Search (82 results, page 1 of 5)

0.106722504 = product of:
  0.42689002 = sum of:
    0.42689002 = weight(_text_:judged in 3885) [ClassicSimilarity], result of:
      0.42689002 = score(doc=3885,freq=4.0), product of:
        0.51079464 = queryWeight, product of:
          7.6410246 = idf(docFreq=57, maxDocs=44421)
          0.06684897 = queryNorm
        0.83573705 = fieldWeight in 3885, product of:
          2.0 = tf(freq=4.0), with freq of:
            4.0 = termFreq=4.0
          7.6410246 = idf(docFreq=57, maxDocs=44421)
          0.0546875 = fieldNorm(doc=3885)
  0.25 = coord(1/4)

Abstract

This article attempts to verify the hypothesis of the document presentation order by an empirical, two-stage experiment. It aims to identify the relationship between number of documents judged and order effect. The results indicate that significant order effect takes place when 15 and 30 documents are presented. Sets with 45 and 60 documents still reveal the order effect. However, subjects are not influenced by order of presentation when the set of documents has 5 and 75 members, respectively.

0.09336275 = product of:
  0.373451 = sum of:
    0.373451 = weight(_text_:judged in 684) [ClassicSimilarity], result of:
      0.373451 = score(doc=684,freq=6.0), product of:
        0.51079464 = queryWeight, product of:
          7.6410246 = idf(docFreq=57, maxDocs=44421)
          0.06684897 = queryNorm
        0.73111767 = fieldWeight in 684, product of:
          2.4494898 = tf(freq=6.0), with freq of:
            6.0 = termFreq=6.0
          7.6410246 = idf(docFreq=57, maxDocs=44421)
          0.0390625 = fieldNorm(doc=684)
  0.25 = coord(1/4)

Abstract

In a novel approach to information finding in networked environments, each user's specific purpose or "quest" can be represented in numerous ways. The most familiar is a list of keywords, or a natural language sentence or paragraph. More effective is an extended text that has been judged as to relevance. This forms the basis of relevance feedback, as it is used in information retrieval. In the "Ant World" project (Ant World, 1999; Kantor et al., 1999b; Kantor et al., 1999a), the items to be retrieved are not documents, but rather quests, represented by entire collections of judged documents. In order to save space and time we have developed methods for representing these complex entities in a short string of about 1,000 bytes, which we call a "Digital Information Pheromone" (DIP). The principles for determining the DIP for a given quest, and for matching DIPs to each other are presented. The effectiveness of this scheme is explored with some applications to the large judged collections of TREC documents

0.09336275 = product of:
  0.373451 = sum of:
    0.373451 = weight(_text_:judged in 2167) [ClassicSimilarity], result of:
      0.373451 = score(doc=2167,freq=6.0), product of:
        0.51079464 = queryWeight, product of:
          7.6410246 = idf(docFreq=57, maxDocs=44421)
          0.06684897 = queryNorm
        0.73111767 = fieldWeight in 2167, product of:
          2.4494898 = tf(freq=6.0), with freq of:
            6.0 = termFreq=6.0
          7.6410246 = idf(docFreq=57, maxDocs=44421)
          0.0390625 = fieldNorm(doc=2167)
  0.25 = coord(1/4)

Abstract

This study systematically compares retrievals on 11 topics across five well-known data bases, with MEDLINE's subject indexing as a focus. Each topic was posed by a researcher in the medical behavioral sciences. Each was searches in MEDLINE, EXCERPTA MEDICA, and PSYCHINFO, which permit descriptor searches, and in SCISEARCH and SOCIAL SCISEARCH, which express topics through cited references. Searches on each topic were made with (1) descriptors, (2) cited references, and (3) natural language (a capabiblity common to all five data bases). The researchers who posed the topics judged the results. In every case, the set of records judged relevant was used to to calculate recall, precision, and novelty ratios. Overall, MEDLINE had the highest recall percentage (37%), followed by SSCI (31%). All searches resulted in high precision ratios; novelty ratios of data bases and searches varied widely. Differences in record format among data bases affected the success of the natural language retrievals. Some 445 documents judged relevant were not retrieved from MEDLINE using its descriptors; they were found in MEDLINE through natural language or in an alternative data base. An analysis was performed to examine possible faults in MEDLINE subject indexing as the reason for their nonretrieval. However, no patterns of indexing failure could be seen in those documents subsequently found in MEDLINE through known-item searches. Documents not found in MEDLINE primarily represent failures of coverage - articles were from nonindexed or selectively indexed journals

0.07623036 = product of:
  0.30492145 = sum of:
    0.30492145 = weight(_text_:judged in 154) [ClassicSimilarity], result of:
      0.30492145 = score(doc=154,freq=4.0), product of:
        0.51079464 = queryWeight, product of:
          7.6410246 = idf(docFreq=57, maxDocs=44421)
          0.06684897 = queryNorm
        0.59695506 = fieldWeight in 154, product of:
          2.0 = tf(freq=4.0), with freq of:
            4.0 = termFreq=4.0
          7.6410246 = idf(docFreq=57, maxDocs=44421)
          0.0390625 = fieldNorm(doc=154)
  0.25 = coord(1/4)

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.

0.07546421 = product of:
  0.30185685 = sum of:
    0.30185685 = weight(_text_:judged in 4275) [ClassicSimilarity], result of:
      0.30185685 = score(doc=4275,freq=2.0), product of:
        0.51079464 = queryWeight, product of:
          7.6410246 = idf(docFreq=57, maxDocs=44421)
          0.06684897 = queryNorm
        0.5909554 = fieldWeight in 4275, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          7.6410246 = idf(docFreq=57, maxDocs=44421)
          0.0546875 = fieldNorm(doc=4275)
  0.25 = coord(1/4)

Abstract

Examines judgements of relevance of document representations to query statements made by people other than the the originators of the queries. A small group of graduate students in the School of Information and Library Studies and undergraduates of Michigan Univ. judges sets of documents that had been retrieved for and judged by real users for a previous study. The assessment of relevance, by the secondary judges, were analysed by themselves and in comparison with the users' assessments. The judges performed reasonably well but some important differences were identified. Secondary judges use the various fields of document records in different ways than users and have a higher threshold of relevance

0.07546421 = product of:
  0.30185685 = sum of:
    0.30185685 = weight(_text_:judged in 6698) [ClassicSimilarity], result of:
      0.30185685 = score(doc=6698,freq=2.0), product of:
        0.51079464 = queryWeight, product of:
          7.6410246 = idf(docFreq=57, maxDocs=44421)
          0.06684897 = queryNorm
        0.5909554 = fieldWeight in 6698, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          7.6410246 = idf(docFreq=57, maxDocs=44421)
          0.0546875 = fieldNorm(doc=6698)
  0.25 = coord(1/4)

Abstract

In the Text REtrieval Conference (TREC) collection - a large full text experimental text collection with varying documents lengths - observes that the likelihood of a document being judged relevant by a user increases with the document length. A retrieval strategy, such as the vector space cosine match, that retrieves documents of different lengths with roughly equal chances, will not optimally retrieve useful documents from such a collection. Presents a modified technique (pivoted cosine normalization) that attempts to match the likelihood of retrieving documents of all lengths to the likelihood of their relevance and shows that this technique yields significant improvements in retrieval effectiveness

0.07546421 = product of:
  0.30185685 = sum of:
    0.30185685 = weight(_text_:judged in 1623) [ClassicSimilarity], result of:
      0.30185685 = score(doc=1623,freq=2.0), product of:
        0.51079464 = queryWeight, product of:
          7.6410246 = idf(docFreq=57, maxDocs=44421)
          0.06684897 = queryNorm
        0.5909554 = fieldWeight in 1623, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          7.6410246 = idf(docFreq=57, maxDocs=44421)
          0.0546875 = fieldNorm(doc=1623)
  0.25 = coord(1/4)

Abstract

Reports results of research to examine users conducting their initial online search on a particular information problem. Findings from 3 separate studies of relevance judgements by 44 initial search users were examined, including 2 studies of 13 end users and a study of 18 user engaged in mediated online searches. Number of items was judged on the scale 'relevant', 'patially relevant' and 'not rlevant'. Results suggest that: a relationship exists between partially rlevant items retrieved anch changes in the users' information problem or question during an information seeking process; partial relevance judgements play an important role for users in the early stages of seeking information on a particular information problem; and 'highly' relevant items may or may not be the only items useful at the early stages of users' information seeking processes

0.07546421 = product of:
  0.30185685 = sum of:
    0.30185685 = weight(_text_:judged in 3032) [ClassicSimilarity], result of:
      0.30185685 = score(doc=3032,freq=2.0), product of:
        0.51079464 = queryWeight, product of:
          7.6410246 = idf(docFreq=57, maxDocs=44421)
          0.06684897 = queryNorm
        0.5909554 = fieldWeight in 3032, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          7.6410246 = idf(docFreq=57, maxDocs=44421)
          0.0546875 = fieldNorm(doc=3032)
  0.25 = coord(1/4)

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.

0.07454643 = product of:
  0.2981857 = sum of:
    0.2981857 = weight(_text_:headings in 2446) [ClassicSimilarity], result of:
      0.2981857 = score(doc=2446,freq=12.0), product of:
        0.32437804 = queryWeight, product of:
          4.8524013 = idf(docFreq=942, maxDocs=44421)
          0.06684897 = queryNorm
        0.91925365 = fieldWeight in 2446, product of:
          3.4641016 = tf(freq=12.0), with freq of:
            12.0 = termFreq=12.0
          4.8524013 = idf(docFreq=942, maxDocs=44421)
          0.0546875 = fieldNorm(doc=2446)
  0.25 = coord(1/4)

Abstract

A recent survey at the Royal Library in the Netherlands showed that subject headings are more efficient than title keywords for retrieval purposes. 475 Dutch publications were selected at random and assigned subject headings. The study showed that subject headings provided additional useful information in 56% of titles. Subsequent searching of the library's online catalogue showed that 88% of titles were retrieved via subject headings against 57% through title keywords. Further precision may be achieved with the help of indexing staff, but at considerable cost

Footnote

Übers. d. Titels: A good title has no need of subject headings, or does it?: a comparative study of title keywords against subject headings

0.0646836 = product of:
  0.2587344 = sum of:
    0.2587344 = weight(_text_:judged in 4893) [ClassicSimilarity], result of:
      0.2587344 = score(doc=4893,freq=2.0), product of:
        0.51079464 = queryWeight, product of:
          7.6410246 = idf(docFreq=57, maxDocs=44421)
          0.06684897 = queryNorm
        0.50653315 = fieldWeight in 4893, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          7.6410246 = idf(docFreq=57, maxDocs=44421)
          0.046875 = fieldNorm(doc=4893)
  0.25 = coord(1/4)

Abstract

Reports preliminary results from a study, conducted at Rutgers Univ., School of Communication, Information and Library Studies, to conduct observations and experiments under real-life conditions on the nature, effects and patterns in the discourse between users and intermediary searchers and in the related computer commands in the context of online searching and responses. The study involved videotaping interactions between users and intermediaries and recording the search logs for 40 questions. Users judged the relevance of output and completed a number of other measures. Data is analysed both quantitatively, using standard and innovative statistical techniques, and qualitatively, through a grounded theory approach using microanalytic and observational methods

0.0646836 = product of:
  0.2587344 = sum of:
    0.2587344 = weight(_text_:judged in 3923) [ClassicSimilarity], result of:
      0.2587344 = score(doc=3923,freq=2.0), product of:
        0.51079464 = queryWeight, product of:
          7.6410246 = idf(docFreq=57, maxDocs=44421)
          0.06684897 = queryNorm
        0.50653315 = fieldWeight in 3923, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          7.6410246 = idf(docFreq=57, maxDocs=44421)
          0.046875 = fieldNorm(doc=3923)
  0.25 = coord(1/4)

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

0.0646836 = product of:
  0.2587344 = sum of:
    0.2587344 = weight(_text_:judged in 4218) [ClassicSimilarity], result of:
      0.2587344 = score(doc=4218,freq=2.0), product of:
        0.51079464 = queryWeight, product of:
          7.6410246 = idf(docFreq=57, maxDocs=44421)
          0.06684897 = queryNorm
        0.50653315 = fieldWeight in 4218, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          7.6410246 = idf(docFreq=57, maxDocs=44421)
          0.046875 = fieldNorm(doc=4218)
  0.25 = coord(1/4)

Abstract

A recent study begun at Case Western Reserve University and continued at Rutgers University compared the transcripts of 200 DIALOG searches conducted by 36 experienced searchers on real questions submitted by academic and industrial researchers. Relevance judgements by the researchers were used to give recall and precision scores to each search result. Findings included: a low degree of overlap between searches on the same question in selection of search terms or items retrieved; the more often an item was retrieved by different searchers, the more likely it was to be judged relevant; recall and precision were not necessarly inversly related; there was a significant positive impact on recall/precision from using more cycles (a sequence from selecting terms to displaying results); serious uncorrectd errors were a major problem in poor searches and proper selection of terms a key to successful searches.

0.0646836 = product of:
  0.2587344 = sum of:
    0.2587344 = weight(_text_:judged in 206) [ClassicSimilarity], result of:
      0.2587344 = score(doc=206,freq=2.0), product of:
        0.51079464 = queryWeight, product of:
          7.6410246 = idf(docFreq=57, maxDocs=44421)
          0.06684897 = queryNorm
        0.50653315 = fieldWeight in 206, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          7.6410246 = idf(docFreq=57, maxDocs=44421)
          0.046875 = fieldNorm(doc=206)
  0.25 = coord(1/4)

Abstract

We investigated the sources and effectiveness of search terms used during mediated on-line searching under real-life (as opposed to laboratory) circumstances. A stratified model of information retrieval (IR) interaction served as a framework for the analysis. For the analysis, we used the on-line transaction logs, videotapes, and transcribed dialogue of the presearch and on-line interaction between 40 users and 4 professional intermediaries. Each user provided one question and interacted with one of the four intermediaries. Searching was done using DIALOG. Five sources of search terms were identified: (1) the users' written question statements, (2) terms derived from users' domain knowledge during the interaction, (3) terms extracted from retrieved items as relevance feedback, (4) database thesaurus, and (5) terms derived by intermediaries during the interaction. Distribution, retrieval effectiveness, transition sequences, and correlation of search terms from different sources were investigated. Search terms from users' written question statements and term relevance feedback were the most productive sources of terms contributing to the retrieval of items judged relevant by users. Implications of the findings are discussed

0.060866904 = product of:
  0.24346761 = sum of:
    0.24346761 = weight(_text_:headings in 4175) [ClassicSimilarity], result of:
      0.24346761 = score(doc=4175,freq=2.0), product of:
        0.32437804 = queryWeight, product of:
          4.8524013 = idf(docFreq=942, maxDocs=44421)
          0.06684897 = queryNorm
        0.7505675 = fieldWeight in 4175, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          4.8524013 = idf(docFreq=942, maxDocs=44421)
          0.109375 = fieldNorm(doc=4175)
  0.25 = coord(1/4)

Footnote

Übers. d. Titels: Title words - subject headings: a comparative research

0.053903006 = product of:
  0.21561202 = sum of:
    0.21561202 = weight(_text_:judged in 175) [ClassicSimilarity], result of:
      0.21561202 = score(doc=175,freq=2.0), product of:
        0.51079464 = queryWeight, product of:
          7.6410246 = idf(docFreq=57, maxDocs=44421)
          0.06684897 = queryNorm
        0.42211097 = fieldWeight in 175, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          7.6410246 = idf(docFreq=57, maxDocs=44421)
          0.0390625 = fieldNorm(doc=175)
  0.25 = coord(1/4)

Abstract

Griesdorf and O'Connor attempt to determine the aspects of a retrieved item that provide a questioner with evidence that the item is in fact on the topic searched independent of its relevance. To this end they collect data from 32 participants, 11 from the business community as well as 21 doctoral students at the University of North Texas each of whom were asked to state if they considered material that approaches a topic in each of 14 specific manners as " on topic" or "off topic." Chi-square indicates that the observed values are significantly different from expected values and the chi-square residuals for on topic judgements exceed plus or minus two in eight cases and plus two in five cases. The positive values which indicate a percentage of response greater than that from chance suggest that documents considered topical are only related to the problem at hand, contain terms that were in the query, and describe, explain or expand the topic of the query. The chi-square residuals for off topic judgements exceed plus or minus two in ten cases and plus two in four cases. The positive values suggest that documents considered not topical exhibit a contrasting, contrary, or confounding point of view, or merely spark curiosity. Such material might well be relevant, but is not judged topical. This suggests that topical appropriateness may best be achieved using the Bruza, et alia, left compositional monotonicity approach.

0.053903006 = product of:
  0.21561202 = sum of:
    0.21561202 = weight(_text_:judged in 98) [ClassicSimilarity], result of:
      0.21561202 = score(doc=98,freq=2.0), product of:
        0.51079464 = queryWeight, product of:
          7.6410246 = idf(docFreq=57, maxDocs=44421)
          0.06684897 = queryNorm
        0.42211097 = fieldWeight in 98, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          7.6410246 = idf(docFreq=57, maxDocs=44421)
          0.0390625 = fieldNorm(doc=98)
  0.25 = coord(1/4)

Abstract

Evaluating Information Retrieval systems is crucial to making progress in search technologies. Evaluation is often based on assembling reference collections consisting of documents, queries and relevance judgments done by humans. In large-scale environments, exhaustively judging relevance becomes infeasible. Instead, only a pool of documents is judged for relevance. By selectively choosing documents from the pool we can optimize the number of judgments required to identify a given number of relevant documents. We argue that this iterative selection process can be naturally modeled as a reinforcement learning problem and propose innovative and formal adjudication methods based on multi-armed bandits. Casting document judging as a multi-armed bandit problem is not only theoretically appealing, but also leads to highly effective adjudication methods. Under this bandit allocation framework, we consider stationary and non-stationary models and propose seven new document adjudication methods (five stationary methods and two non-stationary variants). Our paper also reports a series of experiments performed to thoroughly compare our new methods against current adjudication methods. This comparative study includes existing methods designed for pooling-based evaluation and existing methods designed for metasearch. Our experiments show that our theoretically grounded adjudication methods can substantially minimize the assessment effort.

0.052171633 = product of:
  0.20868653 = sum of:
    0.20868653 = weight(_text_:headings in 5276) [ClassicSimilarity], result of:
      0.20868653 = score(doc=5276,freq=2.0), product of:
        0.32437804 = queryWeight, product of:
          4.8524013 = idf(docFreq=942, maxDocs=44421)
          0.06684897 = queryNorm
        0.64334357 = fieldWeight in 5276, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          4.8524013 = idf(docFreq=942, maxDocs=44421)
          0.09375 = fieldNorm(doc=5276)
  0.25 = coord(1/4)

0.045181964 = product of:
  0.18072785 = sum of:
    0.18072785 = weight(_text_:headings in 386) [ClassicSimilarity], result of:
      0.18072785 = score(doc=386,freq=6.0), product of:
        0.32437804 = queryWeight, product of:
          4.8524013 = idf(docFreq=942, maxDocs=44421)
          0.06684897 = queryNorm
        0.5571519 = fieldWeight in 386, product of:
          2.4494898 = tf(freq=6.0), with freq of:
            6.0 = termFreq=6.0
          4.8524013 = idf(docFreq=942, maxDocs=44421)
          0.046875 = fieldNorm(doc=386)
  0.25 = coord(1/4)

Abstract

Using subject headings for Online Retrieval is an indispensable tool for online system desingners who are developing new systems or refining exicting ones. The book describes subject analysis and subject searching in online catalogs, including the limitations of retrieval, and demonstrates how such limitations can be overcome through system design and programming. The book describes the Library of Congress Subject headings system and system characteristics, shows how information is stored in machine readable files, and offers examples of and recommendations for successful methods. Tables are included to support these recommendations, and diagrams, graphs, and bar charts are used to provide results of data analyses.

0.0430394 = product of:
  0.1721576 = sum of:
    0.1721576 = weight(_text_:headings in 1603) [ClassicSimilarity], result of:
      0.1721576 = score(doc=1603,freq=4.0), product of:
        0.32437804 = queryWeight, product of:
          4.8524013 = idf(docFreq=942, maxDocs=44421)
          0.06684897 = queryNorm
        0.5307314 = fieldWeight in 1603, product of:
          2.0 = tf(freq=4.0), with freq of:
            4.0 = termFreq=4.0
          4.8524013 = idf(docFreq=942, maxDocs=44421)
          0.0546875 = fieldNorm(doc=1603)
  0.25 = coord(1/4)

Abstract

We have investigated the relative merits of searching on titles, subject headings, abstracts, free-language terms, and combinations of these elements. The COMPENDEX data base was used for this study since it combined all of these data elements of interest. In general, the results obtained from the experiments indicate that, as expected, titles alone are not satisfactory for efficient retrieval. The combination of titles and abstracts came the closest to 100% retrieval, with searching of abstracts alone doing almost as well. Indexer input, although necessary for 100% retrieval in almost all cases, was found to be relatively unimportant

0.030433452 = product of:
  0.12173381 = sum of:
    0.12173381 = weight(_text_:headings in 5666) [ClassicSimilarity], result of:
      0.12173381 = score(doc=5666,freq=2.0), product of:
        0.32437804 = queryWeight, product of:
          4.8524013 = idf(docFreq=942, maxDocs=44421)
          0.06684897 = queryNorm
        0.37528375 = fieldWeight in 5666, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          4.8524013 = idf(docFreq=942, maxDocs=44421)
          0.0546875 = fieldNorm(doc=5666)
  0.25 = coord(1/4)

Abstract

Research shows that 65-80% of subject search terms fail to match the appropriate subject heading and one third to one half of subject searches result in no references being retrieved. Examines the subject search terms geberated by 82 school and college students in Princeton, NJ, evaluated the match between the named terms and the expected subject headings, proposes an explanation for match failures in relation to 3 invariant properties common to all search terms: concreteness, complexity, and syndeticity. Suggests that match failure is a consequence of developmental naming patterns and that these patterns can be overcome through the use of metacognitive naming skills