reverse engineering approach in a development of a ...€¦ · reverse engineering approach in a...

Applied Mathematical Sciences, Vol. 6, 2012, no. 108, 5369 - 5382

Reverse Engineering Approach in a

Development of a Decision Support System

for Forensic DNA Analysis

Noor Maizura Mohamad Noor

Department of Computer Science

Faculty of Science and Technology

Universiti Malaysia Terengganu (UMT)

Kuala Terengganu, Terengganu, Malaysia

Ahmad Faiz Ghazali





[email protected]

Md Yazid Mohamad Saman





Zafarina Zainuddin

School of Health Science, Health Campus

Universiti Sains Malaysia (USM)

Kubang Kerian, Kelantan, Malaysia

5370 Noor Maizura Mohamad Noor et al

Abstract

In getting software requirement and specification based on existing system,

reverse engineering (RE) can be done by creating abstraction views of a system in

order to help software engineer maintain and re-engineer it. Requirement

engineering and specification in a development of a decision support system (DSS)

presented in this paper is replaced by RE approach. Before that, software engineer

needs to identify the pattern of design used in an existing system’s

implementation, including programming details. RE can extract information of

software design from source code, but the level of abstraction and documentation

is yet to be discovered. RE can be done at least in three stages; RE of user

interfaces, RE to understand internal process of a system and RE of database.

Keywords: Decision Support System, Forensic DNA Analysis, Reverse

Engineering

1 Introduction

Generally, software is computer programs. Ian Sommerville [23] defined

software not only computer programs but also all the required documentation and

required configuration for computer programs to operate. According to Ian

Sommerville [23], software engineering (SE) is a process of software production,

beginning from software specification until software deployment and testing. As a

program evolves, it becomes increasingly difficult to understand and reason about

changes in the source code. Eventually, if enough changes are made, reverse

engineering (RE) techniques must be used in order to understand the current

behavior of a system. Roger S. Pressman [20] mentioned about the abstraction in

RE;

“Ideally, the abstraction level should be as high as possible. That is, the RE

process should be capable of deriving procedural design representations (a

low-level abstraction), data and control flow models (a relatively high level

of abstraction), and entity relationship models (a high level of abstraction).

As the abstraction level increases, the software engineer is provided with

information that will allow easier understanding of the program”.

Decision support system (DSS) is a concept of application that aims to help

decision-making processes [29]. DSS can means different thing to different

people, hence there is no universally accepted definition of DSS [29]. Turban and

Aronson [29] mentioned that “DSS is also sometimes used as an umbrella terms

to describe any computerized system that supports decision-making in an

organization.”. Today, DSS are being applied in multiple domains, including

forensic, medical, military, financial and engineering [22, 25, 29, 31].

In field SE, RE is a method of design recovery as mentioned a lot by Roger S.

Reverse engineering approach 5371

Pressman [20, 19] but not Ian Sommerville [23]. Using concept RE, where based

only on user interfaces, designs of internal and external data processing in a

system flow can be recovered. Besides, from further analysis, processes,

components, and its relationships among each other can be identified. After that,

the representation of the system is created at a higher level of abstraction or in

another form. In this research, a RE approach has been practiced to study Forensic

DNA Databank of Malaysia (FDDM) used at Royal Malaysian Police (RMP).

User interfaces of the software have been studied. Then, rapid prototype is built in

a platform of Web to ease accessibility. What is the problem encountered during

this design recovery processes? Information provided is very limited, due to

national security issue. The information that initially cannot be recovered

including:

i) Programming languages and technologies used to build the

application.

ii) Source codes

iii) Mathematical equations used for analysis.

iv) The meaning and purposes of the available information and menus

RE can reduce certain costs. In SE, development is a risky and expensive

proposition, hence improving the productivity of software development and the

quality of delivered software will result in significant economic returns [1].

2 Reverse Engineering

Researchers Okolica [18] mentioned about classical definition of reverse

engineering (RE) in 1990 by Cross [8] as “the process of analysing a subject

system to (i) identify its system’s components and their inter-relationships and (ii)

create representations of the system in another form or at a higher level of

abstraction”. RE for software as rephrase by Roger S. Pressman [20] is “the

process of analysing a program in an effort to create a representation of the

program at a higher level of abstraction than source code. Reverse engineering is a

process of design recovery”.

Software RE is a sub-discipline of SE [15]. It creates a suitable representation

of the system in a higher level of abstraction or another form to support the

comprehension of software systems. Figure 1 shows the schematic view of RE

process [20].


Dirty source code

Restructure

code

Clean source code

Extract

abstractions

Initial specification

Refine &

simplify

Final specification

Processing

Interface

Database

Figure 1: Reverse engineering process by Roger S. Pressman [20]

The ‘abstraction level’ here refers to the sophistication of the design information

that can be extracted from source code. Extract abstraction is the core activity of

RE. Three phases in RE are including RE of user interfaces, RE to understand

process, and RE to understand database. This concept is practiced as shown in

section Results and Discussion of this paper.

3 Decision Support System

Decision support system (DSS) has been applied in a wide range of computer

applications that commonly manipulate information in variety fields. This is

including computer-based systems that are being used in handling data in order to

achieve an optimization for decision-making in forensic science. Basic

components in a DSS consist of model base, database, user interface and

knowledge base. Information technology and human resources utilized in field

forensic science and computer science will be able to help supporting

decision-making using computerized forensic DNA analysis and paternity test.


In the early 1970s, the major concepts of DSS was first articulated by Scott

Morton [11] as “interactive computer-based systems, which help decision-makers

utilize data and models to solve unstructured problems”. Keen and Scott Morton

[14] defined DSS as “a computer-based support system for management decision

makers who deal with semi-structured problems”. DSS gather the human

intellectuals and computer’s capabilities to improve the quality of decisions [14].

Scientist working in forensic field will be able to provide DNA data for

analysis. Knowledge that can be interpreted from forensic will be able to help

scientist to make decisions or provide a new alternative for solutions. Other recent

research works in DSS come out with proposed software [13]. Vescoukis et al.

[30] and Miranda et al. [17] proposed architecture in their research work.

Researchers Tremblay et al. [28] doing related works to Web-based DSS concept.

4 Forensic DNA Analysis

Decision theory plays an important role in forensic science (FS) [6, 12, 16, 21,

24-27]. Deoxyribonucleic acid (DNA) profile analysis using computer

programming is not a new research in biology and information technology.

Projects have been done in developed countries and around the world to develop

software tools to analyze DNA data.Aitken and Gammerman [4] initiated the

discussion on the use of graphical probabilistic models for evaluating evidences in

FS, and then, the first example to focus on Bayesian networks (BN) and DNA

evidence is the work by Dawid and Evett [9] because it presents with example

involving blood stains, as reviewed by Biedermann and Taroni [7]. Dawid et al. [9]

discussed an example involves two kinds of scientific evidence which is more

complex than the one discussed by Aitken and Gammerman [4]. A different

approach has been presented by Garbolino et al. [10] where they do not primarily

focus on a specific scenario. The difference between the study by Aitken and

Gammerman [4] and [2, 3] is the nature of knowledge included into the graphical

models of BN. His research works [4] is created based on own judgement for

probability values, but models that he created later [2, 3] is come out based on

data.

Forensic DNA Analysis Software (ForAS) proposed in this research is proven to

be reliable and helpful for paternity test in forensic DNA analysis. The importance

of this research work is it is a great start to develop more applications and extended

software from ForAS for crime investigation. A new Web-based DSS to manage

overall crime cases from the identification of suspects until apprehension of

criminal offenders is currently under extensive development. It is another way of

designing from a different perspective of overall process involved in the related

systems. Technology is not the main focus, but the use and its relevancy to specific

designs are included [5].


5 Results and Discussions

Figure 2 shows the screenshots of implemented software after Stage 1 of

reverse engineering process for the functionality of Forensic DNA Databank of

Malaysia (FDDM). User interface design, especially the frame, looks almost

exactly like FDDM (it is not included here due to security issue). During this

stage, basic DNA profile match can already be performed and results displayed;

but using only mock data.

Proposed Visualization Application for forensic DNA

analysis system (Prototype Version 1.1)


Figure 2: Stage 1- First prototype after reverse engineering the functionality of FDDM

During Stage 2, almost overall user interface are changed from Stage 1

(compare Figure 2 and Figure 3). The location of menu, icons, and banners are

changed, but maintain the same functionality DNA profile match in forensic DNA

analysis.


Figure 3: Stage 2- Second prototype after redesign the user interfaces

Figure 4 shows the comparison of user interfaces design during Stage 1 and

Stage 4 for the core process (main engine) of forensic DNA analysis. The

similarities are both can be used to perform DNA profile match in forensic DNA

analysis. The difference is the software proposed during Stage 4 in this research

work applied more user-friendly interfaces. The arrangements of menu are much

better than the first software proposed during Stage 1 based on the results of

usability study that has been conducted (not included in the scope of this paper).

For each allele match, green colour is applied in the result of matching columns,

or red if vice versa.

DNA profile match in proposed software during Stage 1


DNA profile match in proposed software during Stage 4

Figure 4: Evolution of result’s presentation (Stage 1 and Stage 4)

A lot of programming works has been done during implementation at Stage 3

and Stage 4 of this research work following methodology software reverse

engineering and software reengineering. Figure 5 and Figure 6 shows the

visualization features that has been tried to propose to be included as an

enhancement in Web application for forensic DNA analysis. This is because this

feature is not provided in current FDDM, hence reverse engineering is not occur

for data visualization. The idea here is to find out the way to provide enhancement

compared to the existing forensic DNA analysis tools.

Figure 5: First version - an attempt to visualize allele frequency


Figure 6 shows the final version of visualization using concept visualization tree.

In the first version as in Figure 5, the frequency is tried to be visualized using plot

graph but it does not really helpful to domain expert compared to visualization

tree.

Figure 6: Final version - visualization tree for allele numbers

The Web application proposed is then extended into part specifically built to

handle forensic DNA analysis. It is implemented during the final stage for

enhancement. This paper has presented stages performed for design recovery to

identify the behaviour of user interface, nature of processing data and structure of

system flow. There are several issues that have been identified;

It is crucial to separate the design and implementation step-by step. This is

because, it is more difficult to redesign the process or system flow once it is

implemented compared to before implementation is initiated.

i) The design of architecture for proposed Web-based DSS was first

initialised by process reverse engineering of existing software. This

research has shown that an existing software development can be used

as a starting point to implement new software from the recovered

design.


ii) The unavailability of source code and information in the current

documentation is not an excuse for not being able to recover some of

the designs from the main system flow. An understanding of the

application domain such as software for domain experts like forensic

scientists, customers, or any other specific users can help the recovery

of information about the purpose of a function and its significance.

iii) Sometimes re-engineering has to be done after reverse engineering

because too little or outdated information is provided. There might be a

need to rewrite and recode the system to improve its usability, ease of

maintenance, or further enhancement with more functionality. It is

easier to change a design than source code. Thus, software

re-engineering after software reverse engineering will be useful in

terms of providing better design and implementation after design

recovery.

ForAS presented results of analysis with more detail information of DNA allele

frequencies. Allele numbers, allele frequencies, locus frequencies, estimated DNA

profile frequency, allele matching, locus matching, fractions of locus match and its

percentages are very helpful for forensic scientists. Allele and locus match in

columns of decision table with systematic and detail data representation is applied

with green colour for match and red colour for not match. Systematic and detail data

representation of statistical DNA analysis is very important for the ease and quality

decision-making process of a DSS.

7 Conclusion

In this research, reverse engineering has been done in a few stages. Stage 1

involving the reverse engineered of user interface. Then, reverse engineering of

data structures and algorithms is the crucial part. Statistical forensic DNA analysis

for paternity test is applied for main engine of the software. Contribution on the

concepts and operational basis for DSS that have been done in this research work

is implementing software development process in a Web-based DSS using reverse

engineering and re-engineering methods. Applying software RE approach to DSS

is one of the methods in developing Web-based DSS, besides just following

traditional Software Development Life Cycles (SDLC). The contribution is

architecture using RE approach for developing a Web-based DSS. Through RE of

software, the user interfaces, the data structures, the algorithms and how the

analysis being performed can be recovered. If some basic information of the

software is obtained, then further research can be done to obtain more

information.

Anyway, reverse engineering approach is successfully applied in this research

work. The procedure steps in design recovery can be described as; 5380 Noor Maizura Mohamad Noor et al

i) Collect available information: All possible information that is related

is obtained as much as possible, including source codes, designs and

its documentation if possible. Personnel experience may be included.

ii) Study provided information: Study and extensive reviews must be

done on the collected information so that more information can be

recovered and further plan can be initiated. The structure of internal

and external data can be identified as well.

iii) Sketch-out the windows of system flow: Once understood and

identified, the system flow of the observed software needs to be

written or drawn. Then, the system flow for the proposed software

with enhancement can be provided.

iv) Review again the recovered design: Review and recheck the design to

verify that it represents the structure of the software as desired and

planned. There is also a need to review the recovered design for

consistency with available information and to find out where it can be

improved from time to time.

In summary, this research was successfully using approach reverse engineering

to build a decision support system for forensic DNA analysis from scratch. The

original aim of DSS to use computerized system to support decision makers is

achieved. Hopefully, further investigation based on this research methodology for

software reverse engineering to build new software using concept decision

support system can be practiced. This will contribute to develop and enhance the

concept of DSS and its role to help decision-makers utilize information

technology for problem solving in future.

References

[1] Adolph, S., P. Kruchten, and W. Hall, Reconciling perspectives: A

grounded theory of how people manage the process of software

development. Journal of Systems and Software, 2012. 85(6): p. 1269-1286.

[2] Aitken, C.G.G., et al., Bayesian belief networks with an application in

specific case analysis. Computational Learning and Probabilistic

Reasoning. 1996: John Wiley & Sons.

[3] Aitken, C.G.G., et al., Statistical modelling in specific case analysis.

Science & Justice, 1996. 36: p. 245-255.

[4] Aitken, C.G.G. and A. Gammerman, Probabilistic reasoning in evidential

assessment. Journal of the Forensic Science Society, 1989. 29: p. 303-316.

[5] Bate, I.J. Journal of Systems Architecture - Embedded Software Design.

2012

[6] Biedermann, A., S. Bozza, and F. Taroni, Decision theoretic properties of

forensic identification: underlying logic and argumentative implications.

Forensic Science International, 2008. 177: p. 120-132.


[7] Biedermann, A. and F. Taroni, Bayesian networks for evaluating forensic

DNA profiling evidence: A review and guide to literature. 2012.

[8] Cross, C.a., Reverse Engineering and design recovery: a taxonomy. IEEE

Software, 1990. 7(1).

[9] Dawid, A.P. and I.W. Evett, Using a graphical method to assist the

evaluation of complicated patterns of evidence. Journal of Forensic

Sciences, 1997. 42: p. 226-231.

[10] Garbolino, P. and F. Taroni, Evaluation of scientific evidence using

Bayesian networks. Forensic Science International, 2002. 125(2-3): p.

149-155.

[11] Gorry, G.A. and M.S.S. Morton, A Framework for Management

Information Systems. Sloan Management Review, 1971. 13(1).

[12] Helsloot, I. and J. Groenendaal, Naturalistic Decision Making in Forensic

Science: Toward a Better Understanding of Decision Making by Forensic

Team Leaders. Journal of Forensic Sciences, 2011. 56(4): p. 890-897.

[13] Jullien-Ramasso, S., G. Mauris, and L. Valet, Bolon, A decision support

system for animated film selection based on a multi-criteria aggregation of

referees’ ordinal preferences. 2012. 39(4): p. 4250-4257.

[14] Keen, P.G.W. and M.S.S. Morton, Decision Support Systems: An

Organizational Perspective, Addison-Wesley, Editor. 1978.

[15] Kienle, H.M. and H.A. Muller, Chapter 5 - The Tools Perspective on

Software Reverse Engineering: Requirements, Construction, and

Evaluation. Advances in Computers, 2010. 79: p. 189-290.

[16] Lucena-Molina, J.-J., V. Pardo-Iranzo, and J. Gonzalez-Rodriguez,

Weakening Forensic Science in Spain: From Expert Evidence to

Documentary Evidence. Journal of Forensic Science, 2012.

[17] Miranda, J., P.A. Rey, and J.M. Robles, udpSkeduler : A Web architecture

based decision support system for course and classroom scheduling. 2012.

[18] Okolica, J.S. and G.L. Peterson, Windows driver memory analysis: A

reverse engineering methodology. Computers & Security, 2011. 30(8):

p. 770-779.

[19] Pressman, R.S., Software Engineering - A practitioner’s approach. 2006.

[20] Pressman, R.S., Software Engineering A Practitioner's Approach. Fifth

Edition ed. Computer Science Series, ed. M.-H.I. EDITION. 2001.

[21] Robertson, J., Forensic science, an enabler or dis-enabler for criminal

investigation? Australian Journal of Forensic Sciences, 2012. 44(1): p.

83-91.

[22] Sauter, V.L., Decision Support Systems for Business Intelligence, ed. W.

Sons. 2010.

[23] Sommerville, I., Software Engineering. 8th Edition ed. 2007: Pearson

Education Limited.

[24] Taroni, F., et al., Bayesian Networks and Probabilistic Inference in

Forensic Science, ed. L. John Wiley & Sons. 2006.


[25] Taroni, F., S. Bozza, and C. Aitken, Decision Analysis in Forensic Science

Journal of Forensic Science, 2005. 50(4).

[26] Taroni, F., S. Bozza, and A. Biedermann, Two items of evidence, no

putative source: an inference problem in forensic intelligence. Journal of

Forensic Sciences 2006. 51: p. 1350-1361.

[27] Taroni, F., et al., Data Analysis in Forensic Science: A Bayesan Decision

Perspective. 2010: John Wiley & Sons, Chichester.

[28] Tremblay, M.C., A.R. Hevner, and D.J. Berndt, Design of an information

volatility measure for health care decision making. 2012.

[29] Turban, E. and J.E. Aronson, Decision Support Systems and Intelligent

Systems. 2005.

[30] Vescoukis, V., N. Doulamis, and S. Karagiorgou, A service oriented

architecture for decision support systems in environmental crisis

management. Future Generation Computer Systems, 2012. 28(3): p.

593-604.

[31] Zarate, P., DMSS computer systems and application software.

International Journal of Decision Support System Technology (IJDSST),

2012. 3(4).

Received: May, 2012

reverse engineering approach in a development of a ...€¦ · reverse engineering approach in a...

Documents