Search (55 results, page 1 of 3)

0.017033914 = product of:
  0.03406783 = sum of:
    0.021117793 = weight(_text_:data in 57) [ClassicSimilarity], result of:
      0.021117793 = score(doc=57,freq=2.0), product of:
        0.120893985 = queryWeight, product of:
          3.1620505 = idf(docFreq=5088, maxDocs=44218)
          0.03823278 = queryNorm
        0.17468026 = fieldWeight in 57, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          3.1620505 = idf(docFreq=5088, maxDocs=44218)
          0.0390625 = fieldNorm(doc=57)
    0.012950035 = product of:
      0.02590007 = sum of:
        0.02590007 = weight(_text_:22 in 57) [ClassicSimilarity], result of:
          0.02590007 = score(doc=57,freq=2.0), product of:
            0.13388468 = queryWeight, product of:
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.03823278 = queryNorm
            0.19345059 = fieldWeight in 57, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.0390625 = fieldNorm(doc=57)
      0.5 = coord(1/2)
  0.5 = coord(2/4)

Abstract

Webometric network analyses have been used to map the connectivity of groups of websites to identify clusters, important sites or overall structure. Such analyses have mainly been based upon hyperlink counts, the number of hyperlinks between a pair of websites, although some have used title mentions or URL citations instead. The ability to automatically gather hyperlink counts from Yahoo! ceased in April 2011 and the ability to manually gather such counts was due to cease by early 2012, creating a need for alternatives. This article assesses URL citations and title mentions as possible replacements for hyperlinks in both binary and weighted direct link and co-inlink network diagrams. It also assesses three different types of data for the network connections: hit count estimates, counts of matching URLs, and filtered counts of matching URLs. Results from analyses of U.S. library and information science departments and U.K. universities give evidence that metrics based upon URLs or titles can be appropriate replacements for metrics based upon hyperlinks for both binary and weighted networks, although filtered counts of matching URLs are necessary to give the best results for co-title mention and co-URL citation network diagrams.

Date

6. 4.2012 18:16:22

0.014932535 = product of:
  0.05973014 = sum of:
    0.05973014 = weight(_text_:data in 4279) [ClassicSimilarity], result of:
      0.05973014 = score(doc=4279,freq=16.0), product of:
        0.120893985 = queryWeight, product of:
          3.1620505 = idf(docFreq=5088, maxDocs=44218)
          0.03823278 = queryNorm
        0.49407038 = fieldWeight in 4279, product of:
          4.0 = tf(freq=16.0), with freq of:
            16.0 = termFreq=16.0
          3.1620505 = idf(docFreq=5088, maxDocs=44218)
          0.0390625 = fieldNorm(doc=4279)
  0.25 = coord(1/4)

Abstract

Webometrics, the quantitative study of Web-related phenomena, emerged from the realization that methods originally designed for bibliometric analysis of scientific journal article citation patterns could be applied to the Web, with commercial search engines providing the raw data. Almind and Ingwersen (1997) defined the field and gave it its name. Other pioneers included Rodriguez Gairin (1997) and Aguillo (1998). Larson (1996) undertook exploratory link structure analysis, as did Rousseau (1997). Webometrics encompasses research from fields beyond information science such as communication studies, statistical physics, and computer science. In this review we concentrate on link analysis, but also cover other aspects of webometrics, including Web log fle analysis. One theme that runs through this chapter is the messiness of Web data and the need for data cleansing heuristics. The uncontrolled Web creates numerous problems in the interpretation of results, for instance, from the automatic creation or replication of links. The loose connection between top-level domain specifications (e.g., com, edu, and org) and their actual content is also a frustrating problem. For example, many .com sites contain noncommercial content, although com is ostensibly the main commercial top-level domain. Indeed, a skeptical researcher could claim that obstacles of this kind are so great that all Web analyses lack value. As will be seen, one response to this view, a view shared by critics of evaluative bibliometrics, is to demonstrate that Web data correlate significantly with some non-Web data in order to prove that the Web data are not wholly random. A practical response has been to develop increasingly sophisticated data cleansing techniques and multiple data analysis methods.

0.014166246 = product of:
  0.056664985 = sum of:
    0.056664985 = weight(_text_:data in 2313) [ClassicSimilarity], result of:
      0.056664985 = score(doc=2313,freq=10.0), product of:
        0.120893985 = queryWeight, product of:
          3.1620505 = idf(docFreq=5088, maxDocs=44218)
          0.03823278 = queryNorm
        0.46871632 = fieldWeight in 2313, product of:
          3.1622777 = tf(freq=10.0), with freq of:
            10.0 = termFreq=10.0
          3.1620505 = idf(docFreq=5088, maxDocs=44218)
          0.046875 = fieldNorm(doc=2313)
  0.25 = coord(1/4)

Abstract

Purpose The purpose of this paper is to make an explicit case for the use of data with contextual information as evidence in arts and humanities research evaluations rather than systematic metrics. Design/methodology/approach A survey of the strengths and limitations of citation-based indicators is combined with evidence about existing uses of wider impact data in the arts and humanities, with particular reference to the 2014 UK Research Excellence Framework. Findings Data are already used as impact evidence in the arts and humanities but this practice should become more widespread. Practical implications Arts and humanities researchers should be encouraged to think creatively about the kinds of data that they may be able to generate in support of the value of their research and should not rely upon standardised metrics. Originality/value This paper combines practices emerging in the arts and humanities with research evaluation from a scientometric perspective to generate new recommendations.

0.012670675 = product of:
  0.0506827 = sum of:
    0.0506827 = weight(_text_:data in 4654) [ClassicSimilarity], result of:
      0.0506827 = score(doc=4654,freq=2.0), product of:
        0.120893985 = queryWeight, product of:
          3.1620505 = idf(docFreq=5088, maxDocs=44218)
          0.03823278 = queryNorm
        0.4192326 = fieldWeight in 4654, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          3.1620505 = idf(docFreq=5088, maxDocs=44218)
          0.09375 = fieldNorm(doc=4654)
  0.25 = coord(1/4)

0.012670675 = product of:
  0.0506827 = sum of:
    0.0506827 = weight(_text_:data in 3322) [ClassicSimilarity], result of:
      0.0506827 = score(doc=3322,freq=8.0), product of:
        0.120893985 = queryWeight, product of:
          3.1620505 = idf(docFreq=5088, maxDocs=44218)
          0.03823278 = queryNorm
        0.4192326 = fieldWeight in 3322, product of:
          2.828427 = tf(freq=8.0), with freq of:
            8.0 = termFreq=8.0
          3.1620505 = idf(docFreq=5088, maxDocs=44218)
          0.046875 = fieldNorm(doc=3322)
  0.25 = coord(1/4)

Abstract

Despite the rapid growth in social network sites and in data mining for emotion (sentiment analysis), little research has tied the two together, and none has had social science goals. This article examines the extent to which emotion is present in MySpace comments, using a combination of data mining and content analysis, and exploring age and gender. A random sample of 819 public comments to or from U.S. users was manually classified for strength of positive and negative emotion. Two thirds of the comments expressed positive emotion, but a minority (20%) contained negative emotion, confirming that MySpace is an extraordinarily emotion-rich environment. Females are likely to give and receive more positive comments than are males, but there is no difference for negative comments. It is thus possible that females are more successful social network site users partly because of their greater ability to textually harness positive affect.

Theme

Data Mining

0.010452774 = product of:
  0.041811097 = sum of:
    0.041811097 = weight(_text_:data in 1969) [ClassicSimilarity], result of:
      0.041811097 = score(doc=1969,freq=4.0), product of:
        0.120893985 = queryWeight, product of:
          3.1620505 = idf(docFreq=5088, maxDocs=44218)
          0.03823278 = queryNorm
        0.34584928 = fieldWeight in 1969, product of:
          2.0 = tf(freq=4.0), with freq of:
            4.0 = termFreq=4.0
          3.1620505 = idf(docFreq=5088, maxDocs=44218)
          0.0546875 = fieldNorm(doc=1969)
  0.25 = coord(1/4)

Abstract

New mass publishing genres, such as blogs and personal home pages provide a rich source of social data that is yet to be fully exploited by the social sciences and humanities. Information-centered research (ICR) not only provides a genuinely new and useful information science research model for this type of data, but can also contribute to the emerging e-research infrastructure. Nevertheless, ICR should not be conducted on a purely abstract level, but should relate to potentially relevant problems.

0.009144273 = product of:
  0.03657709 = sum of:
    0.03657709 = weight(_text_:data in 6116) [ClassicSimilarity], result of:
      0.03657709 = score(doc=6116,freq=6.0), product of:
        0.120893985 = queryWeight, product of:
          3.1620505 = idf(docFreq=5088, maxDocs=44218)
          0.03823278 = queryNorm
        0.30255508 = fieldWeight in 6116, product of:
          2.4494898 = tf(freq=6.0), with freq of:
            6.0 = termFreq=6.0
          3.1620505 = idf(docFreq=5088, maxDocs=44218)
          0.0390625 = fieldNorm(doc=6116)
  0.25 = coord(1/4)

Abstract

Broad issue scanning is the task of identifying important public debates arising in a given broad issue; really simple syndication (RSS) feeds are a natural information source for investigating broad issues. RSS, as originally conceived, is a method for publishing timely and concise information on the Internet, for example, about the main stories in a news site or the latest postings in a blog. RSS feeds are potentially a nonintrusive source of high-quality data about public opinion: Monitoring a large number may allow quantitative methods to extract information relevant to a given need. In this article we describe an RSS feed-based coword frequency method to identify bursts of discussion relevant to a given broad issue. A case study of public science concerns is used to demonstrate the method and assess the suitability of raw RSS feeds for broad issue scanning (i.e., without data cleansing). An attempt to identify genuine science concern debates from the corpus through investigating the top 1,000 "burst" words found only two genuine debates, however. The low success rate was mainly caused by a few pathological feeds that dominated the results and obscured any significant debates. The results point to the need to develop effective data cleansing procedures for RSS feeds, particularly if there is not a large quantity of discussion about the broad issue, and a range of potential techniques is suggested. Finally, the analysis confirmed that the time series information generated by real-time monitoring of RSS feeds could usefully illustrate the evolution of new debates relevant to a broad issue.

0.009144273 = product of:
  0.03657709 = sum of:
    0.03657709 = weight(_text_:data in 1401) [ClassicSimilarity], result of:
      0.03657709 = score(doc=1401,freq=6.0), product of:
        0.120893985 = queryWeight, product of:
          3.1620505 = idf(docFreq=5088, maxDocs=44218)
          0.03823278 = queryNorm
        0.30255508 = fieldWeight in 1401, product of:
          2.4494898 = tf(freq=6.0), with freq of:
            6.0 = termFreq=6.0
          3.1620505 = idf(docFreq=5088, maxDocs=44218)
          0.0390625 = fieldNorm(doc=1401)
  0.25 = coord(1/4)

Abstract

Purpose - The purpose of this paper is to claim that it is useful to assess the web connectivity of research groups, describe hyperlink-based techniques to achieve this and present brief details of European life sciences research groups as a case study. Design/methodology/approach - A commercial search engine was harnessed to deliver hyperlink data via its automatic query submission interface. A special purpose link analysis tool, LexiURL, then summarised and graphed the link data in appropriate ways. Findings - Webometrics can provide a wide range of descriptive information about the international connectivity of research groups. Research limitations/implications - Only one field was analysed, data was taken from only one search engine, and the results were not validated. Practical implications - Web connectivity seems to be particularly important for attracting overseas job applicants and to promote research achievements and capabilities, and hence we contend that it can be useful for national and international governments to use webometrics to ensure that the web is being used effectively by research groups. Originality/value - This is the first paper to make a case for the value of using a range of webometric techniques to evaluate the web presences of research groups within a field, and possibly the first "applied" webometrics study produced for an external contract.

0.009144273 = product of:
  0.03657709 = sum of:
    0.03657709 = weight(_text_:data in 2190) [ClassicSimilarity], result of:
      0.03657709 = score(doc=2190,freq=6.0), product of:
        0.120893985 = queryWeight, product of:
          3.1620505 = idf(docFreq=5088, maxDocs=44218)
          0.03823278 = queryNorm
        0.30255508 = fieldWeight in 2190, product of:
          2.4494898 = tf(freq=6.0), with freq of:
            6.0 = termFreq=6.0
          3.1620505 = idf(docFreq=5088, maxDocs=44218)
          0.0390625 = fieldNorm(doc=2190)
  0.25 = coord(1/4)

Abstract

Although there is evidence that counting the readers of an article in the social reference site, Mendeley, may help to capture its research impact, the extent to which this is true for different scientific fields is unknown. In this study, we compare Mendeley readership counts with citations for different social sciences and humanities disciplines. The overall correlation between Mendeley readership counts and citations for the social sciences was higher than for the humanities. Low and medium correlations between Mendeley bookmarks and citation counts in all the investigated disciplines suggest that these measures reflect different aspects of research impact. Mendeley data were also used to discover patterns of information flow between scientific fields. Comparing information flows based on Mendeley bookmarking data and cross-disciplinary citation analysis for the disciplines revealed substantial similarities and some differences. Thus, the evidence from this study suggests that Mendeley readership data could be used to help capture knowledge transfer across scientific disciplines, especially for people that read but do not author articles, as well as giving impact evidence at an earlier stage than is possible with citation counts.

0.009144273 = product of:
  0.03657709 = sum of:
    0.03657709 = weight(_text_:data in 3226) [ClassicSimilarity], result of:
      0.03657709 = score(doc=3226,freq=6.0), product of:
        0.120893985 = queryWeight, product of:
          3.1620505 = idf(docFreq=5088, maxDocs=44218)
          0.03823278 = queryNorm
        0.30255508 = fieldWeight in 3226, product of:
          2.4494898 = tf(freq=6.0), with freq of:
            6.0 = termFreq=6.0
          3.1620505 = idf(docFreq=5088, maxDocs=44218)
          0.0390625 = fieldNorm(doc=3226)
  0.25 = coord(1/4)

Abstract

Scholars writing books that are widely used to support teaching in higher education may be undervalued because of a lack of evidence of teaching value. Although sales data may give credible evidence for textbooks, these data may poorly reflect educational uses of other types of books. As an alternative, this article proposes a method to search automatically for mentions of books in online academic course syllabi based on Bing searches for syllabi mentioning a given book, filtering out false matches through an extensive set of rules. The method had an accuracy of over 90% based on manual checks of a sample of 2,600 results from the initial Bing searches. Over one third of about 14,000 monographs checked had one or more academic syllabus mention, with more in the arts and humanities (56%) and social sciences (52%). Low but significant correlations between syllabus mentions and citations across most fields, except the social sciences, suggest that books tend to have different levels of impact for teaching and research. In conclusion, the automatic syllabus search method gives a new way to estimate the educational utility of books in a way that sales data and citation counts cannot.

0.008959521 = product of:
  0.035838082 = sum of:
    0.035838082 = weight(_text_:data in 5906) [ClassicSimilarity], result of:
      0.035838082 = score(doc=5906,freq=4.0), product of:
        0.120893985 = queryWeight, product of:
          3.1620505 = idf(docFreq=5088, maxDocs=44218)
          0.03823278 = queryNorm
        0.29644224 = fieldWeight in 5906, product of:
          2.0 = tf(freq=4.0), with freq of:
            4.0 = termFreq=4.0
          3.1620505 = idf(docFreq=5088, maxDocs=44218)
          0.046875 = fieldNorm(doc=5906)
  0.25 = coord(1/4)

Abstract

In this article Web issue analysis is introduced as a new technique to investigate an issue as reflected on the Web. The issue chosen, integrated water resource management (IWRM), is a United Nations-initiated paradigm for managing water resources in an international context, particularly in developing nations. As with many international governmental initiatives, there is a considerable body of online information about it: 41.381 hypertext markup language (HTML) pages and 28.735 PDF documents mentioning the issue were downloaded. A page uniform resource locator (URL) and link analysis revealed the international and sectoral spread of IWRM. A noun and noun phrase occurrence analysis was used to identify the issues most commonly discussed, revealing some unexpected topics such as private sector and economic growth. Although the complexity of the methods required to produce meaningful statistics from the data is disadvantageous to easy interpretation, it was still possible to produce data that could be subject to a reasonably intuitive interpretation. Hence Web issue analysis is claimed to be a useful new technique for information science.

0.008959521 = product of:
  0.035838082 = sum of:
    0.035838082 = weight(_text_:data in 1004) [ClassicSimilarity], result of:
      0.035838082 = score(doc=1004,freq=4.0), product of:
        0.120893985 = queryWeight, product of:
          3.1620505 = idf(docFreq=5088, maxDocs=44218)
          0.03823278 = queryNorm
        0.29644224 = fieldWeight in 1004, product of:
          2.0 = tf(freq=4.0), with freq of:
            4.0 = termFreq=4.0
          3.1620505 = idf(docFreq=5088, maxDocs=44218)
          0.046875 = fieldNorm(doc=1004)
  0.25 = coord(1/4)

Abstract

General sentiment analysis for the social web has become increasingly useful for shedding light on the role of emotion in online communication and offline events in both academic research and data journalism. Nevertheless, existing general-purpose social web sentiment analysis algorithms may not be optimal for texts focussed around specific topics. This article introduces 2 new methods, mood setting and lexicon extension, to improve the accuracy of topic-specific lexical sentiment strength detection for the social web. Mood setting allows the topic mood to determine the default polarity for ostensibly neutral expressive text. Topic-specific lexicon extension involves adding topic-specific words to the default general sentiment lexicon. Experiments with 8 data sets show that both methods can improve sentiment analysis performance in corpora and are recommended when the topic focus is tightest.

0.0074662673 = product of:
  0.02986507 = sum of:
    0.02986507 = weight(_text_:data in 1236) [ClassicSimilarity], result of:
      0.02986507 = score(doc=1236,freq=4.0), product of:
        0.120893985 = queryWeight, product of:
          3.1620505 = idf(docFreq=5088, maxDocs=44218)
          0.03823278 = queryNorm
        0.24703519 = fieldWeight in 1236, product of:
          2.0 = tf(freq=4.0), with freq of:
            4.0 = termFreq=4.0
          3.1620505 = idf(docFreq=5088, maxDocs=44218)
          0.0390625 = fieldNorm(doc=1236)
  0.25 = coord(1/4)

Abstract

Web links have been studied by information scientists for at least six years but it is only in the past two that clear evidence has emerged to show that counts of links to scholarly Web spaces (universities and departments) can correlate significantly with research measures, giving some credence to their use for the investigation of scholarly communication. This paper reports an a study to investigate the factors that influence the creation of links to journal Web sites. An empirical approach is used: collecting data and testing for significant patterns. The specific questions addressed are whether site age and site content are inducers of links to a journal's Web site as measured by the ratio of link counts to Journal Impact Factors, two variables previously discovered to be related. A new methodology for data collection is also introduced that uses the Internet Archive to obtain an earliest known creation date for Web sites. The results show that both site age and site content are significant factors for the disciplines studied: library and information science, and law. Comparisons between the two fields also show disciplinary differences in Web site characteristics. Scholars and publishers should be particularly aware that richer content an a journal's Web site tends to generate links and thus the traffic to the site.

0.0074662673 = product of:
  0.02986507 = sum of:
    0.02986507 = weight(_text_:data in 1383) [ClassicSimilarity], result of:
      0.02986507 = score(doc=1383,freq=4.0), product of:
        0.120893985 = queryWeight, product of:
          3.1620505 = idf(docFreq=5088, maxDocs=44218)
          0.03823278 = queryNorm
        0.24703519 = fieldWeight in 1383, product of:
          2.0 = tf(freq=4.0), with freq of:
            4.0 = termFreq=4.0
          3.1620505 = idf(docFreq=5088, maxDocs=44218)
          0.0390625 = fieldNorm(doc=1383)
  0.25 = coord(1/4)

Abstract

Web links have been used for around ten years to explore the online impact of academic information and information producers. Nevertheless, few studies have attempted to relate link counts to relevant offline attributes of the owners of the targeted Web sites, with the exception of research productivity. This article reports the results of a study to relate site inlink counts to relevant owner characteristics for over 400 European life-science research group Web sites. The analysis confirmed that research-group size and Web-presence size were important for attracting Web links, although research productivity was not. Little evidence was found for significant influence of any of an array of factors, including research-group leader gender and industry connections. In addition, the choice of search engine for link data created a surprising international difference in the results, with Google perhaps giving unreliable results. Overall, the data collection, statistical analysis and results interpretation were all complex and it seems that we still need to know more about search engines, hyperlinks, and their function in science before we can draw conclusions on their usefulness and role in the canon of science and technology indicators.

0.0074662673 = product of:
  0.02986507 = sum of:
    0.02986507 = weight(_text_:data in 1883) [ClassicSimilarity], result of:
      0.02986507 = score(doc=1883,freq=4.0), product of:
        0.120893985 = queryWeight, product of:
          3.1620505 = idf(docFreq=5088, maxDocs=44218)
          0.03823278 = queryNorm
        0.24703519 = fieldWeight in 1883, product of:
          2.0 = tf(freq=4.0), with freq of:
            4.0 = termFreq=4.0
          3.1620505 = idf(docFreq=5088, maxDocs=44218)
          0.0390625 = fieldNorm(doc=1883)
  0.25 = coord(1/4)

Abstract

In 2007, the social networking Web site MySpace apparently overthrew Google as the most visited Web site for U.S. Web users. If this heralds a new era of widespread online social networking, then it is important to investigate user behaviour and attributes. Although there has been some research into social networking already, basic demographic data is essential to set previous results in a wider context and to give insights to researchers, marketers and developers. In this article, the demographics of MySpace members are explored through data extracted from two samples of 15,043 and 7,627 member profiles. The median declared age of users was surprisingly high at 21, with a small majority of females. The analysis confirmed some previously reported findings and conjectures about social networking, for example, that female members tend to be more interested in friendship and males more interested in dating. In addition, there was some evidence of three different friending dynamics, oriented towards close friends, acquaintances, or strangers. Perhaps unsurprisingly, female and younger members had more friends than others, and females were more likely to maintain private profiles, but both males and females seemed to prefer female friends, with this tendency more marked in females for their closest friend. The typical MySpace user is apparently female, 21, single, with a public profile, interested in online friendship and logging on weekly to engage with a mixed list of mainly female friends who are predominantly acquaintances.

0.0074662673 = product of:
  0.02986507 = sum of:
    0.02986507 = weight(_text_:data in 2946) [ClassicSimilarity], result of:
      0.02986507 = score(doc=2946,freq=4.0), product of:
        0.120893985 = queryWeight, product of:
          3.1620505 = idf(docFreq=5088, maxDocs=44218)
          0.03823278 = queryNorm
        0.24703519 = fieldWeight in 2946, product of:
          2.0 = tf(freq=4.0), with freq of:
            4.0 = termFreq=4.0
          3.1620505 = idf(docFreq=5088, maxDocs=44218)
          0.0390625 = fieldNorm(doc=2946)
  0.25 = coord(1/4)

Abstract

In both the social sciences and the humanities, books and monographs play significant roles in research communication. The absence of citations from most books and monographs from the Thomson Reuters/Institute for Scientific Information databases (ISI) has been criticized, but attempts to include citations from or to books in the research evaluation of the social sciences and humanities have not led to widespread adoption. This article assesses whether Google Book Search (GBS) can partially fill this gap by comparing citations from books with citations from journal articles to journal articles in 10 science, social science, and humanities disciplines. Book citations were 31% to 212% of ISI citations and, hence, numerous enough to supplement ISI citations in the social sciences and humanities covered, but not in the sciences (3%-5%), except for computing (46%), due to numerous published conference proceedings. A case study was also made of all 1,923 articles in the 51 information science and library science ISI-indexed journals published in 2003. Within this set, highly book-cited articles tended to receive many ISI citations, indicating a significant relationship between the two types of citation data, but with important exceptions that point to the additional information provided by book citations. In summary, GBS is clearly a valuable new source of citation data for the social sciences and humanities. One practical implication is that book-oriented scholars should consult it for additional citations to their work when applying for promotion and tenure.

0.0074662673 = product of:
  0.02986507 = sum of:
    0.02986507 = weight(_text_:data in 4972) [ClassicSimilarity], result of:
      0.02986507 = score(doc=4972,freq=4.0), product of:
        0.120893985 = queryWeight, product of:
          3.1620505 = idf(docFreq=5088, maxDocs=44218)
          0.03823278 = queryNorm
        0.24703519 = fieldWeight in 4972, product of:
          2.0 = tf(freq=4.0), with freq of:
            4.0 = termFreq=4.0
          3.1620505 = idf(docFreq=5088, maxDocs=44218)
          0.0390625 = fieldNorm(doc=4972)
  0.25 = coord(1/4)

Abstract

Sentiment analysis is concerned with the automatic extraction of sentiment-related information from text. Although most sentiment analysis addresses commercial tasks, such as extracting opinions from product reviews, there is increasing interest in the affective dimension of the social web, and Twitter in particular. Most sentiment analysis algorithms are not ideally suited to this task because they exploit indirect indicators of sentiment that can reflect genre or topic instead. Hence, such algorithms used to process social web texts can identify spurious sentiment patterns caused by topics rather than affective phenomena. This article assesses an improved version of the algorithm SentiStrength for sentiment strength detection across the social web that primarily uses direct indications of sentiment. The results from six diverse social web data sets (MySpace, Twitter, YouTube, Digg, Runners World, BBC Forums) indicate that SentiStrength 2 is successful in the sense of performing better than a baseline approach for all data sets in both supervised and unsupervised cases. SentiStrength is not always better than machine-learning approaches that exploit indirect indicators of sentiment, however, and is particularly weaker for positive sentiment in news-related discussions. Overall, the results suggest that, even unsupervised, SentiStrength is robust enough to be applied to a wide variety of different social web contexts.

0.0074662673 = product of:
  0.02986507 = sum of:
    0.02986507 = weight(_text_:data in 1229) [ClassicSimilarity], result of:
      0.02986507 = score(doc=1229,freq=4.0), product of:
        0.120893985 = queryWeight, product of:
          3.1620505 = idf(docFreq=5088, maxDocs=44218)
          0.03823278 = queryNorm
        0.24703519 = fieldWeight in 1229, product of:
          2.0 = tf(freq=4.0), with freq of:
            4.0 = termFreq=4.0
          3.1620505 = idf(docFreq=5088, maxDocs=44218)
          0.0390625 = fieldNorm(doc=1229)
  0.25 = coord(1/4)

Abstract

Data collected by social media platforms have been introduced as new sources for indicators to help measure the impact of scholarly research in ways that are complementary to traditional citation analysis. Data generated from social media activities can be used to reflect broad types of impact. This article aims to provide systematic evidence about how often Twitter is used to disseminate information about journal articles in the biomedical sciences. The analysis is based on 1.4 million documents covered by both PubMed and Web of Science and published between 2010 and 2012. The number of tweets containing links to these documents was analyzed and compared to citations to evaluate the degree to which certain journals, disciplines, and specialties were represented on Twitter and how far tweets correlate with citation impact. With less than 10% of PubMed articles mentioned on Twitter, its uptake is low in general but differs between journals and specialties. Correlations between tweets and citations are low, implying that impact metrics based on tweets are different from those based on citations. A framework using the coverage of articles and the correlation between Twitter mentions and citations is proposed to facilitate the evaluation of novel social-media-based metrics.

0.0074662673 = product of:
  0.02986507 = sum of:
    0.02986507 = weight(_text_:data in 2162) [ClassicSimilarity], result of:
      0.02986507 = score(doc=2162,freq=4.0), product of:
        0.120893985 = queryWeight, product of:
          3.1620505 = idf(docFreq=5088, maxDocs=44218)
          0.03823278 = queryNorm
        0.24703519 = fieldWeight in 2162, product of:
          2.0 = tf(freq=4.0), with freq of:
            4.0 = termFreq=4.0
          3.1620505 = idf(docFreq=5088, maxDocs=44218)
          0.0390625 = fieldNorm(doc=2162)
  0.25 = coord(1/4)

Abstract

Little detailed information is known about who reads research articles and the contexts in which research articles are read. Using data about people who register in Mendeley as readers of articles, this article explores different types of users of Clinical Medicine, Engineering and Technology, Social Science, Physics, and Chemistry articles inside and outside academia. The majority of readers for all disciplines were PhD students, postgraduates, and postdocs but other types of academics were also represented. In addition, many Clinical Medicine articles were read by medical professionals. The highest correlations between citations and Mendeley readership counts were found for types of users who often authored academic articles, except for associate professors in some sub-disciplines. This suggests that Mendeley readership can reflect usage similar to traditional citation impact if the data are restricted to readers who are also authors without the delay of impact measured by citation counts. At the same time, Mendeley statistics can also reveal the hidden impact of some research articles, such as educational value for nonauthor users inside academia or the impact of research articles on practice for readers outside academia.

0.0073912274 = product of:
  0.02956491 = sum of:
    0.02956491 = weight(_text_:data in 1066) [ClassicSimilarity], result of:
      0.02956491 = score(doc=1066,freq=2.0), product of:
        0.120893985 = queryWeight, product of:
          3.1620505 = idf(docFreq=5088, maxDocs=44218)
          0.03823278 = queryNorm
        0.24455236 = fieldWeight in 1066, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          3.1620505 = idf(docFreq=5088, maxDocs=44218)
          0.0546875 = fieldNorm(doc=1066)
  0.25 = coord(1/4)

Abstract

Previous studies of academic web interlinking have tended to hypothesise that the relationship between the research of a university and links to or from its web site should follow a linear trend, yet the typical distribution of web data, in general, seems to be a non-linear power law. This paper assesses whether a linear trend or a power law is the most appropriate method with which to model the relationship between research and web site size or outlinks. Following linear regression, analysis of the confidence intervals for the logarithmic graphs, and analysis of the outliers, the results suggest that a linear trend is more appropriate than a non-linear power law.