doc=3960)
      0.33333334 = coord(1/3)
  0.33333334 = coord(2/6)

Abstract

Purpose - This paper aims to discuss and test the claim that utilization of the personalization techniques can be valuable to improve the efficiency of collaborative information retrieval (CIR) systems. Design/methodology/approach - A new personalized CIR system, called PERCIRS, is presented based on the user profile similarity calculation (UPSC) formulas. To this aim, the paper proposes several UPSC formulas as well as two techniques to evaluate them. As the proposed CIR system is personalized, it could not be evaluated by Cranfield, like evaluation techniques (e.g. TREC). Hence, this paper proposes a new user-centric mechanism, which enables PERCIRS to be evaluated. This mechanism is generic and can be used to evaluate any other personalized IR system. Findings - The results show that among the proposed UPSC formulas in this paper, the (query-document)-graph based formula is the most effective. After integrating this formula into PERCIRS and comparing it with nine other IR systems, it is concluded that the results of the system are better than the other IR systems. In addition, the paper shows that the complexity of the system is less that the complexity of the other CIR systems. Research limitations/implications - This system asks the users to explicitly rank the returned documents, while explicit ranking is still not widespread enough. However it believes that the users should actively participate in the IR process in order to aptly satisfy their needs to information. Originality/value - The value of this paper lies in combining collaborative and personalized IR, as well as introducing a mechanism which enables the personalized IR system to be evaluated. The proposed evaluation mechanism is very valuable for developers of personalized IR systems. The paper also introduces some significant user profile similarity calculation formulas, and two techniques to evaluate them. These formulas can also be used to find the user's community in the social networks.

Date

29. 8.2010 12:59:10

0.017486354 = product of:
  0.10491812 = sum of:
    0.10491812 = weight(_text_:ranking in 5231) [ClassicSimilarity], result of:
      0.10491812 = score(doc=5231,freq=6.0), product of:
        0.20271951 = queryWeight, product of:
          5.4090285 = idf(docFreq=537, maxDocs=44218)
          0.03747799 = queryNorm
        0.51755315 = fieldWeight in 5231, product of:
          2.4494898 = tf(freq=6.0), with freq of:
            6.0 = termFreq=6.0
          5.4090285 = idf(docFreq=537, maxDocs=44218)
          0.0390625 = fieldNorm(doc=5231)
  0.16666667 = coord(1/6)

Abstract

In this issue we begin with a paper where Eastman points out that conventional narrower queries (the use of conjunctions and phrases) in a web engine search will reduce returned number of hits but not necessarily increase precision in the top ranked documents in the return. Thus by precision anomalies Eastman means that search narrowing activity results in no precision change or a decrease in precision. Multiple queries with multiple engines were run by students for a three-year period and the formulation/engine combination was recorded as was the number of hits. Relevance was also recorded for the top ten and top twenty ranked retrievals. While narrower searches reduced total hits they did not usually improve precision. Initial high precision and poor query reformulation account for some of the results, as did Alta Vista's failure to use the ranking algorithm incorporated in its regular search in its advanced search feature. However, since the top listed returns often reoccurred in all formulations, it would seem that the ranking algorithms are doing a consistent job of practical precision ranking that is not improved by reformulation.

0.016153201 = product of:
  0.0969192 = sum of:
    0.0969192 = weight(_text_:ranking in 7463) [ClassicSimilarity], result of:
      0.0969192 = score(doc=7463,freq=2.0), product of:
        0.20271951 = queryWeight, product of:
          5.4090285 = idf(docFreq=537, maxDocs=44218)
          0.03747799 = queryNorm
        0.47809508 = fieldWeight in 7463, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          5.4090285 = idf(docFreq=537, maxDocs=44218)
          0.0625 = fieldNorm(doc=7463)
  0.16666667 = coord(1/6)

Abstract

Compares online searching of DIALOG (a traditional Boolean system), TARGET (a relevance ranking system) and DR-LINK (an advanced intelligent text processing system), in order to establish the differing strengths of traditional and natural language processing search systems. Details example search queries used in comparison and how each of the systems performed. Considers the implications of the findings for professional information searchers and end users. Natural language processing systems are useful because they develop an wider understanding of queries that use of traditional systems may not

0.014277548 = product of:
  0.085665286 = sum of:
    0.085665286 = weight(_text_:ranking in 669) [ClassicSimilarity], result of:
      0.085665286 = score(doc=669,freq=4.0), product of:
        0.20271951 = queryWeight, product of:
          5.4090285 = idf(docFreq=537, maxDocs=44218)
          0.03747799 = queryNorm
        0.42258036 = fieldWeight in 669, product of:
          2.0 = tf(freq=4.0), with freq of:
            4.0 = termFreq=4.0
          5.4090285 = idf(docFreq=537, maxDocs=44218)
          0.0390625 = fieldNorm(doc=669)
  0.16666667 = coord(1/6)

Abstract

With the emerging needs of creating fairness-aware solutions for search and recommendation systems, a daunting challenge exists of evaluating such solutions. While many of the traditional information retrieval (IR) metrics can capture the relevance, diversity, and novelty for the utility with respect to users, they are not suitable for inferring whether the presented results are fair from the perspective of responsible information exposure. On the other hand, existing fairness metrics do not account for user utility or do not measure it adequately. To address this problem, we propose a new metric called FAIR. By unifying standard IR metrics and fairness measures into an integrated metric, this metric offers a new perspective for evaluating fairness-aware ranking results. Based on this metric, we developed an effective ranking algorithm that jointly optimized user utility and fairness. The experimental results showed that our FAIR metric could highlight results with good user utility and fair information exposure. We showed how FAIR related to a set of existing utility and fairness metrics and demonstrated the effectiveness of our FAIR-based algorithm. We believe our work opens up a new direction of pursuing a metric for evaluating and implementing the FAIR systems.

0.014134051 = product of:
  0.084804304 = sum of:
    0.084804304 = weight(_text_:ranking in 399) [ClassicSimilarity], result of:
      0.084804304 = score(doc=399,freq=2.0), product of:
        0.20271951 = queryWeight, product of:
          5.4090285 = idf(docFreq=537, maxDocs=44218)
          0.03747799 = queryNorm
        0.4183332 = fieldWeight in 399, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          5.4090285 = idf(docFreq=537, maxDocs=44218)
          0.0546875 = fieldNorm(doc=399)
  0.16666667 = coord(1/6)

Abstract

Computer based systems can produce enourmous retrieval sets even when good search logic is used. Sometimes this is desirable, more often it is not. Appropriate filters can limit search results, but they represent only a partial solution. Simple ranking techniques are needed that are both effective and easily understood by the humans doing the searching. Optimal search output, whether from a traditional database or the Internet, will result when intuitive interfaces are designed that inspire confidence while making the necessary mathematics transparent. Weighted term searching using powers of 2, a technique proposed early in the history of information retrieval, can be simplifies and used in combination with modern graphics and textual input to achieve these results

0.014134051 = product of:
  0.084804304 = sum of:
    0.084804304 = weight(_text_:ranking in 2032) [ClassicSimilarity], result of:
      0.084804304 = score(doc=2032,freq=2.0), product of:
        0.20271951 = queryWeight, product of:
          5.4090285 = idf(docFreq=537, maxDocs=44218)
          0.03747799 = queryNorm
        0.4183332 = fieldWeight in 2032, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          5.4090285 = idf(docFreq=537, maxDocs=44218)
          0.0546875 = fieldNorm(doc=2032)
  0.16666667 = coord(1/6)

Abstract

This paper investigates two relatively new measures of retrieval effectiveness in relation to the problem of incomplete relevance data. The measures, Bpref and RankEff, which do not take into account documents that have not been relevance judged, are compared theoretically and experimentally. The experimental comparisons involve a third measure, the well-known mean uninterpolated average precision. The results indicate that RankEff is the most stable of the three measures when the amount of relevance data is reduced, with respect to system ranking and absolute values. In addition, RankEff has the lowest error-rate.

0.0121149 = product of:
  0.0726894 = sum of:
    0.0726894 = weight(_text_:ranking in 6262) [ClassicSimilarity], result of:
      0.0726894 = score(doc=6262,freq=2.0), product of:
        0.20271951 = queryWeight, product of:
          5.4090285 = idf(docFreq=537, maxDocs=44218)
          0.03747799 = queryNorm
        0.35857132 = fieldWeight in 6262, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          5.4090285 = idf(docFreq=537, maxDocs=44218)
          0.046875 = fieldNorm(doc=6262)
  0.16666667 = coord(1/6)

Abstract

This contribution reports the results of continuing research in human-information system interactions. Following training and experience with an electronic information retrieval system novice and experienced subject groups responded to questions ranking their value assessments of 7 attributes of information sources in relation to 15 factors describing the search process. In general, novice users were more heavily influenced by the process factors (negative influences) than by the positive attributes of information qualities. Experienced users, while still concerned with process factors, were more strongly influenced by the qualitative information attributes. The specific advantages and contributions of this research are several: higher dimensionality of measured factors and attributes (15 x 7); higher granularity of analysis using a 7 value metric in a closed-end Likert scale; development of bi-directional, firced-choice influence vectors; and a larger sample size (N=186) than previously reported in the literature

0.0121149 = product of:
  0.0726894 = sum of:
    0.0726894 = weight(_text_:ranking in 3961) [ClassicSimilarity], result of:
      0.0726894 = score(doc=3961,freq=2.0), product of:
        0.20271951 = queryWeight, product of:
          5.4090285 = idf(docFreq=537, maxDocs=44218)
          0.03747799 = queryNorm
        0.35857132 = fieldWeight in 3961, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          5.4090285 = idf(docFreq=537, maxDocs=44218)
          0.046875 = fieldNorm(doc=3961)
  0.16666667 = coord(1/6)

Abstract

Modeling multidimensional relevance in information retrieval (IR) has attracted much attention in recent years. However, most existing studies are conducted through relatively small-scale user studies, which may not reflect a real-world and natural search scenario. In this article, we propose to study the multidimensional user relevance model (MURM) on large scale query logs, which record users' various search behaviors (e.g., query reformulations, clicks and dwelling time, etc.) in natural search settings. We advance an existing MURM model (including five dimensions: topicality, novelty, reliability, understandability, and scope) by providing two additional dimensions, that is, interest and habit. The two new dimensions represent personalized relevance judgment on retrieved documents. Further, for each dimension in the enriched MURM model, a set of computable features are formulated. By conducting extensive document ranking experiments on Bing's query logs and TREC session Track data, we systematically investigated the impact of each dimension on retrieval performance and gained a series of insightful findings which may bring benefits for the design of future IR systems.

0.011422038 = product of:
  0.06853223 = sum of:
    0.06853223 = weight(_text_:ranking in 5208) [ClassicSimilarity], result of:
      0.06853223 = score(doc=5208,freq=4.0), product of:
        0.20271951 = queryWeight, product of:
          5.4090285 = idf(docFreq=537, maxDocs=44218)
          0.03747799 = queryNorm
        0.33806428 = fieldWeight in 5208, product of:
          2.0 = tf(freq=4.0), with freq of:
            4.0 = termFreq=4.0
          5.4090285 = idf(docFreq=537, maxDocs=44218)
          0.03125 = fieldNorm(doc=5208)
  0.16666667 = coord(1/6)

Abstract

Ruthven, Lalmas, and van Rijsbergen use traditional term importance measures like inverse document frequency, noise, based upon in-document frequency, and term frequency supplemented by theme value which is calculated from differences of expected positions of words in a text from their actual positions, on the assumption that even distribution indicates term association with a main topic, and context, which is based on a query term's distance from the nearest other query term relative to the average expected distribution of all query terms in the document. They then define document characteristics like specificity, the sum of all idf values in a document over the total terms in the document, or document complexity, measured by the documents average idf value; and information to noise ratio, info-noise, tokens after stopping and stemming over tokens before these processes, measuring the ratio of useful and non-useful information in a document. Retrieval tests are then carried out using each characteristic, combinations of the characteristics, and relevance feedback to determine the correct combination of characteristics. A file ranks independently of query terms by both specificity and info-noise, but if presence of a query term is required unique rankings are generated. Tested on five standard collections the traditional characteristics out preformed the new characteristics, which did, however, out preform random retrieval. All possible combinations of characteristics were also tested both with and without a set of scaling weights applied. All characteristics can benefit by combination with another characteristic or set of characteristics and performance as a single characteristic is a good indicator of performance in combination. Larger combinations tended to be more effective than smaller ones and weighting increased precision measures of middle ranking combinations but decreased the ranking of poorer combinations. The best combinations vary for each collection, and in some collections with the addition of weighting. Finally, with all documents ranked by the all characteristics combination, they take the top 30 documents and calculate the characteristic scores for each term in both the relevant and the non-relevant sets. Then taking for each query term the characteristics whose average was higher for relevant than non-relevant documents the documents are re-ranked. The relevance feedback method of selecting characteristics can select a good set of characteristics for query terms.

0.010095751 = product of:
  0.0605745 = sum of:
    0.0605745 = weight(_text_:ranking in 3608) [ClassicSimilarity], result of:
      0.0605745 = score(doc=3608,freq=2.0), product of:
        0.20271951 = queryWeight, product of:
          5.4090285 = idf(docFreq=537, maxDocs=44218)
          0.03747799 = queryNorm
        0.29880944 = fieldWeight in 3608, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          5.4090285 = idf(docFreq=537, maxDocs=44218)
          0.0390625 = fieldNorm(doc=3608)
  0.16666667 = coord(1/6)

Abstract

There have been 3 Text Retrieval Conferences (TREC) organized by the National Insitute of Standards and Technology (NIST) over the last 3 years which have compared retrieval results on fairly large databases (at least 1 gigabyte). The queries (called topics), relevance judgements and databases were all provided by NIST. The main goal of the tests was to compare various retrieval algorithms using various measures of retrieval effectiveness. When Tague-Sutcliffe performed an analysis of variance on the average precision there is a large group of systems at the top of the ranking which are not significantly different. In addition the queries contribute more to the mean square the systems. To gather further insight into the results, this research investigates the variations in query properties as a partial explanation for the variation in retrieval scores. For each topic statement for the queries, the length (number of content words), langth of various parts and total number of relevant documents are correlated with the average precision

0.010095751 = product of:
  0.0605745 = sum of:
    0.0605745 = weight(_text_:ranking in 4393) [ClassicSimilarity], result of:
      0.0605745 = score(doc=4393,freq=2.0), product of:
        0.20271951 = queryWeight, product of:
          5.4090285 = idf(docFreq=537, maxDocs=44218)
          0.03747799 = queryNorm
        0.29880944 = fieldWeight in 4393, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          5.4090285 = idf(docFreq=537, maxDocs=44218)
          0.0390625 = fieldNorm(doc=4393)
  0.16666667 = coord(1/6)

Abstract

Purpose - This paper aims to investigate how readers assess relevance of retrieved documents in a foreign language they know well compared with their native language, and whether work-task scenario descriptions have effect on the assessment process. Design/methodology/approach - Queries, test collections, and relevance assessments were used from the 2002 Interactive CLEF. Swedish first-language speakers, fluent in English, were given simulated information-seeking scenarios and presented with retrieval results in both languages. Twenty-eight subjects in four groups were asked to rate the retrieved text documents by relevance. A two-level work-task scenario description framework was developed and applied to facilitate the study of context effects on the assessment process. Findings - Relevance assessment takes longer in a foreign language than in the user first language. The quality of assessments by comparison with pre-assessed results is inferior to those made in the users' first language. Work-task scenario descriptions had an effect on the assessment process, both by measured access time and by self-report by subjects. However, effects on results by traditional relevance ranking were detectable. This may be an argument for extending the traditional IR experimental topical relevance measures to cater for context effects. Originality/value - An extended two-level work-task scenario description framework was developed and applied. Contextual aspects had an effect on the relevance assessment process. English texts took longer to assess than Swedish and were assessed less well, especially for the most difficult queries. The IR research field needs to close this gap and to design information access systems with users' language competence in mind.