HCI in Context
What the Words Reveal about It
Samuel Bastos Buchdid and M. Cecília C. Baranauskas
Institute of Computing, State University of Campinas (UNICAMP),
Av. Albert Einstein N1251, Campinas – SP, CEP 13083-852, Brazil
Keywords: HCI Analysis, Conferences, Publications, Tag Clouds.
Abstract: Information and communication technologies are ever more present in our lives. Considerations about daily,
emotional and contextual issues have been necessary in the HCI agenda for a design that is suitable for
contemporary devices and uses, and also for an increasingly diversified audience. Based on an analysis of
titles found in the full program of two major conferences in the field of HCI (ACM CHI and IFIP Interact),
this work intends to identify the main focuses of the contributions over the last few years. Results of
analysis based on tag cloud representations and comparison of collected data are presented and discussed;
they reveal gaps, similarities and differences between what has been discussed in those forums and the
trends indicated by research references in the field.
1 INTRODUCTION
The impact of computing systems has changed since
the human-computer interaction (HCI) field
emerged, not only in the way we work, but also in
the way that we interact and collaborate with others
(Bannon, 2011). Beyond the workplace, technology
is increasingly being used in both public and private
spheres. With the appearance of new devices the
need for systems and connectivity is increasingly
present in our lives (Bødker, 2006).
The growth of techno-dependency is evident in
the way that computers are being incorporated into
objects (e.g., toys, appliances, cars, books, clothes
and furniture) and into everyday environments (e.g.,
airports, garages, malls, homes and offices), along
with the growth in hyperconnectivity that brings
people together as citizens and members of global
communities. Following this dynamism, the user
interface is now embedded in a context of ubiquity,
which establishes the end of interface stability
(Sellen et al., 2009). According to Sellen et. al.
(2009), these transformations redefine our
relationship with technology and change the way we
live by continually increasing digital presence in our
daily lives. This can be seen by the growing passion
of people for capturing more and more information
about other people and becoming increasingly
visible to others. This new behavior leaves digital
footprints for each individual, a process which
represents the end of the ephemeral, since
information about our lives and actions has been
extended. Another highlight is the growth of creative
engagement, which gained ground with the
proliferation of new digital tools (e.g., Web 2.0), and
which allows us to see the world in new ways.
Recognizing the changes, Bannon (2011) argues
that it is necessary to rethink the place of technology
in our values frame, how we live with and through
technology, and to give priority to human values,
activities, tools and environments. For Sellen et. al.
(2009), it is necessary to incorporate truly human
elements, and to conceptualize users as embodied
individuals who have desires and concerns, and who
belong to a social, economic and political ecology.
Furthermore, there must be flexibility, since
people’s engagement with technology and the nature
of their interactions with it change continuously.
Finally, to understand new forms of human-
computer interaction, it is necessary to think about
qualitative issues rather than quantifiable attributes
and capabilities in isolation.
In this sense, Bødker (2006) has drawn attention
to a new wave in the HCI field, and discusses that, to
follow these changes, new elements such as
emotion, aesthetics, motivation, culture, pragmatics
and life experience must gain relevance in the
human-computer relationship. Harrison et. al. (2007)
134
Bastos Buchdid S. and Cecília C. Baranauskas M..
HCI in Context - What the Words Reveal about It.
DOI: 10.5220/0004451301340142
In Proceedings of the 15th International Conference on Enterprise Information Systems (ICEIS-2013), pages 134-142
ISBN: 978-989-8565-61-7
Copyright
c
 2013 SCITEPRESS (Science and Technology Publications, Lda.)

also suggest the creation of a third paradigm in HCI.
Unlike the first and second, which are guided by
ergonomic issues and cognitive factors, respectively,
the third paradigm is guided by a phenomenological
matrix that adopts theories and several points of
view in a simultaneous way, in which the
constructions of meaning of the artefact and its
context are mutually defined and subject to multiple
interpretations.
Considering this context of understandings of the
HCI discipline, this study aimed at mapping out the
main issues that have been addressed in the HCI
field in recent years in order to identify the extent to
which these emerging issues are being recognized in
the main conferences in the field. Finally, this
analysis allows the identification of trends and ways
of thinking, which may engender future studies in
the field.
The analysis is based on the content displayed on
the websites of the two main HCI conferences in the
field. Information from titles of papers, technical
sessions, workshops, tutorials, posters and demos
were considered, extending a preliminary study
conducted to understand HCI in Brazil (Buchdid and
Baranauskas, 2012). As a reference to the trends in
the field, we used the full text of 4 recent papers that
argue about the future of the HCI field. The
discussion is initiated by the analysis of tag clouds
generated with the relevant data. As a contribution,
the paper reveals the individual characteristics of the
conferences, the similarities and differences between
them, the gaps in them, and the potential for future
research on the third wave of HCI (Bødker, 2006),
the third paradigm (Harrison et. al., 2007), new
transformations (Sellen et al., 2009) and human
centeredness (Bannon, 2011) in the field.
The paper is organized as follows: the second
section briefly presents the transformations in HCI
research based on the reference papers and the
conferences analyzed in this work; we also introduce
the related concepts and rationale for the use of tag
clouds as data representation. The third section
describes the methodology used for data extraction,
to create the tag clouds, and to perform the analysis.
The fourth section presents and discusses the
findings from the analysis. The last section presents
the final considerations about the work and
directions for future research.
2 STUDY CONTEXT
Traditionally, HCI has been defined as “a discipline
concerned with the design, evaluation and
implementation of interactive computer systems for
human use and with the study of major phenomena
surrounding them” (ACM SIGCHI, 1996).
Historically, the HCI field emerged in the early
1980s from the confluence of a variety of concerns
about human aspects and their relationship with
computers (Bannon, 2011). Since then, several
conferences, symposiums, workshops, etc. have
been organized to discuss the area’s issues.
According to Harrison et. al. (2007), looking back
over the history of HCI publications, the HCI field
arises from engineering research and, later, from
cognitive science. For Bannon (2011), studies based
on human factors, engineering, and ergonomics all
focused on improving the “man-machine fit,” and
the concern was to maximize industrial productivity
through optimal utilization of technology and the
most effective exploitation of human labor. This
optimization often seemed to fit the person to the
machine, rather than vice-versa, when machines
were expensive, and people at that time were not
able to afford them. For Harrison et. al. (2007) and
Bødker (2006), this scenario (in which the concrete
problems arise during interaction and cause
disruption in the relationship between humans and
computers inside the work places) is the centre of
the first paradigm and the first wave of HCI.
The second paradigm, which is directly oriented
by cognitive science, aims at understanding the
structure and functioning of the human mind, and is
organized around a central metaphor of the mind and
the computer as coupled information processors
(Harrison et al., 2007). For Bødker (2006), the
second wave focuses on groups working with a
collection of applications, where rigid guidelines,
formal methods, and systematic testing were
changed for proactive methods such as a variety of
participatory design workshops, prototyping and
contextual inquiries.
In the third paradigm, the concept of the user
changes because users are immersed into a context
with physical and social situations, and the interface
should be designed for any location, time, social
situation, and surrounding system. For this, a range
of disciplines (from the arts to sociology to politics)
and perspectives appear to establish multiple
interpretations of the site of interaction (Harrison et
al., 2007). In a similar way, the third wave tries to
understand the changes of the nature of human-
computer interaction in face of new technologies
(e.g., pervasive technologies, augmented reality,
small devices, tangible interfaces). The usage
context and application types are broadened and
intermixed. For this, new elements of human life are
HCIinContext-WhattheWordsRevealaboutIt
135

included in the human-computer interaction;
cognitive issues are expanded to include emotional
issues, and pragmatics and culture are embodied in
the user experience.
In this sense, the authors discuss the possibility
of reimagining HCI as a new way to think about the
human-technology relationship. Bannon (2011)
suggests a perspective which considers the user in
many stages of technology development, and it takes
into account his/her understanding, culture, values,
concerns, beliefs and activities. Sellen et. al. (2009),
suggest redefining the three elements that define H-
C-I (human, computer, and interaction).
These papers on the prospective view of HCI
through a third paradigm (Harrison et al., 2007),
third wave (Bødker, 2006), new transformations
(Sellen et al., 2009) and human centeredness
(Bannon, 2011) will be used as a reference for the
analysis of trends in this study.
2.1 Conferences Analyzed
In this work, the analysis was conducted using two
conferences with tradition in the HCI field: ACM
CHI and IFIP Interact.
Since the Conference on Human Factors in
Computing Systems (ACM CHI) was created in
1982, it has held annually, more frequently in the
United States and Canada. Sporadically, the
conference was held in other countries: Italy (2008)
and Holland (1993) (ACM SIGCHI, 2012). In this
paper, the ACM CHI conferences held between
2008 and 2011 were chosen for analysis.
The Conference on Human-Computer Interaction
(IFIP Interact) began in 1984 in the city of London
in the UK, and since then has taken place in
countries on several continents. From 1995 on, it
was held every two years (TC13, 2012). This study
analyzes the editions held in Portugal, Sweden and
Brazil, in the years of 2011, 2009 and 2007,
respectively.
The ACM CHI and IFIP Interact are promoted
by the two largest international associations that
bring together practitioners, researchers and students
interested in HCI: the Association for Computer
Machinery (ACM) and the International Federation
for Information Processing (IFIP) and its Technical
Committee on Human-Computer Interaction
(TC13), respectively (ACM SIGCHI, 2012; TC13,
2012).
2.2 Tag Cloud Representations
and Tools
For an overview of the themes appearing in the
conferences that were analyzed, we used tag cloud
representations. A tag cloud is a visual
representation of a set of words, typically tag words
(labels), which gained notoriety when used in social
software websites as “del.icio.us
®
” and “Flickr
®
”.
Each word is highlighted within the cloud according
to its frequency within the word set, and it is
enhanced through the manipulation of visual
features, such as font size, color, and weight
(Bateman et. al., 2008).
For Rivadeneira et. al. (2007), this format is
useful for quickly providing the most prominent
terms and the relative importance of a specific word
within the analyzed set. Also, it provides a general
impression of all words and the “essence” of the
represented data set. For instance, on social media
websites, tag clouds can give an impression of the
person’s interests or/and expertise.
In this work, tag clouds are used for first
impression formation. There are several tools that
help to create tag clouds from a source text. The tool
used in this study was Wordle (2013). The
occurrence of each word in the source text is
grouped together and the most recurrent words are
visually stressed. One way to prevent similar words
from appearing separately is to apply the Porter
Stemming Algorithm (Porter, 1980) in the source
text, which groups similar words by recurrence, to
organize the words in wordlist with the weight
(frequency) of words (as defined in advanced
options), and to use the advanced options to create
tag clouds.
Compared to a word list, which is equivalent to
the results offered by most search engines, the tag
clouds are less effective for identifying relationships
among concepts. However, they are advantageous
when capturing the essence and when succinctly
presenting a large amount of descriptive
information, thus improving user satisfaction (Kuo
et al., 2007).
This success scenario, and the need for a
summarized presentation of large amount of data,
are the reasons we chose tag clouds as one of the
resources in the analysis conducted in this study. For
more accurate analysis, other representations were
used to complement the analysis.
ICEIS2013-15thInternationalConferenceonEnterpriseInformationSystems
136

3 THE STUDY METHOD
Considering that the title of a text “indicates the
general subject” (Merriam-Webster, 2013) and
summarizes the essence of a publication, this study
used titles from the complete programs of the
conferences, more specifically, titles of papers,
technical sections, workshops, tutorials, posters, and
demos.
Figure 1: Steps in the Method.
The method involved data collection and refinement,
tag cloud generation, word quantification and the
comparison of sets. For this, the data was divided
into three large blocks, as illustrated in Figure 1. In
the “Data Refinement” stage, the goal was to gather
information about existing content in papers and
titles from the data present on the conference
websites. In the “Tag Cloud Analyses” stage, the
idea was to create tag clouds from the refined data.
The “Comparison of sets” stage was important for
developing the relationships among the lists created
from the tag clouds and for identifying differences
and similarities among the sets of word.
The titles were extracted from the full program
of the conferences, which were available online for
public access (item 1 in Figure 1). The data that was
irrelevant to the analysis, such as authors’ names,
presentation times, and affiliations, were
disregarded. Depending on the way data was
available (item 2 in Figure 1), the titles were
extracted either manual (for programs with less than
7000 words) or automatically (for programs with
more than 7000 words).
Regarding the reference papers, we extracted the
full texts of the four papers that indicated the trends
in the HCI field (Bannon, 2011; Bødker, 2006;
Harrison et. al., 2007; Sellen et. al., 2009) (item 1’
in Figure 1). In this case, the texts were extracted
manually (item 3 in Figure 1) and it was necessary
to remove the hyphen of separated words found at
line breaks.
As a result of “Data Refinement”, the data was
ready for tag cloud generation (item 5 in Figure 1). It
contained all of the titles, and they were gathered in
text files organized by conference/year. In addition,
the full texts of the four reference papers were
included, along with the title, abstract, keywords and
acknowledgements, but excluding the references.
Tag clouds were created with Wordle (2013)
(item 6 in Figure 1) for each group of data (IFIP
Interact and ACM CHI and Reference papers set).
However, before this, the Porter Stemming generate
Table 1: Analized sets of words.
Selectedwordsets Explanation
(a) IFIPInteract∩Ref.Papers
(b) ACMCHI∩Ref.Papers
(c) IFIPInteract∩ACMCHI
It shows the words that are common to the two
conferences, or to a specific conference and the
referencepapers.
(d) IFIPInteract∩ACMCHI∩Ref.Papers It shows the common words that appear in the
conferencesandreferencepapers
(e) IFIPInteract‐(Ref.Papers∩IFIPInteract∩ACMCHI)
(f) ACMCHI‐(Ref.Papers∩IFIPInteract∩ACMCHI)
(g) Ref.Papers‐(Ref.Papers∩IFIPInteract∩ACMCHI)
Itshowsthewordsthatappearexclusivelyinaspecific
conferenceorgroupofpapers.
HCIinContext-WhattheWordsRevealaboutIt
137

wordlists with the weight of each word. With the
wordlists ready, the advanced features were used,
along with parameters such as language of text
(defined as “English”), layout (defined as rounder
edges, black color and horizontal orientations),
prefer alphabetical order (checked as true) and
maximum words (defined as 100) were used in order
to generate tag clouds that were more coherent and
suitable for further analysis.
As a result of the “Tag Cloud Analyses” stage, a
comprehensive overview of the conferences was
visible by comparing the different images created
(item 7 in Figure 1). Moreover, it was possible to
extract the list of the top 100 most recurrent words
in each tag cloud (item 8 in Figure 1) to develop
three sets of words.
To emphasize the highlights observed in the tag
clouds, some sets of words were compared in the
“Comparison of Sets” stage. To make the
comparison among the groups, we defined an initial
subset of words (item 9 in Figure 1). The subset
defined in this paper is shown in the Table 1.
To facilitate the comparison among the word
sets, code was generated to match similar words
(item 10 in Figure 1) and to separate the different
words in different groups.
To refine the analysis, the words were allocated
to the sub-areas of the H-C-I field (Human,
Computer and Interaction) (item 11 in Figure 1).
Table 2 describes the criteria used to classify the
words in the HCI sub-areas. The definitions of the
classes were extracted from the reference papers set
(Bannon, 2011); (Bødker, 2006); (Harrison et al.,
2007); (Sellen et al., 2009).
Table 2: H-C-I sub- areas.
Area Definitions
Interaction
Userinterfaceandfeatures,typesof
interaction,concepts,challengesand
otherman‐machinerelationships.
Human
Users,experiences,activitiesand
behaviors,cultural,socialandwork
issuesthataredirectorindirectly
associatedtousers.
Computer
Devices,documents,software
applications,methodsandformalissues
relatedtotechnology.
The analysis of sets and subsets (item 12 in Figure
1), along with the tag clouds created (item 7 in
Figure 1), support the discussion (item 13 in Figure
1) in the following sections.
4 SYNTHESIS OF RESULTS
AND DISCUSSION
Based on the material produced during the Data
Refinement stage, Table 3 presents the total number
of words contained in all titles for each
conference/year. All together, around 2,700 titles
containing more than 27,000 words were gathered.
In Table 3, the cell content represents the number
of words in the conference analyzed in that year.
The cells marked with “---” indicate that there was
no conference in the year indicated (e.g., IFIP
Interact 2008 and 2010) or that the data was not
accessible for this study (e.g., ACM CHI 2007).
Table 3: Number of words per conference/year.
ACMCHI IFIPInteract Total
2007 ‐‐‐ 1619 1619
2008 3044 ‐‐‐ 3044
2009 6386 2222 8608
2010 3938 ‐‐‐ 3938
2011 7423 2714 10137
Total 20791 6555 27346
The papers considered as reference in the analysis
have a total number of words represented in Table 4.
All together, over 27,000 words were extracted.
Table 4: Number of words per reference paper.
(Harrisonet.
al.,2007)
(Bødker,
2006)
(Sellenet.
al.,2009)
(Bannon,
2011)
Total
11820 5349 5475 4672 27316
The tag clouds created from the titles (Figures 2 and
3) bring together the titles of all years of the ACM
CHI and IFIP Interact conferences, respectively, in
which the number of words is presented in the last
line of Table 3.
In a comparative analysis between the two
images (Figures 2 and 3) we observe that:
• “Design” and “Interaction” appear more
prominently, followed by the word “User”, for
both conferences, as expected.
• “Evaluation” and “Interfaces” are more salient in
IFIP Interact and are less emphasized in ACM
CHI.
• Both conferences show the word “Mobile”
although there seems to be more emphasis on it in
ACM CHI.
• The term “Social” appears in both conferences, at
the second prominence level in ACM CHI, while
at the fourth level in IFIP Interact.
ICEIS2013-15thInternationalConferenceonEnterpriseInformationSystems
138

Figure 2: Tag cloud of titles for the ACM CHI conference.
Figure 3: Tag cloud of titles for the IFIP Interact conference.
Figure 4: Tag cloud for reference papers content.
• “Usability” is the third most salient term in IFIP
Interact and fourth in ACM CHI.
• The word “Visualization” appears only in the IFIP
Interact tag cloud. The correspondent word
(“Visual”) appear with less emphasis in ACM
CHI.
• “Experience” and “Information” appear more
frequently in ACM CHI than in IFIP Interact.
• “Study” appears in both conferences, although
with more emphasis in ACM CHI.
• “Collaboration” and “Communication” are less
emphasized in ACM CHI. The opposite occurs for
the word “Online”. The word “Web” appears in
both conferences with the same frequency.
• “Computing”, “Display”, and “System” appear
more prominently in ACM CHI conferences. The
opposite occurs with the words “Applications” and
“Techniques”. “Supporting/Support” keeps the
same proportion in both.
The tag clouds created from the reference papers
content, represented by Figure 4, are composed of
the content of the four articles presented in Table 4.
Figure 4 shows that:
• “HCI” and “Design” are the most prominent
words.
• “Human”, “Interaction”, “Paradigm” and “Work”
are highlighted at the second salient level.
• “Approaches”, “Computer”, “People” and
“Technology” appear at the third salient level
• “User”, “Values” and “Wave” appear at the fourth
prominence level.
Looking at Figures 2, 3 and 4, we can see some
common words, as expected (e.g., “Design” and
“User”). Some words associated with technology
were also seen (e.g., “Computer/Computing”,
“Technology” and “System” itself). More
importantly, it is also possible to observe that some
highlighted words appear only in the reference
papers: “Paradigm”, “People”, “Work”, “Values”
and “Wave”. These new words are good
representatives for the subjects addressed in the
reference papers, and some of them, such as
“People” and “Values”, may indicate a difference in
HCI perspectives between the conferences and the
HCIinContext-WhattheWordsRevealaboutIt
139

new demands pointed out by the reference papers.
As a result of the “Comparison of Word Sets”, it
was possible to extract relationships among the
subgroups formed by the lists of the 100 most
relevant words of the conferences and reference
papers set (item 8 of Figure 1).
Table 5 indicates the number of words found at
the intersection of the set of words formed from
ACM CHI, IFIP Interact and reference papers set
(e.g., the IFIP Interact and ACM CHI tag clouds
have 64 words in common). Table 5 data suggests
that:
Table 5: Number of words in the intersection of sets.
(a)IFIP
Interact∩
ACMCHI
(b)Ref.
Papers∩
IFIPInteract
(c)Ref.
Papers∩
ACMCHI
(d)Ref.Papers
∩IFIPInteract
∩ACMCHI
64words 25words 28words 20words
• ACM CHI and IFIP Interact have approximately
65% of their most frequent words in common
(column 1 in Table 5), suggesting a good
alignment of research between conferences.
• The intersection between the reference papers and
IFIP Interact (column 2 in Table 5), and the
reference papers and ACM CHI (column 3 in
Table 5), have around 25% of the most frequent
words in common. This finding might suggest that
there is a place for HCI research that has yet to be
filled by the conferences in terms of new demands
for the field.
• The intersection of the words presented in ACM
CHI, IFIP Interact and reference papers (column 4
of Table 5) is even smaller. Only 1/5 of the most
frequent words are common to all sets. This
reinforces the opportunities that have not yet been
explored in the conference works regarding new
trends in the HCI field.
Tables 6, 7, 8 and 9 present four subsets of words
defined in Table 1 and classified in the H-C-I sub-
areas. Table 6 shows the word subsets (item 11 of
Figure 1) of the intersections among the two
conferences and the reference papers, allocated to
the H-C-I sub-areas. The percentage was calculated
with respect to every word classified.
In Table 6, we can see that most of the words
(over 46%) are associated in the “Computer”
column, which suggests that words sets in common
largely involve the area of technology. All together,
5 words cannot be allocated.
Tables 7, 8 and 9 show what the IFIP Interact,
ACM CHI and reference papers set are publishing,
respectively, resulting in sets of 80 words each. The
main objective is to show the individual
characteristics of each data set and to make
comparisons among them.
Table 6: Allocation of the words ((d) subset - Table 1).
(d)Ref.Papers∩IFIPInteract∩
ACMCHI
Total=20words
Interaction Human Computer
Experience,
interaction.
Human,people,
personal,
cognitive,social,
user.
Design,digital,
evaluation,
methods,process,
research,
technology.
13.3%(2words) 40%(6words) 46.6%(7words)
5unclassifiedwords: understanding,hci,multiple,
information,work.
Tables 7, 8 and 9 show what the IFIP Interact, ACM
CHI and reference papers set are publishing,
respectively, resulting in sets of 80 words each. The
main objective is to show the individual
characteristics of each data set and to make
comparisons among them.
Table 7: Allocation of the words ((e) subset - Table 1).
(e)IFIPInteract‐ (Ref.Paper∩IFIP
Interact∩ACMCHI)
Total=80words
Interaction Human Computer
Gestures,multi‐
touch,touch,
tangible…
Awareness,children,
context,cultural,
privacy,public...
3d,devices,
games,mobile,
video,networks,
web…
24,6%(15words) 24,6%(15words) 50,8%(31words)
19unclassifiedwords:based,combining,making,novel…
Table 7 refers to the words appearing exclusively in
IFIP Interact; we observe that more than 50% of the
words are in the “Computer” column, which
represents the technology sub-area. The “Human”
and “Interaction” columns contain 24.6% of the
valid words. Of the 80 words used in the rating, 19
words were not classified.
Table 8: Allocation of the words ((f) subset - Table 1).
(f) ACMCHI‐ (Ref.Paper∩IFIP
Interact∩ACMCHI)
Total=80words
Interaction Human Computer
Multi‐touch,
space,touch,
tangible…
Behavior,
children,group,
privacy,world...
Applications,audio,
devices,mobile,
tabletop,software…
18,2%(12words) 34.9%(23words) 46,9%(31words)
14unclassifiedwords:improving,towards,field,real,study…
Regarding the words that were exclusive to ACM
CHI, Table 8 shows that the words are distributed
incrementally among “Interaction” (over 18%),
“Human” (over 34%), and “Computer” (over 46%).
All together, 14 words were not classified within the
80 words related to ACM CHI.
ICEIS2013-15thInternationalConferenceonEnterpriseInformationSystems
140

Table 9: Allocation of the words ((g) subset - Table 1).
(g)Ref.Papers‐(Ref.Paper∩IFIP
Interact∩ACMCHI)
Total=80words
Interaction Human Computer
1st,2nd,3rd,
human‐centered,
wave…
Communities,
context,life,lives,
perspective,values...
Desktop,machine,
physical,model,
systems…
35,3%(18words) 37,3%(19words) 27,4%(14words)
29unclassifiedwords:argue,moved,rather,seems,shift…
Regarding data that was exclusive to reference
papers set, Table 9 shows the “Human” column with
more valid words (over 37%), followed by
“Interaction” column (over 35%), and finally the
“Computer” column (over 27%). In this set of data,
29 words were not classified.
Analyzing data from Tables 7, 8 and 9:
• If we compare the conferences (Tables 7 and 8)
with the reference papers (Table 9), there is an
inversion on the number of valid words between
the classes relative to “Computer”, “Human” and
“Interaction”. While the conferences are more
focused on the technology class (“Computer”), the
reference papers set seem to shed more light on the
“Human” and “Interaction” classes.
• Around 50% of words that were exclusive to either
ACM CHI or IFIP Interact fall in “Computer”
columns in Tables 7 and 8.
5 CONCLUSIONS
Computational technology in the modern world is
changing the way we interact and communicate. The
design of new interaction and communication
devices and their presence in people’s lives have
required new theoretical and methodological
frameworks to support HCI professionals in a
context far more complex than those of the first
decades of the discipline. Getting an overview of the
main issues that have been addressed in recent years
is a way of identifying whether the main discussions
around the HCI field are aligned to this
contemporaneity of the technology in our lives. In
this work, we conducted an analysis on the main
focuses of research addressed by the ACM CHI and
IFIP Interact conferences using the words coming
from contribution titles compared to the demands for
future topics to be addressed as argued by some
reference papers. Informal tests conducted for the
same publications, which included as input data,
titles, abstracts, and keywords, have shown no
significant difference in the tag clouds generated
only with the publication titles.
The main findings show that the ACM CHI and
IFIP Interact contributions seem to be aligned in
terms of their research focuses (approximately 65%
of common words). “Design”, “Interaction” and
“User” are terms that appear with more emphasis in
both conferences. Some differences are related with
frequency of terms as “Evaluation” and “Interfaces”
(more highlighted in IFIP Interact), and “Social”
(more highlighted in ACM CHI). Also, the results
suggest that both conferences show more
contributions on technology issues (e.g., “Digital”,
“Computer” and “Technology”), while the reference
papers seem to place more emphasis on human
issues and interaction in terms of the future of the
field (e.g., “Perspective” and “Life”).
Finally, the results showed that the demands of
our current life with technology are still being
modestly explored in the conferences that we
studied. Words such as “Values”, “People” and
“Lives” are not yet prominent in the conference
works. This suggests opportunities of research for
the discipline. As further work, we intend to analyze
whether the focus has shifted over the years and to
extend the analysis to include other HCI
communities.
ACKNOWLEDGEMENTS
This work is partially funded by CNPq through the
EcoWeb Project (#560044/2010-0).
REFERENCES
ACM SIGCHI, 1996. Curricula for Human-Computer
Interaction by Hewett, Baecker, Card, Carey, Gasen,
Mantei, Perlman, Strong and Verplank Copyright ©
1992.
ACM SIGCHI, 2012. Retrieved from:
http://www.sigchi.org, on Jun 15, 2012.
Bannon, L., 2011. Reimagining HCI: Toward a More
Human-Centered Perspective. Interactions, 18(4), p.
50-57.
Bateman, S., Gutwin, C., Nacenta, M., 2008. Seeing things
in the clouds: the effect of visual features on tag cloud
selections. In HT’08, Nineteenth ACM Conference on
Hypertext and Hypermedia. ACM Press. p. 995-998.
Bødker, S., 2006. When second wave HCI meets third
wave challenges. In NordiCHI’06, 4th Nordic
Conference on Human-computer Interaction:
Changing Roles. ACM Press. p. 1-8.
Buchdid, S.B., Baranauskas M.C.C., 2012. IHC em
contexto: o que as palavras relevam sobre ela. In
IHC’12, 11th Brazilian Symposium on Human Factors
in Computing Systems. ACM Press. p. 199-208.
Harrison, S., Tatar, D., Sengers, P., 2007. The three
HCIinContext-WhattheWordsRevealaboutIt
141

paradigms of HCI. In Alt.CHI, CHI’07. ACM Press. p.
1-18.
Merriam-Webster, 2013. Retrieved from:
http://www.merriam-webster.com, on Jun 05, 2012.
Kuo, B. Y-L., Hentrich., T., Good., B. M., Wilkinson., M.
D., 2007. Tag clouds for summarizing web search
results. In WWW200, 16th International Conference
on World Wide. ACM Press. p. 1203-1204.
Porter, M.F., 1980. An algorithm for suffix stripping.
Program, 14 no. 3. p. 130-137
Rivadeneira, A.W., Gruen D. M., Muller, M. J., Millen D.
R., 2007. Getting our head in the clouds: toward
evaluation studies of tagclouds. In CHI’07, SIGCHI
Conference on Human Factors in Computing Systems.
ACM Press. p. 995-998.
Sellen, A., Rogers, Y., Harper, R., Rodden, T., 2009.
Reflecting Human Values in the Digital Age.
Communications of the ACM, 52(3), p.58-66.
TC13, 2012. Retrieved from: http://www.tc13.org, on Jun
10, 2012.
Wordle. 2013. Retrieved from: http://www.wordle.net, on
Jan 10, 2013.
ICEIS2013-15thInternationalConferenceonEnterpriseInformationSystems
142