Towards an Approach for Improving Exploratory Testing Tour

Assignment based on Testers’ Proﬁle

Let

ıcia De Souza Santos

, Rejane Maria Da Costa Figueiredo

1,2 a

, Rafael Fazzolino Pinto Barbosa

Auri Marcelo Rizzo Vincenzi

, Glauco Vitor Pedrosa

1,2 b

and John Lenon Cardoso Gardenghi

1,2 c

Information Technology Research and Application Center (ITRAC), University of Brasilia (UnB), Brasilia, DF, Brazil

Post-Graduate Program in Applied Computing, University of Brasilia (UnB), Brasilia, DF, Brazil

Federal University of Sao Carlos (UFSCar), Sao Carlos, SP, Brazil

john.gardenghi@unb.br

Keywords:

Recommendation Systems, Exploratory Testing, Proﬁle of Testers.

Abstract:

This work presents an empirical study on the relationship between the testers’ proﬁle and their efﬁciency and

preference in the application of tours with tourist metaphor for exploratory software testing. For this purpose,

we developed and applied a questionnaire based model to gather as much as possible information about the

knowledge, expertise and education level from a group of testers. The results indicated that, in fact, the testers’

proﬁle have impact on the application of tours used in the tourist metaphor: there are differences between the

tours preferred by different levels of education and most of testers tend to choose those tours based on what

they believed to have the shortest execution time. This work raises a valuable discussion about a humanized

process of assigning test tasks in order to improve the efﬁciency of software testing.

1 INTRODUCTION

Software testing is an arduous and expensive activity.

In this sense, there are opportunities for developing

strategies to reduce the test execution time and to in-

crease defect ﬁndings. One approach is to allocate test

cases according to the testers proﬁle in a way to max-

imize testing productivity. However, optimizing the

allocation of manual test cases is not a trivial task: in

large companies, test managers are responsible for al-

locating hundreds of test cases among several testers.

Studies such as (Anvik et al., 2006) and (Miranda

et al., 2012) have shown that it is possible to employ

recommendation systems to allocate tasks to speciﬁc

proﬁles based on the analysis of previous allocations.

So, a recommendation system can assist in assigning

test to a team of testers, seeking to contribute to the

teams’ productivity.

Through the Exploratory Testing (ET), it is pos-

sible that the tester does not depend on a set of pre-

designed test cases, as ET contains the steps of design

and test execution. To systematize the ET process,

https://orcid.org/0000-0001-8243-7924

https://orcid.org/0000-0001-5573-6830

https://orcid.org/0000-0003-4443-8090

the work in (Whittaker, 2009) proposed the Tourist

Metaphor, which draws an analogy between software

testing and a tourist visiting in a city. Tourism is a

mixture of structure and freedom, just like the Ex-

ploratory Test.

The tester’s proﬁle inﬂuences the application of

tours used in the Tourist Metaphor (Miranda et al.,

2012). In this sense, a way to maximize productiv-

ity in a test team is to allocate tasks according to the

testers proﬁle. This is due to the understanding that

the generated test cases and their sequences vary a lot

depending on the tester.

This paper presents an empirical study to support

the identiﬁcation of testers’ proﬁle for the use of the

Exploratory Testing approach considering the Tourist

Metaphor. A group of 60 testers was interviewed to

gather information about their education level, exper-

tise, computational knowledge and preference among

those tours considered in the tourist metaphor. The

idea is to raise correlations between this information

in order to develop a test recommendation system.

In summary, the main contributions of this work

are:

• Literature review: we raised references on the re-

lationship between the testers’ proﬁle and the as-

signment of test tasks;

Santos, L., Figueiredo, R., Barbosa, R., Vincenzi, A., Pedrosa, G. and Gardenghi, J.

Towards an Approach for Improving Exploratory Testing Tour Assignment based on Testers’ Proﬁle.

DOI: 10.5220/0011113800003179

In Proceedings of the 24th International Conference on Enterprise Information Systems (ICEIS 2022) - Volume 2, pages 183-190

ISBN: 978-989-758-569-2; ISSN: 2184-4992

183

• The development of a questionnaire for tester’s

proﬁle identiﬁcation;

• Experimental study on the relationship between

the testers’ and their efﬁciency in the application

of tours in the context of Exploratory Tests with

the Tourist Metaphor;

• Data analysis to support the deﬁnition of a Rec-

ommendation System for automatic assignment of

test tasks with the Tourist Metaphor.

The remaining of this paper is organized as fol-

lows: Section 2 discusses related works and presents

some deﬁnitions concerned to our work; Section 3 de-

scribes the methodology adopted in our work; Section

4 presents the proﬁle of the group of professionals in-

volved with our data collection; Section 5 presents the

results and discussions and Section 6 ﬁnalizes with

the conclusion and future works.

2 BACKGROUND

The theoretical background of this study is based on

the software testing and the exploratory test with the

tourist metaphor. In the following, we present some

deﬁnitions and correlated works that motivated the

development on this work.

2.1 Software Testing

The testing activity is essential for Software Engineer-

ing, as it is a tool to ensure the quality of the software

product from the identiﬁcation of failures during its

development (Myers et al., 2004) when fault correc-

tion is cheaper. The feedback of real behaviors makes

testing a fundamental quality assurance analysis tech-

nique in the industry, although it may require a lot of

human work and the scientiﬁc community considers

more the use of automated tests (Bertolino, 2007).

Testing by human testers is relevant to real-world

software development as it allows the identiﬁcation of

new BUGs, especially in interactive systems. Human

testers have advantages over machines, given their ca-

pacity for knowledge, learning and adaptation to new

situations, which facilitates the process of efﬁcient

recognition of problems (Itkonen et al., 2015).

Through the Exploratory Testing (ET), it is pos-

sible that the tester does not depend on a set of pre-

designed test cases, since this test approach contains

the steps of design and test execution, in which testers

are constantly learning and adapting activities. In a

practical way, the tester learns iteratively about the

product and its failures, designs and executes the tests

dynamically and systematically (Whittaker, 2009).

2.2 Exploratory Test with the Tourist

Metaphor

To systematize the ET process, the work in (Whit-

taker, 2009) presented the Tourist Metaphor, which

draws an analogy between software testing and a

tourist visiting in a city. According to (Whittaker,

2009), tourism is a mixture of structure and freedom,

just like the exploratory test.

In the analogy presented, the software features are

separated into “districts”, that one decides to explore.

Each district has a set of “tours”, which represent the

different ways of going trough the characteristics and

functionalities of the software. This analogy helps the

test team to communicate about what should be tested

and how.

In this work, we considered 23 tours present in 6

districts. Among these 23, 16 were used in the inter-

view with testers. Therefore, the descriptions of the

16 tours chosen by the participants are presented in

the following.

• Tours Trough the Business District

– Intellectual Tour: Run the software with inputs

so that it operates under conditions of maxi-

mum load or that require more processing.

– Landmark Tour: Determines what are the main

characteristics of the product (reference points)

and in which order to visit them.

– Garbage Collector’s Tour: Test all menu items,

all error messages, among others, and to visit

each one in a methodical way, going through

the shortest path.

– Guidebook Tour: Follow the user manual.

– FedEx Tour: Look for identiﬁes where the data

is changed in order to assess if they are being

corrupted on the way.

• Tours Trough the Historical District

– Bad-neighborhood Tour: Visit areas of code

full of defects. Focus test effort on areas with

the highest concentration of defects.

• Tours Trough the Entertainment District

– Back Alley Tour: Visit the less attractive func-

tionalities from the user’s point of view.

– Supporting Actor Tour: Regardless of the sales-

people’s efforts, the customer often ends up be-

ing more interested in peripheral characteristics

than the main ones.

– All-nighter Tour: Challenge the software seek-

ing to popularize the same data and force con-

secutive readings and writes of the values of the

variables.

ICEIS 2022 - 24th International Conference on Enterprise Information Systems

184

• Tours Trough the Tourist District

– Supermodel Tour: Look for superﬁcial defects

in the software product related to its appear-

ance.

– Collector’s Tour: Visit every possible location

of the software and document every output ob-

tained.

• Tours Trough the Hotel District

– Couch Potato Tour: Work as little as possible.

Even if the tester isn’t doing much, it doesn’t

necessarily mean that the software isn’t.

– Rained-out Tour: Identify a list of time-

consuming operations to perform. Start an op-

eration and then stop it.

• Tours Trough the Seedy District

– Antisocial Tour: Do the opposite of what is ex-

pected in the software.

– Saboteur: Limit access or exclude required re-

sources.

– Obsessive-Compulsive Tour: Repeat, redo,

copy, paste, borrow, the same action several

times in a row.

2.3 Impact of Testers’ Proﬁle in

Software Testing

As a human-based activity, the results of a software

product test are dependent on human factors and pose

challenges for software development teams, such as

the search for a more effective way to increase testers’

motivation and satisfaction (Deak et al., 2016).

For the past 50 years, Software Engineering has

been concerned with the inﬂuence of human person-

ality on individual work tasks, as the systematic lit-

erature review done by (Cruz et al., 2011) points

out. In fact, some works on exploratory tests con-

clude that the human personality can inﬂuence this

test method (Bach, 2003; Whittaker, 2009; Itkonen

et al., 2015; Itkonen et al., 2012; Shoaib et al., 2009).

The actions of testers during the application of ex-

ploratory tests can vary signiﬁcantly from one person

to another, that is, the methodology adopted is directly

related to the personality traits of each tester (Shoaib

et al., 2009).

The experiment carried out by (Shoaib et al.,

2009) was designed to identify the testers who can

achieve the best result during the application of ex-

ploratory tests. The results showed a positive relation-

ship between the exploratory test and human person-

ality traits. In addition, organizations have adopted al-

ternative methodologies and workforces to efﬁciently

deliver software (Dubey et al., 2017).

The work in (Berner et al., 2005) claims that au-

tomated testing can never completely replace manual

testing. (Martin et al., 2007) presents reports which

state that the problems related to software testing in

the industry involve the company’s socio-technical

environment and organizational structure.

The relationship between software testing and the

human aspect was studied by (Shah and Harrold,

2010) in the context of a service-based company. The

results showed that the attitudes of older professionals

can signiﬁcantly inﬂuence the attitudes of more inex-

perienced people.

In this sense, as already addressed by (Cruz et al.,

2011), the team’s performance may vary depending

on its members and their personalities and experi-

ences. This contributes strongly to the present work,

since the laboratory involved is an environment com-

posed of team members at different levels of educa-

tion and experience and in continuous turnover.

Based on the literature reviewed and the tech-

niques adopted in previous studies, this work seeks

to apply concepts related to the proﬁle of the tester

in order to gather data to develop a recommendation

system for case test based on the Tourist Metaphor.

3 METHODOLOGY

The methodology adopted by this work comprises

four basic phases: (1) research planning, (2) data col-

lection, (3) data analysis, and (4) reporting the results.

The interviews were developed with undergradu-

ate course of Software Engineering at our university,

engaging 40 participants, and a specialization course

in the area of Computer Science, involving 20 partic-

ipants. These participants were the object of study.

In the data collection phase, the research procedures

employed were: documentary research; bibliographic

research; and action research. A questionnaire was

applied to each participant, in order to identify their

characteristics.

The process of digital transformation in the con-

text of this work suggests a large number of low com-

plexity services, which allows for a large number of

test cycles in a short period of time. This character-

istic favors learning and evolution related to testing

activities.

Figure 1 presents the continuous learning pro-

cess, together with the methodology proposed for this

work. The left part of Figure 1 refers to the Analy-

sis and is linked to the collection of the proﬁle of the

tester and the analysis of the data collected after each

test cycle. The right part of Figure 1 lists what is in-

tended to be used as an experiment, which makes it

Towards an Approach for Improving Exploratory Testing Tour Assignment based on Testers’ Proﬁle

185

possible to record the test cycles and propose Attribu-

tion.

Figure 1: Approach to the Recommendation System.

The proﬁle of each tester will be drawn from the

information extracted from the questionnaire applied.

The recommendation strategy suggested in this re-

search can use information from the preferences for

each tour by the participants of the test dynamics and,

thus, determine which tours are most suitable for cer-

tain testers, based on the history of tests carried out

and reported preferences.

It is important to highlight that the execution of a

tour by a tester gives rise to a set of tests generated

from the tour . The assignment of tours and testers

should be done dynamically, with a learning process

during each test cycle. In other words, the attribution

of tours and testers must consider, in addition to char-

acteristics, the learning obtained during the previous

test cycles. Learning can maximize the efﬁciency of

applying test cases. In this context, the word ”efﬁ-

ciency” refers to the failure identiﬁcation rate during

the application of the test cases it is given by Equa-

tion 1.

efﬁciency =

number of failures identiﬁed

number of test cases of the tour

(1)

In this work, as proposed in (Miranda et al., 2012),

the assignment is represented throughout the text as a

set of test and tester pairs, represented by {CTn;Tn}

(Test Case n and Tester n). The user is represented by

a typical test manager and recommendations are made

by comparing a speciﬁc test case with the proﬁle of a

tester.

A correlation is found between the efﬁciency in

the test process and the different variables that make

up the proﬁle of the tester. This proﬁle will be iden-

tiﬁed based on different questions answered by the

testers in a digital questionnaire.

After the questionnaires are answered, the next

step is to collect data on the efﬁciency of each tester

in carrying out the test activities. To collect this data,

correlation tests are performed to identify whether

there is any relationship between the proﬁle variables

and the efﬁciency in the tests of a given tester or the

testing team as a whole.

The data explain the impact of the variables of the

proﬁle of the tester on the efﬁciency of the tests. Af-

ter this stage, it is possible to use the questionnaire

to make a correlation test between the variables sur-

veyed.

4 DATA COLLECTION

The interviews to collect information on the testers’

proﬁle were conducted with professionals and stu-

dents engaged with our university. In summary, the

collected data contains information from:

• 40 undergraduate students of Software Engineer-

ing

• 20 professionals of a specialization course in the

area of Computer Science

The collected data interviews were developed with

undergraduate course of Software Engineering at our

university, engaging 40 participants, and a specializa-

tion course in the area of Computer Science, involving

20 participants. These participants were the object of

study. In the data collection phase, the research pro-

cedures employed were: documentary research; bib-

liographic research; and action research. A question-

naire was applied to each participant, in order to iden-

tify their characteristics.

To extract information to support an automatic test

task assignment process, the action research proce-

dure was selected. Therefore, activities were carried

out in a participatory and interactive way by the re-

searcher and the participants.

The following steps were taken: identiﬁcation and

survey of personal characteristics to be addressed in

the questionnaire; extraction of information about the

personal characteristics of each tester; monitoring the

test dynamics of a ﬁctitious service created with dif-

ferent characteristics; data analysis; and proposal of a

ﬁltering strategy to consider future recommendations.

It is possible that the participants with different

background knowledge and experiences in tests are

not considered when assigning test tasks performed

using Exploratory Tests. Therefore, students who are

at the beginning of their undergraduate course and

have no experience in tests, can be allocated to test

tasks that have the same level of difﬁculty as test tasks

allocated to experienced testers.

ICEIS 2022 - 24th International Conference on Enterprise Information Systems

186

The questionnaire applied in this study was based

on the survey conducted by (Geras et al., 2004).

Based on this, in order to characterize software tests

and quality assurance practices, the questionnaire was

divided into two categories: (i) personal and profes-

sional proﬁle; and (ii) speciﬁc knowledge about test-

ing;

The ﬁrst category was developed to collect per-

sonal information and the professional proﬁle of the

respondent, with an interest in understanding their

level of education, undergraduate courses and expe-

riences with programming languages.

The second category sought to evaluate the degree

of familiarity of the testers with the test activity. The

questionnaire addressed techniques and testing crite-

ria, for example, to assess the experience of testers

in this regard. In an application of the recommenda-

tion process proposed in this research, a third category

could encompass an open question, which asks the

tester to report their experience. The research team

intends to include this category in future works.

A ﬁctitious service was developed so that testers

could make tours from the Tourist Metaphor and ﬁnd

as many defects as possible. In order for the partic-

ipants to perform the test, a Guide for the Applica-

tion of Exploratory Tests was created. All the partic-

ipants already had minimal knowledge about the Ex-

ploratory Testing approach.

5 RESULTS AND DISCUSSIONS

The information collected from the questionnaire is

presented in graphic form, and refers to level of ed-

ucation; technologies most used; experience in Soft-

ware Testing; experience in Application Testing; ex-

perience in Exploratory Tests; and experience in each

test phase.

Figure 2 reveals the distribution of participants

involved in this research in relation to the level of

education. These levels varied considerably within

the three participating groups, and the largest number

compromised postgraduate students.

Figure 2: Distribution of Participants by Education Level.

The performance of tests considering different

professional expertise was an important information

gathered in our study, as some of the participants have

already had experiences in the labor market, while

others are in the middle of the Software Engineer-

ing course, which decreases their level of expertise

in comparison to graduate participants.

Figure 3 presents the techniques that the partici-

pants had some knowledge of or were skillful in. It

help us to predict how testers would be able to per-

form tests that would require further exploration of

software or, in the case of this research, a government

service.

Figure 3: Number of Participants by Technology.

Figure 4 presents the test phases in which the three

participant groups have greater experience. In the

software testing discipline, the students’ greatest ex-

perience was concentrated in Unit Tests, which had

already been practiced by 90% of the respondents.

Approximately 57% of the respondents declared to

have experience in Acceptance Tests, and 33%, in In-

tegration Tests. Meanwhile, about 20% declared ex-

perience in System Testing and only 10% declared to

have no experience in any test phase.

All respondents in the Experimental Software En-

gineering discipline declared that they had already

carried out unit tests, while 60% had already taken

Acceptance tests, 40% in Integration Tests and 30%

in System Tests.

The diversiﬁcation of the level of education allows

us to understand the different ways of looking at the

software and its possible defects. This statement be-

comes more evident in the discussion about the an-

swers obtained after performing the tests in the ﬁcti-

tious service.

The graduate class showed a level of 70% in unit

test knowledge, 50% in both integration tests and sys-

tem tests and 20% in acceptance tests. Also, 15% said

they had not worked in any of the phases indicated.

The results of the dynamics also showed the test-

ing techniques in that the members of each discipline

have experience. In the Software Testing discipline,

about 87% of the participants had already performed

tests using the Functional Test technique (black box

test), 57% had already used the structural test tech-

Towards an Approach for Improving Exploratory Testing Tour Assignment based on Testers’ Proﬁle

187

Figure 4: Level of experience in testing phases of the mem-

bers of each discipline.

nique (white box), 23% had performed a defect-based

testing technique, and 3% had never used any of the

testing techniques presented.

Regarding the Experimental Software Engineer-

ing class, all the respondents declared that they al-

ready had experience in some of the testing tech-

niques. 90% declared to have used the functional test

technique, 30% declared to have used structural test

and 10% had already used defect-based testing.

In general, respondents in the graduate course had

more experience in functional testing (70%). With

50% positive responses, the structural test was the

second technique mastered by students of the disci-

pline, while 30% had already tested it based on de-

fects. 15% stated that they had never performed the

testing techniques.

Table 1 shows the results of the tours most used by

students involved in the dynamics of Tests. The most

used one was the anti-social tour, done by 50 of the 60

students who participated in the dynamic. This tour

had already been noticed in the work of (Blinded Au-

thor(s), 0000), which presents a number of tests and

failures identiﬁed by tour, and reports the creation of

a process for validating services produced by digital

transformation.

A ranking was made with the tours most used

in each of the disciplines. Group 1 participants

showed greater interest in Antisocial, Couch Potato

and Obsessive-Compulsive tours.

It is possible to relate the choices of tours with the

proﬁles outlined by the questionnaire, given the levels

of knowledge. Group 2 and 3 participants presented

more consistent answers about their knowledge of the

types, phases, techniques, criteria and testing tools.

Their answers to the question were more complete

than those of Group 1.

The maturity with which the dynamics were

treated by the groups also inﬂuenced the data obtained

when taking the educational level into account. Group

1 presented a large number of responses in the dynam-

Table 1: Table with List of Tours Used by Testers.

Selected Tours

Group 1 Group 2

Group 3 Total

Antissocial 30 6 12 48

Couch Potato 30 4 8 42

Obsessive-Compulsive 23 3 4 30

Intelectual 18 0 10 28

Supermodel 17 4 4 25

Collector 18 2 3 23

Landmark 16 4 2 22

Saboteur 11 5 5 21

Garbage Collector 14 1 3 18

Guidebook 15 0 2 17

Back Alley 7 1 1 9

Rained-out 5 0 3 8

FedEx 8 0 0 8

Supporting Actor 3 1 0 4

Bad-neighborhood 3 0 0 3

All-nighter 0 1 1 2

ics; however, not all participants answered the pro-

ﬁle questionnaire. In addition, many of the defects

found were more related to an ad-hoc way that stu-

dents ended up testing, than to a way of testing de-

ﬁned by some tour, for example.

With the application of the Test dynamics in the

three participating groups, it was possible to perceive

the preferences in relation to the tours of the Tourist

Metaphor. This information was necessary to support

the creation of a test task recommendation system that

is based on the Exploratory Testing approach.

All the 30 Software Engineering undergraduate

participants in the Software Testing class (Group

1) made use of the Antissocial Tour and the

Couch Potato Tour, and 77% chose the Obsessive-

Compulsive Tour. The others demonstrated a more

widespread interest among the listed tours, as shown

in Table 1 .

Of the 10 Software Engineering undergraduate re-

spondents in the Experimental Software Engineering

discipline (Group 2), 60% opted for the Antisossial

Tour and 40% opted for the Intellectual tour, while

the rest of tours were more spread among the stu-

dents’choices.

Of the 20 participants in the graduate class (Group

3), 60% chose the Antisocial Tour, 50% took the In-

tellectual Tour, while the rest dissipated among the

other tours.

It is evident that some of the tours had a greater

adherence by the participants of the three groups.

This choice is related to the description of the tour

which, in the cases of the most listed tours, indicate

practical ways of testing software, explain how to per-

form the opposite of what was expected in a function-

ality (Antisocial), or put more load in some ﬁeld than

it should be able to support (Intellectual).

The Couch Potato Tour was strongly preferred in

ICEIS 2022 - 24th International Conference on Enterprise Information Systems

188

the Software Testing class, which consequently has

the majority of the younger and less experienced stu-

dents in terms of years in Software Testing. Most

of the tours that were little, or not chosen, required

more testing time. In view of the determined time of

the dynamics, the students chose the tours based also

on what they believed to have the shortest execution

time.

As shown in Figure 1, after collecting and ana-

lyzing proﬁle data from testers, it was be possible to

use the approach and, based on a future use of his-

torical data on test cases already registered, propose

an assignment of test tasks using the tours that were

ranked by the subjects.

Although there are differences between the tours

preferred by different levels of education, the ranking

presents options that could be proposed to different

levels of knowledge about tests, as shown by (Blinded

Author(s), 0000), who carried out his experiment with

a team composed mostly of undergraduate students.

After assigning the best tours ranked, it is possi-

ble to execute and record the test cases in order to

generate inputs for a future recommendation of tours,

concluding the ﬁrst cycle of a continuous process pro-

posed in this work and presented in Figure 1.

6 CONCLUSIONS AND FUTURE

WORKS

This study aimed to identify proﬁles of testers to sup-

port the creation of a test task recommendation sys-

tem based on the Exploratory Testing approach with

the Tourist Metaphor. For this, we sought to gather as

much relevant information as possible to assign test

tasks based on the proﬁle of the testers.

The information comes from both literature re-

view and empirical analysis, with a sample from three

groups of different levels of education, related to IT

and linked to the academy. This enabled the collec-

tion of information about proﬁles and the achieve-

ment of testing dynamics based on the Exploratory

Testing approach.

The personal characteristics of each tester inﬂu-

ence his work with software tests and deﬁne a basic

strategy for structuring a test process that is based on

human characteristics in order to direct the attribution

of test tasks. This strategy should consider both the

test history of each tester and their proﬁle, which are

incremented with each test cycle.

This study raises a valuable discussion about a hu-

manized process of assigning test tasks in order to

generate data for the deﬁnition of a recommendation

system for automatic assignment of test tasks based

on the proﬁle of each tester. In addition to testing

tasks, this strategy can be extended to development

contexts, given that the proﬁle of each developer, and

tester, can also inﬂuence the effectiveness of the ac-

tivity and the degree of satisfaction of the developer.

It is possible to highlight two main future works

derived from this research. The ﬁrst one is to extract

the proﬁles with more testers as a sample, in order to

follow the exploratory testing process carried out, to

build a consistent database on proﬁles. From a more

solid database, the second future work is to apply ar-

tiﬁcial intelligence algorithms for automatic assign-

ment of test tasks based on the proﬁle of testers.

Finally, consolidate the implementation of a rec-

ommendation system for assigning test tasks based on

the testers’ proﬁle.

REFERENCES

Anvik, J., Hiew, L., and Murphy, G. C. (2006). Who should

ﬁx this bug? In Proceedings of the 28th international

conference on Software engineering, pages 361–370.

ACM.

Bach, J. (2003). Exploratory testing explained.

Berner, S., Weber, R., and Keller, R. K. (2005). Observa-

tions and lessons learned from automated testing. In

Proceedings of the 27th international conference on

Software engineering, pages 571–579. ACM.

Bertolino, A. (2007). Software testing research: Achieve-

ments, challenges, dreams. In 2007 Future of Software

Engineering, pages 85–103. IEEE Computer Society.

Blinded Author(s) (0000). Blinded title. In Blinded Confer-

ence, pages 00–00.

Cruz, S. S., da Silva, F. Q., Monteiro, C. V., Santos, P.,

and Rossilei, I. (2011). Personality in software engi-

neering: Preliminary ﬁndings from a systematic litera-

ture review. In 15th annual conference on Evaluation

& assessment in software engineering (EASE 2011),

pages 1–10. IET.

Deak, A., St

alhane, T., and Sindre, G. (2016). Challenges

and strategies for motivating software testing person-

nel. Information and software Technology, 73:1–15.

Dubey, A., Singi, K., and Kaulgud, V. (2017). Personas

and redundancies in crowdsourced testing. In 2017

IEEE 12th International Conference on Global Soft-

ware Engineering (ICGSE), pages 76–80. IEEE.

Geras, A. M., Smith, M. R., and Miller, J. (2004). A sur-

vey of software testing practices in alberta. Cana-

dian Journal of Electrical and Computer Engineering,

29(3):183–191.

Itkonen, J., M

antyl

a, M. V., and Lassenius, C. (2012). The

role of the tester’s knowledge in exploratory software

testing. IEEE Transactions on Software Engineering,

39(5):707–724.

Itkonen, J., M

antyl

a, M. V., and Lassenius, C. (2015).

Test better by exploring: Harnessing human skills and

knowledge. IEEE Software, 33(4):90–96.

Towards an Approach for Improving Exploratory Testing Tour Assignment based on Testers’ Proﬁle

189

Martin, D., Rooksby, J., Rounceﬁeld, M., and Som-

merville, I. (2007). ’good’organisational reasons

for’bad’software testing: An ethnographic study of

testing in a small software company. In Proceedings

of the 29th international conference on Software En-

gineering, pages 602–611. IEEE Computer Society.

Miranda, B., Aranha, E. H. d. S., and Iyoda, J. M. (2012).

Recommender systems for manual testing: deciding

how to assign tests in a test team. In Proceedings of

the ACM-IEEE international symposium on Empirical

software engineering and measurement, pages 201–

210. ACM.

Myers, G. J., Badgett, T., Thomas, T. M., and Sandler, C.

(2004). The art of software testing, volume 2. Wiley

Online Library.

Shah, H. and Harrold, M. J. (2010). Studying human and

social aspects of testing in a service-based software

company: case study. In Proceedings of the 2010

ICSE Workshop on Cooperative and Human Aspects

of software Engineering, pages 102–108. ACM.

Shoaib, L., Nadeem, A., and Akbar, A. (2009). An empir-

ical evaluation of the inﬂuence of human personality

on exploratory software testing. In 2009 IEEE 13th In-

ternational Multitopic Conference, pages 1–6. IEEE.

Whittaker, J. A. (2009). Exploratory software testing: tips,

tricks, tours, and techniques to guide test design. Pear-

son Education, London, England.

ICEIS 2022 - 24th International Conference on Enterprise Information Systems

190