DEVELOPMENT AND EVALUATION OF A COMPUTER

GAME FOR TEACHING VOCABULARY

TO CHILDREN WITH AUTISM

Rafael Moreira Cunha and Simone Diniz Junqueira Barbosa

Informatics Department, PUC-RIO, Rio de Janeiro, Brazil

Keywords: Autism, Education, HCI, Computer Games.

Abstract: Autism is a global development disorder that affects children's ability to communicate, to establish

relationships and to respond appropriately to the environment. Within Computer Science, the area of

Human-Computer Interaction (HCI) can assist in building high quality tools that can help the development

of children with autism. This work aims to analyze the impact of a computer game developed exclusively to

help children with autism in vocabulary acquisition. The results obtained so far show that children were able

to effectively learn new words. Evidence of generalization and retention has been shown.

1 INTRODUCTION

Autism was first identified by (Kanner, 1943). He

described a group of seriously injured children who

had certain common characteristics. The most

noticeable was the inability to relate to people. Since

then, autism has been a subject of research in

different areas within universities.

According to the 10th International Statistical

Classification of Diseases and Related Health

Problems of 1992 (ICD-10), autism is considered a

Pervasive Developmental Disorder, which would

present a pattern of abnormal development and/or

impairment, manifested before the age of three. The

abnormal function of the child's skills can be related

to three areas: social interaction, communication and

imagination (Figure 1). This triad represents a

broader view of the autistic spectrum and forms the

basis for current diagnostic criteria, as described in

the fourth edition of the Diagnostic and Statistical

Manual of Mental Disorders (DSM-IV). The

diagnostic criteria expressed in the DSM-IV Autistic

Disorder are consistent with the criteria in the ICD-

10 for Infantile Autism, both listed under the general

heading of Pervasive Developmental Disorder. The

DSM-IV includes a diagnostic category of ‘Autistic

Disorder’ in the section titled Pervasive

Developmental Disorders. The diagnostic criteria for

the DSM-IV and the ICD-10 for autism are almost

identical, although in the ICD-10 it is known as

‘Childhood Autism’. For a diagnosis of autism, at

least two symptoms of impairment in social

interaction, one symptom of impairment in

communication, and one symptom of restricted and

repetitive behavior must be present.

These

characteristics should be evident until the age of

three, although diagnosis is usually done later.

Figure 1: Autistic triad.

According to the fourth edition of the Diagnostic

and Statistical Manual of Mental Disorders Revised

(APA, 2000) this developmental disorder affects

1:100 children, with higher incidence in males (3:1,

against 4:1 in female children). In Brazil,

epidemiological studies of the disorder have not

been done, but based on statistical studies conducted

in other countries (Fombonne, 2010), it is estimated

that in Brazil there are 500,000 people under twenty

years old with this disorder. And approximately

386

Moreira Cunha R. and Diniz Junqueira Barbosa S..

DEVELOPMENT AND EVALUATION OF A COMPUTER GAME FOR TEACHING VOCABULARY TO CHILDREN WITH AUTISM.

DOI: 10.5220/0003940403860392

In Proceedings of the 8th International Conference on Web Information Systems and Technologies (WEBIST-2012), pages 386-392

ISBN: 978-989-8565-08-2

 2012 SCITEPRESS (Science and Technology Publications, Lda.)

115,000 of these are children aged 0 to 4 years.

Although incurable, it is possible to ameliorate the

developmental delay caused by the disorder. This is

possible through early detection of the disorder

coupled with specialized treatments (Schopler et al.,

1995). Ideally, treatment should begin preferably

even before three years of age (Robins et al., 2001;

Wetherby et al., 2004).

Treatment should be performed by a

multidisciplinary team consisting of professionals

from different areas of health. In addition, other

approaches are commonly used in treatment, such as

therapeutic horseback riding, music therapy, speech

therapy, swimming, contact with animals and,

recently, interventions using the computer.

Computers are considered promising tools to

assist people with autism. Some of the early

researches showed positive results in using the

computer to treat people with autism dates back to

1973 (Colby, 1973). Currently there are several

works that explore the use of computers to support

the development of people with autism. There are

jobs in the area of virtual reality (Charitos et al.,

2000; Kerr, 2002; Parsons and Mitchell, 2002),

robotics (Dautenhahn, 1999), face recognition

(Baron-Cohen and Golan, 2006), cooperative games

(Piper, 2006), and several others that use the

technology for this purpose.

One of the main factors that motivate the use of

computers to treat these people is that with

computers you can create controlled environments

that are interesting and without distractions. These

features are considered important for success in

treating people with autism (Dautenhahn, 2000;

Putnam and Chong, 2008).

The aim of this work is to develop and evaluate a

computer game created specifically to teach

vocabulary to children with autism aged 5 to 10

years old.

2 THE VOCABULARY GAME

The process of language comprehension and

acquisition is complex. A person must have many

skills to be able to understand and comprehend the

meaning of the spoken words. This means

understanding the meaning and having the ability to

process the sounds of the spoken word. This process

is the basis of the language use in communication.

The acquisition of vocabulary is an important

component in this process (Gupta and Macwhinney,

1997). Studies show that the breadth and depth of

vocabulary knowledge is related to the ability of

good reading (Stanovich, 1986; Wood, 2001), and

also to academic success (Vermeer, 2001).

2.1 Game Characteristics

In designing software for children with autism it is

especially important that the interface should be

designed in a socially responsible way. Although

children with autism are a diverse group, there are

characteristics commonly found in autism (NAS,

2004) which were considered in the design of this

game. The following items summarize the design

decisions (DD) related to certain characteristics:

• They are likely to prefer predictable, structured

and controlled procedures and environments

and, possibly as a consequence, they are often

attracted to inanimate objects, machines and

computers. DD: A computer version of a paper-

based vocabulary game is developed.

• They may have difficulty with remote object

references (proto-imperative pointing), and may

have difficulty understanding the use of a

mouse. DD: The game is presented using a

touch sensitive screen, allowing direct

manipulation using a finger. The possible

answers in the lower part of the screen can be

draggable, giving a strong analogy with the

physical world. Mouse or touchpad can also be

used.

• They are generally highly visual. DD: Figures

can be used throughout the whole training.

• They tend to focus on particular details and

prefer local than global integration (Happé,

1997). Consequently, a child with autism might

focus exclusively on some apparently irrelevant

details, e.g. great interest might be taken in the

experimenter’s glasses, or a program can be

used not for its primary purpose, but just for the

pleasure of the accompanying noises. DD:

Screen design is kept very simple with no

complex features.

• They may be highly sensitive to noise, finding

intolerable noise, which is barely perceptible or

unremarkable to others (Bogdashina, 2003).

DD: Sound features can be turned off.

• They may have problems with manual dexterity.

DD: A simple docking function aids the final

placement of the child’s chosen answer.

• They may find failure very debilitating (Jordan

and Powell, 1995). DD: There is no penalty for

a wrong answer. The accompanying adult may

encourage another attempt if appropriate to the

DEVELOPMENTANDEVALUATIONOFACOMPUTERGAMEFORTEACHINGVOCABULARYTOCHILDREN

WITHAUTISM

387

particular child at the time.

Challenges, needs, and skills vary by age of a

child and severity of their diagnosis, and each case

of autism is unique. So, the game can be

personalized for each child. It is possible to change

its options and adapt them over time. Goals can be

set for each child, individually.

2.1.1 Game Procedure

The game is based on matching-to-sample

procedure, which is a form of conditional

discrimination.

The basic operations in the matching-to-sample

task are: first, the sample stimulus and an array of

choice stimuli are presented; second, the participant

responds by selecting a stimulus from the array

which he/she believes that corresponds to the sample

stimulus; third, informative feedback is given as to

whether the correct choice was made.

The essential components of the task consist of

the sample stimulus followed by a set of choice

stimuli, which includes a stimulus that corresponds

to the sample. Based on some rule or relationship,

the participant should respond by choosing the

correct one.

In the actual application, the sample and choice

stimuli are linguistic stimuli. Although the sample

and choice stimuli have a correspondence, they may

vary widely in terms of their actual properties. For

example, a very simple task may be matching

picture to picture, while a more complex one would

consist of matching a word to picture. Table I shows

all the existing possibilities of matching in the game.

The game proceeds in the following the sequence:

• The stimulus item and the array of choice

items are presented on the screen.

• The child is (verbally) asked to locate the

identical item using the touch screen, mouse

or touchpad.

• If the child chooses the correct response,

there is a display congratulating him/her.

Otherwise, the incorrect item wiggles, as if

saying “No, not me”.

• The trial is repeated until the child reach

criterion.

2.1.2 Game Training Cycle

The game training is based on a Discrete Trial

Training (DTT) cycle. DTT is the primary teaching

method for a number of the behaviourally based

interventions used in teaching children with autism.

Table 1: Matching-to-Sample Possibilities.

Sample stimulus is a word

and the choice stimuli are a

set of pictures

Sample stimulus is a picture

and the choice stimuli are a

set of pictures

Sample stimulus is a picture

and the choice stimuli are a

set of words

Sample stimulus is a word +

picture and the choice

stimuli are a set of pictures

Sample stimulus is a word +

picture and the choice stimuli

are a set of words

Sample stimulus is a word

and the choice stimuli are a

set of words

Children with autism often face many deficits and

difficulties in learning (Schopler et al., 1995).

Discrete-trial training can help to compensate for

these difficulties.

A discrete trial is a single cycle of a behavior

based instruction routine. A particular trial may be

repeated several times in succession, several times a

day, over several days (or even longer) until the skill

is mastered. There are four parts, and an optional

fifth, to a discrete trial.

• Discriminative stimulus (S

) - The instruction

or environmental cue to which the teacher

would like the child to respond.

• Prompting stimulus (S

) - A prompt or cue

from the teacher to help the child respond

correctly (optional).

WEBIST2012-8thInternationalConferenceonWebInformationSystemsandTechnologies

388

• Response (R) - The skill or behavior that is the

target of the instruction, or a portion thereof.

• Reinforcing stimulus (S

) - A reward designed

to motivate the child to respond and respond

correctly.

• Inter-trial interval (ITI) – A brief pause

between consecutive trials.

The parts of the discrete trial are often

represented symbolically in the following format,

adopted in the remainder of this work:

Î R Î S

Î ITI

)

Essentially, this illustrates the order of a discrete

trial. First there is the teacher's instruction (S

). If

the teacher thinks the child may need some help

responding correctly, she will give him/her a little

prompt, cue, or model to help him/her out (S

Then, either with help or without, the child gives

some response to the instruction (R). If the child

responds incorrectly, the teacher might correct

him/her, and then give him/her another chance. If

the child responds correctly, or close to correctly,

the teacher might give him/her some reward or

praise to encourage him/her (S

). After that

sequence is completed, the teacher might want to

pause for a little while before continuing, to let the

child know that they have completed one cycle and

have moved on to the next (ITI).

The game followed the same idea behind DTT.

Initially, the main character makes a request (S

). If

there is delay in response time, the character

reinforces the request (S

). A guess (S

) can be

customized to be presented sometimes during the

game. When a correct response occurs (R), the

character congratulates the child and an animation

appears.

2.2 Game Goals

Research on treatments to help children with autism

should include the retention and generalization of

behaviors (Schreibman and Ingersoll, 2005).

Behaviors refer to any skill or knowledge gained by

an individual during treatment. Learning new words

is a behavior that can be taught and measured.

Retention and generalization can occur separately

for a behavior; however, it is desirable to ensure that

both occur (Bregman et al., 2005). This investigation

measured the retention and generalization of learned

material after a period of time.

2.2.1 Retention

Retention is the ability to remember over a period of

time (Smith, 1994). Retention of a new word would

entail remembering the word after a period of time

with no exposure or training in that word.

There is little value to learning skills that are

forgotten (Powers, 2005). Time spent teaching skills

that are not maintained is time wasted. If the skill

was worth teaching in the first place, the skill should

be either automatically maintained by the child in

the natural environment or it should be learned so

well that it is easily remembered when necessary.

Retention could include periodic review or practice

of material so that it is remembered.

2.2.2 Generalization

Generalization is the ability of an individual to

demonstrate a behavior outside the treatment

environment (Bregman et al., 2005). Behaviors that

are generalized can be used with different stimuli,

people, and situations. Generalization of new words

would be evident by the ability to recognize a word

in a different form, such as other pictures or objects,

which represent the word that was previously taught

by the computer. Other aspects of generalization

include the use of a new word that is spoken by a

different person or used in a different setting. New

knowledge must be transferred to different materials

and environments to be useful. Words learned on the

computer should be useable in the real world. This

investigation evaluated the ability of the participant

to generalize his learning from the computer game to

objects and flash cards.

3 METHODOLOGY

3.1 Participants

The criteria for selection of participants included

prior diagnosis of autism, availability of family, and

a trained professional to mediate the search. To date,

this research involved three children aged five to

nine years. The vocabulary known by the children

was verified using receptive tests for identification

of approximately ten among 105 significant figures.

In the receptive tests, children were only asked to

point at the correct word, and they did not need to

say it aloud.

DEVELOPMENTANDEVALUATIONOFACOMPUTERGAMEFORTEACHINGVOCABULARYTOCHILDREN

WITHAUTISM

389

3.2 Ethical Issues

The research was carefully designed to maximize

student learning and avoid subjecting them to

situations of discomfort. Their parents have carefully

read and signed the consent form.

3.3 Procedure

The study consisted of four phases: pre-tests, pre-

training, training and post-tests.

3.3.1 Pre-Tests

This phase consisted in the identification of words

unknown to the child. 105 words were used in the

same size cutouts of small-format cards.

The words represented by the figures correspond

to nouns of different categories: fruit, transportation,

animals, insects, furniture, electronics, cooking,

music, tools, buildings, food, clothing, hygiene, and

school.

The pictures were displayed in pairs and the

child was asked to receptively indicate (i.e., to just

point at) one of the figures through requests such as:

"Show me the apple." The criterion to consider the

word correct was the hit of at least three consecutive

times in which the same word was shown alongside

different sets of figures. Three consecutive errors

indicated otherwise. Since it was just a pre-test,

there were no scheduled consequences for hit and

miss and no help or correction procedure was used

during this phase.

This phase was considered finished when

approximately ten words were identified as

unknown to the child.

3.3.2 Pre-Training

Before starting the training, participants were taught

to use the proposed game. At this stage the

participant used only the known vocabulary. The

criterion for completion of this phase was to hit five

times the target word.

3.3.3 Training

At this stage, children trained words that were

unknown as first identified in the pre-test phase. The

training included ten sessions. Generally, sessions

took fifteen to thirty minutes each.

To teach new words, we use both an identity and

an arbitrary model. In the identity model, both the

word to be learned and the array of stimuli are in the

same type of representation, i.e., both images or both

words. In the arbitrary model, there is a combination

of words and images.

The game issued for each hit a compliment for the

participant to keep performing the activities until the

end of the training; these interventions were

contingent on successes. There were no programmed

consequences for error situations.

The criterion for completion of the task was to

correctly answer at least five consecutive attempts

involving the same word without an error. While the

criterion was not reached, the word was not

switched.

3.3.4 Post-Tests

At this stage, we verified if the retention of the

taught words, i.e., the extent to which these words

were remembered over time.

(Biemiller and Boote, 2006) tested retention after

two periods of non-exposure treatment, 2 weeks and

6 weeks. In their study, the retention period was 4

weeks.

For our assessment, we used the same figures

and procedures of the pre-tests. Generalization tests

also occurred during this phase. We followed the

same procedure as the pre-tests, including physical

objects.

The game only required a receptive response, not

an expressive one. The generalization from receptive

to expressive is not guaranteed with children with

disabilities (Biemiller and Boote, 2006). Thus, the

participant was tested to determine if the words that

he could select receptively in the game could also be

labeled expressively.

To determine whether generalization occurred

also in a significant way, we conducted tests in

which a picture was presented to the child and

requests were made or questions like “What is it?”

were posed. The child should say the name of what

was presented. Objects were also used in this way.

4 RESULTS

The evaluation assessed the extent to which

developed the game contributed to the acquisition,

retention, and generalization of vocabulary, as well

as a subjective evaluation of each child's response to

the program. Data were captured and analyzed in

two distinct moments of the study: training sessions

and post-tests. The results obtained so far show that

children were able to significantly learn some

vocabulary, with retention of about 94% of new

words after 30 days. Table II provides the individual

WEBIST2012-8thInternationalConferenceonWebInformationSystemsandTechnologies

390

performance of each child. As can be seen in the

table, children were also able to generalize the

acquired knowledge (M = 0.94, SD = 0.10).

The three participating children seemed to enjoy

working with the game. They came for the

evaluation sessions requesting the "squirrel", which

is the game’s main character. We have also observed

that the praises fired by the main character made the

children happy.

Table 2: Number of trained words, total number of words

retained and generalized after a period of training and total

training time for each participating child after 10 sessions.

Child Trained Retained Generalized

Training

Time

A 9 9 9 0:53:30

B 12 12 12 1:08:28

C 11 9 9 1:05:15

Total 32 30 30 3:07:13

Average 10.67 10.00 10.00 1:02:24

SD 1.53 1.73 1.73 0:07:53

5 CONCLUSIONS

The evaluation of the known vocabulary is an

indicator of cognitive development. Acquiring

words means not only to be able to speak, but also

understand the spoken word and look it up in

memory. There is always a need for individuals to

use their cognitive ability to understand and interact

to the world. Moreover, without a large repertoire of

words, it is difficult for children to create new

conceptual categories (Bee, 2003). Therefore, this

study aimed to analyze the impact of a computer

program developed exclusively to help children with

autism in vocabulary acquisition.

The computer program developed for this

research combined visuals, sound effects, incentives,

and reinforcements. Preliminary results showed that

children could learn at least two nouns at each

session. Generalization and retention tests also

showed that children were able to recall the nouns

trained and to apply the acquired knowledge in

novel situations. In addition, they were highly

motivated to interact with the computer.

This study is limited by the small sample and by

the potential variance in responses by children with

autism. Additional research should be conducted to

examine more children with autism with different

abilities. The effects of computer instruction, the use

of collaborative as well as individual computer

activities, and the specific visual and auditory

production techniques that promote interest and

learning should also be examined.

In conclusion, we hope that this work can

contribute to and promote other projects in the area

of information technology in special education,

especially in the education of children with autism.

ACKNOWLEDGEMENTS

We are very grateful to parents and children who

participated in the research. Special thanks to Dr.

Carolina Lampreia, from Department of Psychology

at the Catholic University of Rio de Janeiro (PUC-

Rio). We are also grateful to the speech therapists

Luciana de Almeida Moreira Reis e Regina de

Souza Pires.

This research was made possible by support

from Globo.com (Globe Network).

REFERENCES

Kanner, L., 1943. Autistic disturbances of affective

contact. Nerv Child 2:217-250.

World Health Organisation, 1992. ICD-10 Classifications

of Mental and Behavioural Disorder: Clinical

Descriptions and Diagnostic Guidelines. Geneva.

World Health Organisation.

American Psychiatric Association, 2000. Diagnostic and

statistical manual of mental disorders (4th ed., text

rev.). Washington, DC: Author.

Fombonne, 2010. Encontro Brasileiro para Pesquisa em

Autismo, 1., 2010, Porto Alegre. The epidemiology of

Autism and PDDs. Porto Alegre: Eric Fombonne,

2010.

Schopler, E., Mesibov, G. B. and Hearsey, K., 1995.

Learning and Cognition in Autism (New York:

Plenum Press, 1995). Part of the series Current Issues

in Autism. ISBN: 0306448718.

Robins, D. L., Fein, D., Barton, M. L., Green, J. A., 2001.

‘The Modified Checklist for Autism in Toddlers: An

Initial Study Investigating the Early Detection of

Autism and Pervasive Developmental Disorders’,

Journal of Autism and Developmental Disorders 31

(2): 131–44, 2001.

Wetherby, A.M., Woods, J., Allen, L., Cleary, J.,

Dickinson, H., Lord, C., 2004. Early Indicators of

Autism Spectrum Disorders in the Second Year of

Life, Journal of Autism and Developmental Disorders

34 (5): 473–93, 2004.

Colby., K. M., 1973. The rationale for computer-based

treatment of language difficulties in nonspeaking

autistic children. Journal of Autism and Childhood

Schizophrenia. 1973, 3~254-260.

Charitos D., Karadanos G., Sereti E., Triantafillou S.,

Koukovinouand S. and Martakos D., 2000. Employing

Virtual reality for aiding the organisation of autistic

DEVELOPMENTANDEVALUATIONOFACOMPUTERGAMEFORTEACHINGVOCABULARYTOCHILDREN

WITHAUTISM

391

children behaviour in everyday tasks. in. Proc. 3rd Intl

Conf. Disability, Virtual Reality and Assoc. Tech.,

Alghero, Italy, pp. 147-152, 2000.

Kerr, S., J., 2002. Scaffolding: design issues in single &

collaborative virtual environments for social skills

learning. In Proceedings of the Workshop on Virtual

Environments 2002. W. Stürzlinger and S. Müller,

Eds. ACM International Conference Proceeding

Series, vol. 23, pages 81–91. ACM Press, May 2002.

Parsons, S., Mitchell, P., 2002. The potential of virtual

reality in social skills training for people with autistic

spectrum disorders. Journal of Intellectual Disability

Research, 46(5):430–443, June 2002.

Dautenhahn K., 1999. Robots as social actors: Aurora and

the case of autism. In Proc. CT99, The Third

International Cognitive Technology Conference,

August, San Francisco, pages 359-374, 1999.

Baron-Cohen S. and Golan O., 2006. Systemizing

empathy: Teaching adults with asperger syndrome or

high-functioning autism to recognize complex

emotions using interactive multimedia. Development

and Psychopathology, 18(2):591–617, June 2006.

Piper, A. M., O’Brien, E., Morris, M. R., Winograd, T.,

2006. SIDES: A Cooperative Tabletop Computer

Game for Social Skill Development. In: Proc.

CSCW’06, Banff (Canada), pp.1–10. ACM Press,

New York (2006).

Dautenhahn K., 2000. Design issues on interactive

environments for children with autism. In Proceeding

3rd Intl Conf. Disability, Virtual Reality & Assoc,

page 153159, Alghero, Italy, 2000.

Putnam C., Chong L., 2008. Software and technologies

designed for people with autism: what do users want?

Proceedings of the 10th international ACM

SIGACCESS conference on Computers and

accessibility, October 13-15, 2008, Halifax, Nova

Scotia, Canada.

Gupta, P., & MacWhinney, B., 1997. Vocabulary

acquisition and verbal short-term memory:

computation and neural bases. Brain and Language,

59, 267–333.

Stanovich, K. E., 1986. Matthew effects in reading: Some

consequences of individual differences in the

acquisition of literacy. Reading Research Quarterly,

21, 360–406.

Wood, J., 2001. Can software support children’s

vocabulary development? Language Learning &

Technology, 5, 166–201.

Vermeer, A., 2001. Breadth and depth of vocabulary in

relation to L1/L2 acquisition and frequency of input.

Applied Psycholin- guistics, 22, 217–234.

NAS. National autistic society: Last accessed April 2004,

avail- able from: http://www.nas.org.uk

Happé, F., 1997. Central coherence and theory of mind in

autism: reading homographs in context. Br. J. Dev.

Psychol. 15, 1–12.

Bogdashina, O., 2003. A reconstruction of the sensory

world of autism. In: Proceedings of the 7th

International Autism-Europe Congress.

Jordan, R., Powell, S. D., 1995. Understanding and

Teaching Children with Autism. Wiley, New York.

Schreibman, L., & Ingersoll, B., 2005. Behavioral

interventions to promote learning in individuals with

autism In F.R. Volkmar, R. Paul, A. Klin, & D. Cohen

(Eds.), Handbook of Autism and Pervasive

Developmental Disorders, Vol 2. (pp 882- 896).

Hoboken, NJ: John Wiley & Sons, Inc.

Bregman, J.D., Zager, D., & Gerdtz, J., 2005. Behavioral

interventions. In F.R. Volkmar, R. Paul, A. Klin, & D.

Cohen (Eds.), Handbook of Autism and Pervasive

Developmental Disorders, Vol 2. (pp 897-924).

Hoboken, NJ: John Wiley & Sons, Inc.

Smith, T., 1994. Improving memory to promote

maintenance of treatment gains in children with

autism. The Psychological Record, 44(4), 459.

Powers, M. D., 2005. Behavioral assessment of

individuals with autism: A functional ecological

approach. In F. R. Volkmar, R. Paul, A. Klin, & D.

Cohen (Eds.), Handbook of Autism and Pervasive

Developmental Disorders, Vol 2. (pp 817- 830).

Hoboken, NJ: John Wiley & Sons, Inc.

Biemiller, A. and Boote, C., 2006. An Effective Method

for Building Meaning Vocabulary in Primary Grades.

Journal of Educational Psychology, 98(1), 44-62.

Bee, H., 2003. A criança em desenvolvimento (9a ed.).

Porto Alegre: Artmed.

WEBIST2012-8thInternationalConferenceonWebInformationSystemsandTechnologies

392