Revealing Interdisciplinarity in Nanoscience and Technology Queries: A Transaction Log Analysis Approach
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Knowl. Org. 38(2011)No.2 135
A. Shiri. Revealing Interdisciplinarity in Nanoscience and Technology Queries: A Transaction Log Analysis Approach
Revealing Interdisciplinarity in
Nanoscience and Technology Queries:
A Transaction Log Analysis Approach
Ali Shiri*
School of Library and Information Studies, University of Alberta,
3-09 Rutherford South, Edmonton, T6G 2J4, Alberta, Canada,
Ali Shiri is an Associate Professor in the School of Library and Information Studies at the University of
Alberta, Edmonton, Canada. Ali’s research areas and interests include: user search and interaction be-
havior, digital library user interfaces, knowledge organization systems in digital libraries, institutional
repositories, and subject gateways and social tagging trends and interfaces.
Shiri, Ali. Revealing Interdisciplinarity in Nanoscience and Technology Queries: A Transaction Log
Analysis Approach. Knowledge Organization, 38(2), 135-153. 86 references.
ABSTRACT: The study reported here investigated the search behaviour patterns of nanoscience and
nanotechnology searchers as revealed by transaction log analysis of the NANOnetBASE electronic
book digital library. This paper examines the patterns and strategies of nano searchers’ query formulation and reformulation, then
explores the extent of interdisciplinarity in search queries using the INSPEC and Compendex thesauri. The results show certain
query formulation patterns associated with searching in an emerging and interdisciplinary area of nanotechnology such as: the
use of multiword and compound query terms, extensive use of search terms beginning with the prefix “nano,” hyphenated terms,
spelling variations, a large number of query reformulations, and the use of acronyms. The results also indicate that 62% of the
unique top terms resulting from mapping users’ query terms to the INSPEC Classification codes represented two or more disci-
plines, specifically terms associated with the Classification code “A” representing “physics.” The results have implications for in-
formation organization and representation, user interface design and federated searching in digital libraries and multi-subject da-
tabases.
Received 11 February 2010; Revised 22 May 2010; Accepted 22 May 2010
* The author would like to gratefully thank Thane Chambers for her assistance in data analysis and Randy Reichardt, Research
Services Librarian (Engineering) at the University of Alberta, who provided me with the transaction log data for the
NANOnetBASE e-book library.
1.0 Introduction ficiently identify information resources and to carry
out inclusive and effective searches in a diverse and
Nanoscale science and technology have seen rapid heterogeneous range of digital libraries, web-based da-
growth and expansion in new areas in recent years. tabases, and search engines. It is essential to understand
Nanoscience and technology are characterized by nano the problems and issues of interdisciplinary searching
researchers in an increasingly interdisciplinary domain, in a strategically and commercially significant area such
drawing upon such disciplines as chemistry, physics, as nanotechnology to support and enhance researchers’
materials science, and computer, electrical, mechanical, scholarly productivity and innovativeness, and to in-
and biomedical engineering. A key challenge faced by form knowledge organization and representation prac-
nano researchers, as well as the information profes- tices in nanotechnology-related library and informa-
sionals providing related information services, is to ef- tion services. Kutner (2000), in examining challenges
https://doi.org/10.5771/0943-7444-2011-2-135
Generiert durch IP '46.4.80.155', am 23.11.2021, 19:11:56.
Das Erstellen und Weitergeben von Kopien dieses PDFs ist nicht zulässig.136 Knowl. Org. 38(2011)No.2
A. Shiri. Revealing Interdisciplinarity in Nanoscience and Technology Queries: A Transaction Log Analysis Approach
and issues in searching in interdisciplinary areas, argues 2.2 Citation analysis and domain mapping
that, although electronic bibliographic databases that
provide sophisticated searching capabilities and multi- Most information science studies published relating
ple access points to the scholarly literature have been a to nanoscience and technology focus on citation ana-
boon to the interdisciplinary researcher, problems con- lysis, scientometric, and bibliometric studies of jour-
tinue to exist in terms of lack of consistent interfaces nals and publications in the field of nanoscience and
and consistent controlled vocabularies across databases. technology (Meyer and Persson 1998; Meyer 2000,
While there exist many studies of general information 2001; Eto 2003; Morillo et al 2003; Schummer 2004a;
seeking behaviour of interdisciplinary areas of humani- Calero et al 2006; Leydesdorff 2007), mapping and
ties, the sciences, women’s studies, and social sciences, visualization of the area of nanoscience and technol-
very few studies have examined the nature of queries ogy (Boyack et al 2002; Zitt and Bassecoulard 2006;
and query formulation and reformulation strategies of Bassecoulard 2007; Milojević 2007), and identifying
interdisciplinary searchers in nanoscience and technol- core journals in the area of nanotechnology (Leydes-
ogy. dorff 2007). A special issue of Scientometrics (Volume
70, Number 3, March 2007) focuses on various as-
2.0 Related work pects of nanoscience and technology-related trends
and issues. Suggestions for the development of an on-
This literature review focuses on the following areas: tology for the area of nanoscience and technology
the interdisciplinary nature of the sciences; nanos- have also been made by LIS researchers such as Ta-
cience and technology citation analysis and domain naka (2005). Citation analysis studies have also been
mapping; information search behaviour of interdisci- conducted to show the interdisciplinarity of other
plinary researchers; information search behaviour of scientific areas such as medical and behavioural sci-
scientists and engineers using electronic information ences (McCain 1989), environmental sciences (Steele
resources; and representative transaction log analysis and Stier 2000), biotechnology (McCain 1995), and
studies. chemical engineering (Peters et al 1995).
2.1 Interdisciplinarity 2.3 Information search behaviour of interdisciplinary
researchers
A number of studies have investigated the notion of
interdisciplinarity in the context of science and tech- Klein (1996) notes that meeting the interdisciplinary
nology. Qin et al (1997) investigated types and levels needs of today's library users begins with understand-
of collaboration in interdisciplinary research in the ing the activities that create them and their place in
sciences; they concluded that biology and medical sci- the knowledge system. Drawing on the studies car-
ences are highly interdisciplinary. Pierce (1999) identi- ried out by Mote (1962), as well as Packer and Soer-
fied three types of interdisciplinary information trans- gel (1979), Bates (1996) concluded that studying re-
fer: Borrowing, Collaboration, and Boundary cross- searcher information seeking in interdisciplinary
ing. Using bibliometric techniques, Morillo et al fields may tell us not only about the needs and prob-
(2003) measured interdisciplinarity in the sciences lems of people in those fields—something we very
based on the Institute for Scientific Information (ISI) much need to learn about—but also about what fac-
multi-assignation of journals in subject categories. tors, in general, contribute to ease and difficulty in
They found that a high degree of interdisciplinarity scholarly information seeking. She notes that inter-
was common in engineering, physics, and chemistry disciplinary researchers constitute a significant and
and concluded that new, emerging disciplines are distinctive class of scholars, much deserving of re-
highly interdisciplinary and show a predominance of search on their particular information needs and in-
“big interdisciplinarity.” Haythornthwaite (2006) formation-seeking behavior. Many other researchers
studied exchanges among three distributed interdisci- have investigated the challenges and issues of search-
plinary teams (one science and two social science ing in interdisciplinary areas (Kutner 2000; Palmer
teams) and found the highest indication of mutual ex- 1999; Palmer and Neumann 2002; Foster 2004; Weis-
change in reports of joint research in the science team. gerber 1993; White 1996).
Leydesdorff (2007) has introduced a new indicator
called “betweenness centrality” as a measure of inter-
disciplinarity of scientific journals.
https://doi.org/10.5771/0943-7444-2011-2-135
Generiert durch IP '46.4.80.155', am 23.11.2021, 19:11:56.
Das Erstellen und Weitergeben von Kopien dieses PDFs ist nicht zulässig.Knowl. Org. 38(2011)No.2 137
A. Shiri. Revealing Interdisciplinarity in Nanoscience and Technology Queries: A Transaction Log Analysis Approach
2.4 Information search behavior of scientists and 2.5 Transaction log analysis studies
engineers using electronic information resources
Transaction log analysis involves the following three
Several studies have focused on the ways in which new steps: TLA involves the following three major stages,
technologies such as email, electronic publishing, and which are as follows: a) collection: the process of col-
the Internet have affected the information seeking be- lecting the interaction data for a given period in a
haviour of petroleum engineers (Joseph 2001), as well transaction log, b) preparation: the process of cleaning
as scientists and engineers (Llull 1991); informal in- and preparing the transaction log data for analysis; and
formation sharing among scientists and engineers (Po- c) analysis: the process of analyzing the prepared data
land 1991); and information sources used by design, (Jansen 2006). Peters (1993) provides a review of
process, and manufacturing engineers (Kwasitsu 2003). transaction log analysis library and experimental IR
Pinelli (1991), for instance, reviews the literature of in- systems used in the 1980s and early 1990s. Transaction
formation seeking practices of scientists and engineers log analysis has been employed to study users of the
to discuss some of the differences in the information Internet in public libraries (Curry 2005); search en-
seeking patterns of scientists and engineers. Allen gines (Jansen and Pooch 2001; Cothey 2002; Ke et al
(1977) comments that engineers are more likely to use 2002; Wen et al 2002; Jansen et al 2005; Mat-Hassan
specific forms of literature such as handbooks, stan- and Levene 2005; Sanderson and Dumais 2007); com-
dards, specifications, and technical reports. King et al parative evaluation of web search engines (Jansen and
(1994) provide a comprehensive review of engineers’ Spink 2005); OPACs (Millsap and Ferl 1993; Blecic et
information needs, seeking processes, and use. Leckie al 1998; Lau and Goh 2006); library websites (Ghaph-
et al (1996) provide an analysis and interpretation of ery 2005); digital libraries (Jones et al 2000; Mahoui
empirical studies on the information habits and prac- and Cunnigham 2000; Marchionini 2000; Nicholas et
tices of three groups: engineers, health care profession- al 2006a; Zuccala et al 2007); electronic journals
als, and lawyers to propose a general model of informa- (Eason et al 2000; Davis and Solla 2003; Jamali et al
tion seeking. Other researchers have investigated the 2005; Nicholas et al 2005; Nicholas et al 2006b); elec-
information seeking patterns of engineers and research tronic books (Hughes and Buchanan 2001; Connaway
scientists in an industrial environment (Ellis and Hau- and Snyder 2005); and organizational interfaces
gan 1997); astronomers, chemists, mathematicians, and (Marchionini 2002). Peters (1993) provides a useful
physicists (Brown 1999); environmental scientists discussion of the history and development of transac-
(Murphy 2003); engineers (Hertzum and Pejtersen tion log analysis. Troll Covey (2002) provides a very
2000); and interdisciplinary scholars in the areas of the useful guide on the use of transaction log analysis for
arts, social sciences, and science and information media studying users of digital libraries, OPACs, and library
studies (Spanner 2001). Palmer (2005, 2006, and websites. Jansen (2006) presents transaction log analy-
Palmer et al 2007) discusses different levels of research sis as a methodological framework along with strengths
work and information use within an interdisciplinary and limitations of this method in research. In the pre-
brain research context. Chau et al (2006) developed a sent study, transaction logs of an electronic book digi-
knowledge portal for nanoscience and technology in- tal library have been analyzed to investigate the search
formation called NanoPort incorporating collection behaviour patterns of nanoscience and technology
building, meta-searching, keyword suggestion, and searchers, in particular the extent of interdisciplinarity
various content analysis techniques such as document in the queries formulated by nanoscience and technol-
summarization, document clustering, and topic map ogy searchers.
visualization. User evaluation of NanoPort showed
that NanoPort could provide results with higher preci- 3.0 Methodology
sion than benchmarking search engines such as Google
and NanoSpot (Qin et al 2004). The literature on the information search behaviour of
The aim of present research is to investigate the in- nanoscience and technology information searchers is
formation search behaviour of nanoscience and tech- scarce. The overarching objective of this study was to
nology researchers through analyzing transaction investigate the information search behaviour of nano
logs of a nano electronic book digital library called researchers interacting with a nanoscience and tech-
NANOnetBASE. In particular, it examines the extent nology electronic book digital library. A more specific
of interdisciplinarity in the queries formulated by na- objective was to examine whether the data can pro-
no researchers. vide a basis for investigating the interdisciplinary na-
https://doi.org/10.5771/0943-7444-2011-2-135
Generiert durch IP '46.4.80.155', am 23.11.2021, 19:11:56.
Das Erstellen und Weitergeben von Kopien dieses PDFs ist nicht zulässig.138 Knowl. Org. 38(2011)No.2
A. Shiri. Revealing Interdisciplinarity in Nanoscience and Technology Queries: A Transaction Log Analysis Approach
ture of the queries formulated by nano researchers. found to be exact matches for searchers’ query terms
As a result, the nature of this research is exploratory were traced to their top terms using their broader
rather than hypothesis-driven. terms. All of the top terms were then compared with
the INSPEC Classification codes. The classification
3.1 Research questions codes were used due to their discipline-based categori-
zation. A detailed description of the dataset and data
The research questions addressed in this study are analysis methods is provided in the following sections.
listed below. The first three focus on search charac-
teristics and strategies of nano researchers while the 3.2 The NANOnetBASE e-book digital library
remaining two are concerned with interdisciplinarity
in the searchers’ queries. NANOnetBASE is an e-book database for nano-
technology and nanoscience researchers consisting of
– What are the search behaviour characteristics of nanoscience and technology references in an online
nano searchers, including session length, query collection. This full-text database contains 45 titles
characteristics, and the use of query operators? that can be accessed in a variety of ways. Researchers
– What query formulation and reformulation strate- can browse titles from subject lists (e.g., Biomedical
gies did users employ? Engineering, Computer Engineering, Electrical Engi-
– What search strategies were adopted by the users neering, Electrical Engineering Communications,
(e.g., basic vs. advanced search features, failed and Electronics, General, Industrial Engineering & Manu-
erroneous queries, and spelling issues)? facturing, Laser & Optical Engineering, Mechanical
– Can disciplinary and interdisciplinary queries be Engineering, and Nanoscience/Nanotechnology) or
identified? What subject areas are covered by the search within all titles using keywords.
queries formulated by nano researchers?
– Are thesauri capable of providing a basis for identi- 3.3 Transaction log dataset
fying interdisciplinary queries in the area of
nanoscience and technology? The data used in this study consisted of transaction
logs from the NANOnetBASE digital library be-
Table 1 shows the variables used in this study along tween July 2004 and October 2006. The data was ac-
with their descriptions and the coding method. quired from a large Canadian university with strong
The data analysis was based on Jansen’s (2006) nanoscience and technology research profile where
transaction data analysis method, which includes term more than 140 nano researchers are currently active.
level, query level, and session level analyses. The IN- They include faculty members as well as graduate
SPEC thesaurus and the INPSEC Classification students and post-doc researchers. It should be noted
Codes were used to map the disciplines represented that the actual users were not involved in this study.
by the query terms. The INPSEC terms that were Rather, their interactions with the digital library, cap
Variable Definition Coding
Session Session length and the type of interactions users had Manual coding: identifying and recording
with the system (e.g., time spent on formulating and sessions using IP addresses
submitting a query and time spent on viewing the re-
sults). A transaction log may contain many sessions
in which users go to digital library for purposes other
than searching.
Query terms Number of terms per query Manual count
Search strategies Construction of queries using Identified and counted queries with op-
Boolean operators; queries with acronyms erators and acronyms
Query reformulation Any change to the initial query terms Manual comparison of initial vs. reformu-
lated terms
Subject areas of queries Discipline-based analysis of queries Comparison of query terms with two
thesauri, INSPEC and Compendex
Table 1. Definition of variables studies
https://doi.org/10.5771/0943-7444-2011-2-135
Generiert durch IP '46.4.80.155', am 23.11.2021, 19:11:56.
Das Erstellen und Weitergeben von Kopien dieses PDFs ist nicht zulässig.Knowl. Org. 38(2011)No.2 139
A. Shiri. Revealing Interdisciplinarity in Nanoscience and Technology Queries: A Transaction Log Analysis Approach
Table 2. Snippet from a sample transaction log
tured in the NANOnetBASE server, were used as re- Session level analysis: includes session length and
search data in this study. In total, 1921 transactions the type of interactions users had with the system
were analyzed. Table 2 provides a sample of transac- (e.g., time spent on formulating and submitting a
tion logs that were analyzed in the study. The table query and time spent on viewing the results). This
has the following four columns: session definition is similar to the definition of a
unique visitor that is used by commercial search
– Date and time of activity. engines and organizations to measure Web site traf-
– User identifier (in the form of IP address). fic (Jansen 2006).
– Activity detail (query or book viewed).
– Activity type: book viewed, advanced search, quick 4.0 Results
search, search results viewed, or managed account
information. The results and findings presented here are based on
the order of the research questions.
3.4 Data analysis In order to provide a brief overview of the general
results, Table 3 shows the number of sessions, que-
Jansen (2006) suggests three common levels of analy- ries, reformulated queries, and queries using advanced
sis for examining transaction logs (i.e., term, query, search features.
and session), so we based our data analysis on these.
Before the transactions were analyzed, the transac- Variable Number
tion logs were converted into a spreadsheet to facili- Sessions 552
tate data manipulation. Queries 393
Queries using acronyms 40
Term level analysis: the number, nature, and fre- Reformulated queries 155
quency of user terms. Queries using advanced search features 22
Query level analysis: includes query characteristics Table 3. Summary of session and query-related data
as follows:
– Query formulation and reformulation (and the 4.1 Session length
reasons for query reformulations),
– Query construction strategies (use of Boolean Jansen et al (2005) noted that a session is the entire se-
operators, quick vs. advanced search strategies ries of queries submitted by a user during one interac-
etc.), tion with the Web search engine. This definition was
– Query classification (based on disciplines). taken into consideration while analyzing the dataset in
this study. Session length information for the dataset
https://doi.org/10.5771/0943-7444-2011-2-135
Generiert durch IP '46.4.80.155', am 23.11.2021, 19:11:56.
Das Erstellen und Weitergeben von Kopien dieses PDFs ist nicht zulässig.140 Knowl. Org. 38(2011)No.2
A. Shiri. Revealing Interdisciplinarity in Nanoscience and Technology Queries: A Transaction Log Analysis Approach
was only available for the years 2005 and 2006. Table 4 Table 5 shows that approximately 60% of queries
shows the average time spent on each search step by an consisted of two or more terms. This finding is par-
individual user; the overall search session length ranged ticularly interesting when compared with previous
from 30 seconds to more than 6 hours. This result studies that found users tend to use very short que-
should be contextualized by the fact that this is in a ries. This can be explained by the very specific cover-
digital library that is an electronic book digital library age of this digital library and, based on closer exami-
with more than 44 titles. Some users spent more time nation of queries, that nano researchers tended to
on consulting the content than querying. The time start with very specific, but long, muti-term queries
spent on each query averaged 2 minutes and fifty-one and then reformulated their queries if they were con-
seconds, but users spent more time looking at particu- sidered to be too restrictive in terms of retrieval. A
lar books and also viewing the results of their searches. hypothesis could be formulated to test whether in-
terdisciplinary searchers tend to formulate longer
Activity type Time spent queries than disciplinary searchers.
Average Session 1:51:00
Average Query 0:02:51 4.3 Search strategies
Average Book Viewed 0:08:04
Average Search Results Viewed 0:05:57 One advantage to using transaction logs is the ability
to see the entire search process. After performing a
Table 4. Search session length query, the user receives search results from the NA-
NOnetBASE with the following information: number
It is important to note the difference between the aver- of times the search string appears in the document and
age session of 1 hour 51 minutes and the individual ac- a list of documents (usually chapter titles) indicating
tivities that occurred during the sessions; that is, most book title, authors, and file size. Unfortunately, there
users performed several activities during their interac- is little context for the user to determine whether re-
tion with the system. Users who chose to limit their trieved results are actually relevant without opening
session to one single action normally viewed a single the document itself. No abstract or brief annotation is
book. Of the 187 sessions, 161 of these involved view- provided for any of the documents found. Figure 1
ing a book. This is not surprising since it reflects the shows the NANOnetBASE search results screen.
nature of the digital library resources itself. The search process begins with the query. During
the time period of this study, most users started with a
4.2 Query terms quick search easily carried out from the homepage. It
is interesting to note that, in 2005, the number of Ad-
The average number of terms used per query was 2.11; vanced Searches and Quick Searches performed were
the maximum terms used was 8. Most users were look- almost equal (i.e., 94 Advanced Searches and 90 Quick
ing for subjects, and their queries reflected looking for Searches), but Quick Searches were far more popular
subjects in a direct away. For example, if someone were in 2004 (almost twice as many) and 2006 (almost four
looking for information about “textiles” and nano- times as many). It is also interesting that most users
technology, the query would be simply “textile.” Que- did not move from Quick Search to Advanced Search,
ries with more terms tended to be searches for book ti- but preferred to reformulate their search within the
tles, which are usually longer than one or two words. search type they started with. This may suggest that
the advanced search functionality was not viewed as
Number of terms Number of Percentage of
desirable, relevant, or intuitive to the users.
used in queries Queries Queries
NANOnetBASE has many search functions in-
1 156 40%
cluding single character, stemming, fuzzy search,
2 137 35%
phonic search, synonym search, numeric range, and
3 53 14%
4 19 5%
variable term weighting. These search functions,
5 9 2%
however, are not well-used by users of this database.
6 13 3% Only 22 queries out of 393 used any advanced search
7 3 0.76% functions and the majority of these just used Boo-
8 3 0.76% lean. Table 6 provides a summary of various search
Total 393 strategies adopted by searchers along with the fre-
quency of those strategies in the dataset.
Table 5. Number of terms in queries
https://doi.org/10.5771/0943-7444-2011-2-135
Generiert durch IP '46.4.80.155', am 23.11.2021, 19:11:56.
Das Erstellen und Weitergeben von Kopien dieses PDFs ist nicht zulässig.Knowl. Org. 38(2011)No.2 141
A. Shiri. Revealing Interdisciplinarity in Nanoscience and Technology Queries: A Transaction Log Analysis Approach
Figure 1. The NANOnetBASE search results screen
Search features No. Quick Search where those functions can be used but
Queries using advanced search functions 22 are not promoted in any obvious way, the simple and
Queries using more than one advanced search 4 easy access to free text searching, or users’ frustration
function when advanced search features retrieve no results. In-
Queries using 2 advanced search function 4 terestingly, many consecutive searches revealed users
Queries using Boolean operators 20 searching for single and plural versions of the same
Queries using AND 19 word, so the use of wildcard characters or truncation
Queries using OR 1 has potential time-saving benefits.
Queries using phrase searching 4
Queries using parentheses 1 4.4 Search for acronyms
Queries using truncation 1
About 10% of the queries made use of some form of
Table 6. Search strategies acronym. This suggests that, as in many scientific
fields, acronyms are a commonly-used and intrinsic
Although the ‘?’ character can be used to match any language form within the literature of nanoscience
single character, no queries in this dataset used it. and technology. Table 7 shows the extent of the use
Furthermore, there were no uses of stemming, fuzzy of acronyms in the searches.
search, phonic search, synonym search, numeric range, The search function of NANOnetBASE does not
or variable term weighting. Among the variety of handle acronyms intelligently, so users must know and
reasons users chose not to use these functions could use their full forms as well. If an acronym is entered,
be a lack of understanding how they work, inaccessi- only a small number of results are retrieved because the
bility of descriptions or instructions, reliance on system does not offer simultaneous full-form and ac-
https://doi.org/10.5771/0943-7444-2011-2-135
Generiert durch IP '46.4.80.155', am 23.11.2021, 19:11:56.
Das Erstellen und Weitergeben von Kopien dieses PDFs ist nicht zulässig.142 Knowl. Org. 38(2011)No.2
A. Shiri. Revealing Interdisciplinarity in Nanoscience and Technology Queries: A Transaction Log Analysis Approach
Queries containing acronyms % of 393 Reasons for query reformulation No of
queries instances
Queries using acronyms 9.67 Query changed (spelling errors) 9
Queries using 1 acronym by itself 4.33 Query expansion (additional terms added) 32
Queries using acronyms and full form 4.58 Major shifts 38
Queries using 2 acronyms 0.76 Query narrowed (narrower term used) 21
Query broadened ( broader term used) 26
Table 7. Queries using acronyms Query restructured (query terms moved 6
around)
ronym searches to make the document selection pro- Synonymous terms used 23
cess easier. To ensure that all relevant documents are re- Total 155
trieved, users must combine the acronym with its full-
Table 9. Type and number of reformulation strategies
form equivalent using the Boolean operator OR.
Figure 2 provides a graphical representation of the
4.5 Query reformulation
reasons for query reformulation and the number of
instances for each reformulation reason.
A key question of this research was to examine how
One interesting observation regarding the nature of
users reformulated their queries and the reasons for
the queries submitted by nano searchers was that, of
doing so. Query reformulation can be interpreted in a
132 terms starting with the prefix “nano,” a few were
variety of ways. For instance, adding terms to the ini-
very general (e.g., nanotechnogy or nanoscience) but
tial search terms is considered to be one of the ways in
most represented very specific query terms; for exam-
which a query can be expanded whereas dropping any
ple, “nano-onion,” “nonoparticle,” “nanofabrication,”
terms from the initial query terms would be consid-
“nanotubes,” and “nanocones.” Due to a lack of con-
ered query refinement or modification. Through the
sistency in the ways these terms appear in different
analysis of users’ terms and the various ways in which
documents (e.g., sometimes hyphenated, sometimes
they reformulated their initial queries, we identified a
not), some searchers were not successful at retrieving
number of query reformulation strategies discussed in
any documents using these terms if they happened to
this section. Table 8 shows basic statistics about the
try the wrong form. Figure 2 refers to a shift to a to-
number of total queries and the reformulated queries
tally different topic. Table 10 below shows examples
in the analyzed dataset. One important observation is
of query reformulations.
that approximately 40% of queries were reformulated.
4.6 Interdisciplinarity in queries
Query details No. of
queries
Another key question of this research was to examine
Number of sessions 552
whether disciplinary or interdisciplinary queries can
Total number of queries 393
be identified in the search behaviour of nano re-
Total number of reformulated queries 155
searchers. This is a complex and multidimensional
Table 8. Number of Initial and reformulated queries question because interdisciplinarity can be inter-
preted from a variety of perspectives and using vari-
Table 9 shows types of reformulation strategies and ous methods (Palmer 1999; Morillo et al 2003). Pre-
the number of reformulations carried out by users. vious research on the interdisciplinary nature of
One interesting pattern was the use of phrases in nanoscience and technology has largely focused on
searching, and many reformulated queries were, in citation analysis and domain mapping (Meyer and
fact, the result of phrase searching. The “major shift” Persson 1998; Meyer 2010; Zitt 2006). Here the in-
category refers to a reformulation strategy where the tention is to explore whether the data allows us to
user tries a new query related to the initial query but address this issue from a search behaviour perspec-
not by any hierarchical or semantic relationships. tive. Some specific questions in this regard might be:
“Expansion” and “major shifts” queries constitute
37% of the 155 total reformulated queries. Also, 37% – Are there any queries that represent more than one
of reformulated queries were hierarchically or syn- discipline?
onymously related to the initial queries.
https://doi.org/10.5771/0943-7444-2011-2-135
Generiert durch IP '46.4.80.155', am 23.11.2021, 19:11:56.
Das Erstellen und Weitergeben von Kopien dieses PDFs ist nicht zulässig.Knowl. Org. 38(2011)No.2 143
A. Shiri. Revealing Interdisciplinarity in Nanoscience and Technology Queries: A Transaction Log Analysis Approach
Figure 2. Query reformulation strategies
Initial query Spelling Query ex- Query modi- Query modi- Query synonyms Major shift
error pansion fication fication restructured
(narrower) (broader)
Protiein Protein
Carbon
nanotube
nantubes
DNA
DNA
scaffolding
Microchannel
Heat sink
heat sink
DNA wire Wiring DNA
Nanoparticle Nanoparticle
toxicology toxicity
Transistor
Intel
size
Table 10. Examples of query reformulation strategies
– Are there any specific characteristics associated ity with their actual search behaviour, their research
with interdisciplinary queries? topics, and the results of the query analysis. How-
– Are interdisciplinary queries more complex and ever, in order to analyze the data to support our
sophisticated than disciplinary queries? analysis of the extent of interdisciplinarity in the que-
ries, it was decided to conduct some analyses of user
Because we did not have access to the users them- terms and compare them with thesauri in two com-
selves or their research and educational backgrounds, mercial databases, namely INSPEC and Compendex,
it was not possible to base any evaluation on their and then compare them to the INSPEC classification
perception of interdisciplinarity. It would be instruc- codes that represent disciplines.
tive to compare users’ perception of interdisciplinar-
https://doi.org/10.5771/0943-7444-2011-2-135
Generiert durch IP '46.4.80.155', am 23.11.2021, 19:11:56.
Das Erstellen und Weitergeben von Kopien dieses PDFs ist nicht zulässig.144 Knowl. Org. 38(2011)No.2
A. Shiri. Revealing Interdisciplinarity in Nanoscience and Technology Queries: A Transaction Log Analysis Approach
4.7 Mapping user terms to INSPEC Terms representing disciplines No. of top %
and Compendex thesauri terms
Top terms representing one 24 37.5
The first type of analysis was to map users’ terms discipline
with the terms in the two thesauri attached to the Top terms representing two 22 34.3
INSPEC and Compendex databases. This has been disciplines
reported in another paper (Shiri & Chambers 2008) Top terms representing 18 28.2
three or more disciplines
showing that approximately 50% of terms entered by
users matched those of the Inspec thesaurus and that Total 64
around 62% of users’ search terms were matched Table 11. Percentage of top terms representing disciplines
those of the Compendex thesaurus.
The top terms mapped to the INSPEC Classification
4.8 Mapping top terms to the Codes represent the following disciplines: A - Physics,
INSPEC Classification codes B - Electrical Engineering & Electronics, C - Com-
puters & Control, and E - Manufacturing & Produc-
The second type of analysis investigated whether any tion Engineering.
element of interdisciplinarity could be observed in One important observation about queries repre-
users’ queries. To analyze this, the INPSEC thesaurus senting two or three disciplines is that most had the
terms that were exact matches for searchers’ query INSPEC codes A and B for Physics and Electrical
terms were traced to their top terms using broader Engineering & Electronics, respectively. As Table 11
term indications in the thesaurus. All the top terms shows, approximately 62% of the top terms represent
were then compared with the INSPEC Classification two or more disciplines. Our findings also suggest
codes, which represent discipline-based categoriza- that some query terms represented a greater degree of
tions. Appendix A shows the classification structure interdisciplinarity. In other words, there may be que-
and its treatment of various domains. ry terms that occur in more than one discipline or are
Each INSPEC top term was searched in the subject discussed, researched, and searched from a variety of
field of the INSPEC database to retrieve records for perspectives. A closer examination of the top terms
each term with their subject fields containing the top and some of the associated classification codes pro-
term. Following the search, details of up to 10 docu- vides another interesting aspect of query interdisci-
ments were examined to ensure that the top term was, plinarity. Among the terms with a higher degree of
in fact, among the first three terms assigned to each interdisciplinarity, there were such terms as “coat-
document in the database. Then, classification codes ings,” “fluidics,” “control equipment,” “interface
for those records were examined and recorded to ana- phenomena,” “surface treatment,” and “vacuum tech-
lyze the disciplines and sub-disciplines they repre- niques.”
sented. The rationale for this method is that it takes Another interesting observation relates to code A8
into account the document collection as context for in appendix “A” or “Cross-disciplinary physics & re-
top terms; in other words, top terms associated with lated areas of science & technology.” There are a
classification codes and the document collection number of top terms that have been assigned this
formed the basis for the analysis of interdisciplinary classification code thus demonstrating another di-
queries. mension of interdisciplinarity. That is, terms such as
Top terms and classification codes were entered “electrochemical devices,” “transistors,” “disperse
into an MS Excel spreadsheet and analyzed (Appen- systems,” and “electric field effects” appear in docu-
dix A lists the INSPEC top terms and the INSPEC ments that show more interdisciplinarity than re-
Classification Codes utilized). In total, 64 unique top flected in the codes assigned to them. Another inter-
terms were used for searching NANOnetBASE and esting observation is that 52 out of the 64 top terms
were compared with the INSPEC Classification had the classification code “A” (i.e., physics) which
Codes. Table 11 summarizes the analysis of top terms suggests physics plays a key role in nanoscience and
with regards to their disciplinary or interdisciplinary technology and that a majority of nanoscience and
origins. technology search queries are, in fact, associated with
this discipline.
https://doi.org/10.5771/0943-7444-2011-2-135
Generiert durch IP '46.4.80.155', am 23.11.2021, 19:11:56.
Das Erstellen und Weitergeben von Kopien dieses PDFs ist nicht zulässig.Knowl. Org. 38(2011)No.2 145
A. Shiri. Revealing Interdisciplinarity in Nanoscience and Technology Queries: A Transaction Log Analysis Approach
5.0 Discussion gine users. Wolfram (2008) found that query length
varies in different web-based information retrieval
Interdisciplinary searching in the area of nanotechnol- environments; users tend to formulate shorter que-
ogy is an interesting topic as it sheds light on the ways ries in bibliographic databanks or specialized search
in which knowledge organization systems, digital li- services (1.81 and 1.78 terms per query, respectively)
braries, portals, and other web services can be devel- and longer queries when using OPACs or search en-
oped. The following discussion brings together and gines (3.66 and 2.62 terms per query respectively). In
discusses our results with findings reported previously. the present study, the average number of terms per
Session length indicates the level of users’ engagement query was 2.11. An interesting observation in this
and also their interaction with the system. Jones et al study was that approximately 24% of queries had
(2000) suggest that over half (54.34%) of user sessions three or more terms. This can be explained by users
they studied had a duration of five or fewer minutes, carrying out searches in an electronic book collection
while two thirds (66.43%) had a duration of ten min- specific to the area of nanoscience and technology,
utes or less. Marchionini (2002) reported that many re- where users could formulate very specific but some-
searchers use 30 minutes as a sensible idle time interval times long and hyphenated queries. Some queries
for a session delimiter..Spink and Jansen (2004) noted formulated by nano researchers are multifaceted and
that the mean session time associated with an identifier are more sophisticated compared to queries submit-
was more than two hours for some search engine data- ted to general search engines.
sets but most sessions lasted less than 15 minutes. In total, 22 queries made use of Boolean operators,
Wolfram (2008) found that users were more likely to but it must be noted that the interface does not ex-
engage in extensive searching using the OPAC and plicitly provide search options using Boolean opera-
specialized search system as compared to search en- tors, thus partly explaining the limited use. Gerhard,
gines and bibliographic databases. In the present study, Su, and Rubens (1998) note that despite keyword,
session length was found to be in line with those from Boolean, and other sophisticated search techniques
Wolfram (2008) and Spink & Jansen (2004); the aver- making it easier to access interdisciplinary and newly
age session lasted for an hour and fifty-one minutes. emerging bodies of literature, poorly constructed
This finding should be viewed in light of users interact- keyword searching still results in the retrieval of
ing with an electronic book digital library so a signifi- much irrelevant material. In this study, query refor-
cant amount of time was spent viewing individual elec- mulations and consecutive searches were related to
tronic books. It can be argued that, in the context of the poor performance of keyword searching. A close
electronic book libraries, users generally spend more examination of users’ queries showed that approxi-
time compared to OPAC searches, search engine use, mately 10% of their queries included some form of
and bibliographic databases consultation. acronym. Because nanotechnology queries may rep-
One search pattern related to search modes found in resent multifaceted and interdisciplinary search
this study was that approximately 40% of searches that terms, the use of acronyms should be considered one
were carried out in 2004 to 2006 used the advanced search strategies that digital libraries in this field must
search mode. This may be partly explained by the support. In the present study, users showed variabil-
NANOnetBASE interface not being flexible enough ity in their use of search terms, including American
to support users’ retrieval of partially relevant results in and British spellings. The matching algorithm and the
their initial searches. Furthermore, approximately 85% interface should be designed to support users by pro-
of searches were subject searches reflecting the impor- viding cross references for variant spelling forms and
tance of providing subject-based search support for us- acronyms, particularly the disambiguation of homo-
ers of e-book digital libraries. All of the query terms graphs such as ‘PVD’ to distinguish between Physical
were checked against the electronic book titles that Vapour Deposition and Peripheral Vascular Disease.
were available in the NANOnetBASE digital library. One key search strategy that can be used to infer
This observation is particularly important as users users’ search patterns and behaviours is query refor-
could easily browse the list of titles available in the li- mulation. The analyzed data showed that approxi-
brary, hence the focus on subject-based searches. mately 40% (155 of 393) of the queries were subject
Number of terms per query and query length rep- to reformulations. Some reasons for query reformula-
resent search behaviour characteristics of interest to tions were spelling errors, hyphenated and combined
researchers. Jansen et al (2005) reported 2.35 average words, narrowing the search, broadening the search,
number of terms per query for Altavista search en- restructuring the query, and use of phrase searching
https://doi.org/10.5771/0943-7444-2011-2-135
Generiert durch IP '46.4.80.155', am 23.11.2021, 19:11:56.
Das Erstellen und Weitergeben von Kopien dieses PDFs ist nicht zulässig.146 Knowl. Org. 38(2011)No.2
A. Shiri. Revealing Interdisciplinarity in Nanoscience and Technology Queries: A Transaction Log Analysis Approach
and synonyms. The reformulation reasons noted ear- nology. Examples of these patterns are the use of
lier indicate that nanotechnology information search- multiword and compound query terms, hyphenated
ers require terminological support to carry out suc- terms, spelling variations, query reformulation, and
cessful reformulations and to explore the conceptual the use of acronyms. One hypothesis that can be
structure of the collection. The use of multiword formulated based on this study is that interdiscipli-
search terms and the order of the terms pose a chal- nary searchers tend to carry out more query reformu-
lenge to nanotechnology searchers in terms of formu- lation and modification compared with disciplinary
lating queries. It is of particular importance that the searchers. This hypothesis should be tested by a user-
interface and the search system be flexible enough to centred study, in which user contextual factors, such
guide the user throughout the search process. Inter- as task, intent, education level, and computer and in-
active term suggestion facilities based on the docu- formation literacy could be captured. This study also
ment collection can alleviate this problem. demonstrated that knowledge structures such as
Braun et al (1997) analyzed the early growth of thesauri attached to multidisciplinary databases and
nanoscience research during 1986-1995 and counted the INSPEC Classification Codes can contribute to
the occurrence of nano-prefix terms in the title of the identification of interdisciplinary queries. In par-
journal papers; they found 4,152 such papers during ticular, the INSPEC and Compendex thesauri can
1985-1995, which indicated a fast growth rate. provide interactive term suggestion facilities to sup-
Schummer (2004b) measured the growth of “nano- port users in exploratory browsing and searching of
title-papers” in terms of annual growth rate and dou- nanotechnology information collections by providing
bling times in different bibliographic databases and a structure for consecutive or reformulated searches.
various disciplines. In the present study, approxi- In addition, the knowledge structures in thesauri con-
mately 132 of the search terms used had the prefix textualize terms and how they are used in different
“nano,” thus stressing the importance of incorporat- disciplines and subject areas. Other findings suggest
ing ways to handle these terms by the search system. that the use of acronyms and full form of terms
McCain (1989) studied descriptor versus citation- should be accommodated to allow users to disam-
based retrieval in the medical behavioural sciences lit- biguate their search terms; the user interface/system
erature and found the overlap between the results re- should provide opportunities to use either hyphen-
trieved using descriptors and the results retrieved ated or non-hyphenated words without penalizing
through citation-based searches was low. The use of the user by zero hits. The user terms that will be
INSPEC thesaurus terms along with INSPEC Classi- matched with the terms in the thesaurus, either ex-
fication Codes provided an analytical framework for actly or partially, could be shown to the user in addi-
this type of categorization in this study. Sixty-two tion to the discipline or disciplines in which these
percent of the 64 unique top terms resulting from terms are used. This way, the thesauri could be used
mapping users’ query terms to the INSPEC thesau- as an interdisciplinary map of several different do-
rus represented two or more disciplines and these mains.
queries were associated with the disciplines of Physics The NANOnetBASE user interface does not pro-
and Electrical Engineering & Electronics. One hy- vide query reformulation or modification facilities:
pothesis based on this finding is that queries related interdisciplinary digital library services should offer
to Physics and Electrical Engineering & Electronics interactive query reformulation support such as term
represent a considerable number of nanoscience and suggestions or a document collection map as well as
technology search queries. This finding concerning ranking algorithms that account for the importance
the interdisciplinarity of queries is particularly impor- of query terms in various disciplinary or interdiscipli-
tant, indicating that some queries are, in fact, more nary documents. This area also calls for revisiting in-
interdisciplinary than others. formation organization and representation strategies
based on the notion of disciplinarity. For instance,
6.0 Conclusion subject terms can be categorized, contextualized, and
presented based on their relevance and importance in
This study was based on an analysis of transaction different disciplines. From a searcher’s standpoint,
logs from an electronic book digital library in the area search terms can be used to retrieve documents in a
of nanoscience and technology. The results show that collection based on disciplinary or interdisciplinary
there are certain patterns associated with searching in ranking or focus. If we can prove that some queries
an emerging and interdisciplinary area of nanotech- are more interdisciplinary than others, then it should
https://doi.org/10.5771/0943-7444-2011-2-135
Generiert durch IP '46.4.80.155', am 23.11.2021, 19:11:56.
Das Erstellen und Weitergeben von Kopien dieses PDFs ist nicht zulässig.Knowl. Org. 38(2011)No.2 147
A. Shiri. Revealing Interdisciplinarity in Nanoscience and Technology Queries: A Transaction Log Analysis Approach
be possible to design algorithms that match terms to References
documents based on disciplinary or interdisciplinary
focus. The interface could then categorize and display Allen, Thomas. J. 1977. Managing the flow of technol-
the retrieved results based on the degree of interdis- ogy: technology transfer and the dissemination of
ciplinarity or the dominant disciplines that have a technological information within the R&D organiza-
higher degree of association with the terms searched. tion. Cambridge, MA: MIT Press.
Furthermore, the user interface could be designed Bassecoulard, Elise 2007. Mapping nanosciences by
such that terms and their relationships as well as their citation flows: a preliminary analysis. Scientomet-
associated disciplines be shown to the user for further rics 70: 859-80.
exploration and examination. This type of design calls Bates, Marcia. J. 1996. Learning about the informa-
for a more enriched use of the knowledge structures tion seeking of interdisciplinary scholars and stu-
in thesauri within search user interfaces, by drawing dents. Library trends 45n2: 155-64.
upon both thesaurus term relationships and classifica- Blecic, Deborah., Bangalore, Nirmala S., Dorsch, Jo-
tion categories. For instance, thesaurus terms and sephine L., Henderson, Cynthia L., Koenig,
their broad level facets could be shown to the user Melissa H., and Weller, Ann C. 1998. Using trans-
along with the documents retrieved to allow users to action log analysis to improve OPAC retrieval re-
form disciplinary and interdisciplinary views of terms sults. College and research libraries 59: 39-50.
and documents. Faceted browsing and exploratory Boyack, Kevin W., Wylie, Brian N., and Davidson,
search interfaces that make use of taxonomies, the- George S. 2002. Domain visualization using VxIn-
saurus terms or high level categories and facets, are sight for science and technology management.
becoming particularly popular. For instance, Fla- Journal of the American Society for Information Sci-
menco Search Interface Project at the University of ence and Technology 53: 764-74.
Berkeley School of Information (Hearst 2006) pro- Braun, Tibor., Schubert, A., and Zsindely, Sandor.
vides interesting and innovative ways of incorporat- 1997. Nanoscience and nanotechnology on the
ing facets into search user interfaces to support users’ balance. Scientometrics 38: 321-25.
navigation and exploration of information. Brown, Ceceila M. 1999. Information-seeking behav-
This study was limited to transaction log analysis iour of scientists in the electronic information age:
and did not involve real users. Further research should astronomers, chemists, mathematicians, and physi-
examine the interdisciplinary search behaviour of real cists. Journal of the American Society for Informa-
users and their perceptions and impressions of their tion Science 50: 929-43.
research topics, search terms, and vocabulary prob- Calero, Clara, Buter, Renald, Valoes, Ceclia. C., and
lems. Vocabulary problems are considered one of the Noyons, Ed. 2006. How to identify research
key issues in interdisciplinary searching and can be groups using publication analysis: an example in the
analyzed from two different perspectives. One per- field of nanotechnology. Scientometrics 66: 365-76.
spective would identify the difficulties that nano- Chau, Michael, Huang, Zan., Qin, Jialun, Zhou, Yilu,
science and technology searchers experience as they and Chen, Hsinchun. 2006. Building a scientific
search for a particular term, while the other would ex- knowledge web portal: The NanoPort experience.
amine the ways in which digital libraries and informa- Decision support systems 42: 1216-38.
tion retrieval systems in this area represent informa- Connaway, Lynn S., 2005, February. Transaction
tion. Further research should investigate the value and log analysis of electronic (e-book) usage. Against
the use of co-occurrence analysis of terms in interdis- the grain 85-89. Available http://www.oclc.org/
ciplinary areas such as nanotechnology to provide research/publications/archive/2005/connaway-
support for interdisciplinary searchers. The analytical snyder-atg.pdf
framework developed in this study to classify queries Cothey, Vivian. 2002. A longitudinal study of World
based on their subject disciplines can be used to study Wide Web users’ information-searching behavior.
other interdisciplinary domains; a particular strength Journal of American Society for Information Science
of this framework is the combined use of real search and Technology 53: 67-78.
terms entered by interdisciplinary searchers with the- Curry, Ann. 2005. Reaching beyond the reference
saurus terms and classification codes to provide a solid books in public libraries. International journal of
basis for the study of query interdisciplinarity. web based communities (IJWBC) 1: 450-63.
Davis, Philip. M., and Solla, Leah. 2003. An IP-Level
analysis of usage statistics for electronic journals in
https://doi.org/10.5771/0943-7444-2011-2-135
Generiert durch IP '46.4.80.155', am 23.11.2021, 19:11:56.
Das Erstellen und Weitergeben von Kopien dieses PDFs ist nicht zulässig.148 Knowl. Org. 38(2011)No.2
A. Shiri. Revealing Interdisciplinarity in Nanoscience and Technology Queries: A Transaction Log Analysis Approach
chemistry: making inferences about user behav- Jansen, Bernard J., and Pooch, Udo. 2001. Web user
iour. Journal of the American Society for Informa- studies: A review and framework for future work.
tion Science and Technology 54: 1062-8. Journal of the American Society of Information Sci-
Eason, Ken, Richardson, Sue, and Yu, Liangzhi. 2000. ence and Technology 52: 235−46.
Patterns of use of electronic journals. Journal of Jansen, Bernard J., and Spink, Amanda. 2005. How
documentation 56: 477-504. are we searching the World Wide Web? A compari-
EI thesaurus description 2008. Available http://www. son of nine search engine transaction logs. Infor-
ei.org/databases/other.html. mation processing & management 42: 248−63.
Ellis, David, and Haugan, Merete. 1997. Modelling Jansen, Bernard J., Spink, Amanda, and Pedersen, Jan.
the information seeking patterns of engineers and 2005. A temporal comparison of AltaVista Web
research scientists in an industrial environment. searching. Journal of the American Society for In-
Journal of documentation 53: 384-403. formation Science and Technology 56: 559-70.
Eto, Hajime. 2003. Interdisciplinary information in- Jones, Steve, Cunningham, Sally Jo, McNab, Rodger,
put and output of a nano-technology project. Sci- and Boddie, Stefan. 2000. A transaction log analy-
entometrics 58: 5-33. sis of a digital library. International journal on digi-
Foster, Allan. 2004 A nonlinear model of informa- tal libraries 3: 152-69.
tion-seeking behavior. Journal of the American So- Joseph, Lura E. 2001. Information-seeking and com-
ciety for Information Science and Technology 55: munication behavior of petroleum geologists. Sci-
228-37. ence & technology libraries 21n3/4: 46-62.
Gerhard, Kristin H., Su, Mila C., & Rubens, Charlot- Ke, Hao-Ren, Kwakkelaar, Rolf, Tai, Yu-Min, and
te C. 1998. An empirical examination of subject Chen, Li-Chun. 2002. Exploring behaviour of e-
headings for women's studies core materials. Col- journal users in science and technology: Transaction
lege and research libraries 59: 130-38. log analysis of Elsever's ScienceDirect OnSite in
Ghaphery, Jimmy. 2005. Too quick? Log analysis of Taiwan. Library & information science research 24:
quick links from an academic library website. 265-91.
OCLC systems & services 21n3: 148-55. King, Donald W., Caste, J., and Jones, H. 1994. Com-
Haythornthwaite, Caroline. 2006. Learning and munication by engineers: a literature review of engi-
knowledge networks in interdisciplinary collabo- neers’ information need, seeking processes, and use.
rations. Journal of the American Society for Infor- Washington, DC: Council on Library Resources.
mation Science and Technology 57: 1079-92. Klein, Julie T. 1996. Interdisciplinary needs: The cur-
Hearst, Marti. 2006. Design recommendations for hi- rent context. Library trends 45n2: 134–54.
erarchical faceted search interfaces. In Association Kutner, Laurie A. 2000. Library instruction in an in-
for Computing Machinery Special Interest Group on terdisciplinary environmental studies program:
Information Retrieval (SIGIR) Workshop on Fac- Challenges, opportunities, and reflections. Issues in
eted Search, pp. 26-30. science and technology librarianship 28. Available
Hertzum, Morten, and Pejtersen, Annelise M. 2000. http://www.library.ucsb.edu/istl/00-fall/ .
The information-seeking practices of engineers: Kwasitsu, Lishi. 2003. Information-seeking behaviour
searching for documents as well as for people. In- of design, process, and manufacturing engineers.
formation processing & management 36: 761-78. Library & information science research 25: 459-76.
Hughes, Carol A., and Buchanan, Nancy L. 2001. Lau, Eng P., and Goh, Dion H. 2006. In search of
Use of electronic monographs in the humanities query patterns: A case study of a university OPAC.
and social sciences. Library hi tech 19n4: 368-375. Information processing & management 42: 1316-29.
INSPEC User Guide (n.d.). Available http://www2. Leckie, Gloria J., Pettigrew, Karen E., and Sylvain,
theiet.org/publish/support/inspec/document/User Christian. 1996. Modelling the information-seeking
G/UserGuide.pdf of professionals: a general model derived from re-
Jamali, Hamid R., Nicholas, David, and Huntington, search on engineers, health care professionals, and
Paul. 2005. The use and users of scholarly e- lawyers. Library quarterly 66: 161-93.
journals: A review of log analysis studies. Aslib Leydesdorff, Loet. 2007. “Betweenness centrality” as
Proceedings 57: 554-71. an indicator of the interdisciplinarity of scientific
Jansen, Bernard J. 2006. Search log analysis: What it journals. Journal of the American Society for Infor-
is, what's been done, how to do it. Library & in- mation Science and Technology 58: 1303-19.
formation science research 28: 407-32.
https://doi.org/10.5771/0943-7444-2011-2-135
Generiert durch IP '46.4.80.155', am 23.11.2021, 19:11:56.
Das Erstellen und Weitergeben von Kopien dieses PDFs ist nicht zulässig.You can also read
