0% found this document useful (0 votes)
18 views10 pages

Document Verification Using Quick Response Code With Modified

Uploaded by

oyedejihiqmat
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
18 views10 pages

Document Verification Using Quick Response Code With Modified

Uploaded by

oyedejihiqmat
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 10

Indonesian Journal of Electrical Engineering and Computer Science

Vol. 28, No. 1, October 2022, pp. 470~479


ISSN: 2502-4752, DOI: 10.11591/ijeecs.v28.i1.pp470-479  470

Document verification using quick response code with modified


secure hash algorithm-1 and modified blowfish algorithm

Rogel Ladia Quilala1, Theda Flare G. Quilala2


1
Master in Information Technology Department, College of Computer Studies, Tarlac State University, Tarlac City, Philippines
2
Information Technology Department, College of Computer Studies, Tarlac State University, Tarlac City, Philippines

Article Info ABSTRACT


Article history: A previous study has been conducted integrating modification on secure
hash algorithm 1 (SHA-1) to document integrity verification of printed
Received Mar 25, 2022 documents using quick response (QR) codes. However, encryption is
Revised Jul 5, 2022 warranted as data is transmitted in plaintext directly to the server to prevent
Accepted Jul 27, 2022 hacking and ensuring not only data integrity but data security as well. A
more secured document integrity verification using QR code was designed
and developed by successfully incorporating a better hashing algorithm–
Keywords: modified SHA-1 and integrating a modern encryption algorithm–modified
blowfish algorithm. By integrating both, data integrity and data
Blowfish algorithm confidentiality is assured as compared to previous research. The developed
Data integrity software was checked against user requirements to check the acceptability of
Document verification the software. Error rate and accuracy were also checked to see how the
Encryption software performed. Based on the testing conducted, it has been found that
SHA-1 the document integrity verification software using QR code with the
integration of modified SHA-1 hash and Modified Blowfish encryption was
acceptable, accurate, and more secure.
This is an open access article under the CC BY-SA license.

Corresponding Author:
Theda Flare G. Quilala
Information Technology Department, College of Computer Studies, Tarlac State University
Romulo Blvd. San Vicente, Tarlac City 2300, Philippines
Email: [email protected]

1. INTRODUCTION
Competency, skills, and qualifications for a certain craft are demonstrated by acquiring
certifications. These documents act as your credentials and are considered by employers when hiring for
personnel. But recently, people get involved in the creation of fraudulent academic credentials therefore its
prevalence [1]. Verification of these documents is quite challenging because they cannot be traced as quickly
and accurately [2]. Thus, a system is needed to inspect the authenticity and validity of such document
certification rapidly and precisely [3].
Due to the rise in the usage of smartphones with cameras combined with the ease of scanning a
quick response (QR) code using this device, studies have explored the use of QR codes as a cheap alternative
to other tag-based systems [4]. QR codes have been used for authentication on printed documents for fraud
identification. One study tried embedding watermark objects with QR codes to determine printed document
validity but challenges were identified such as the preparation of validation links, watermark image (logo)
configuration, and size restriction [5], [6] making the embedding of an image in QR unattractive. Others have
embedded a blockchain technology pattern into a QR code for authentication [7] and encrypted lossless
compression [8], but failed to incorporate hashing in their verification thus neglecting to consider the data
integrity of the data being transmitted. One study analyzes the impact of blockchain on academic certificates

Journal homepage: http://ijeecs.iaescore.com


Indonesian J Elec Eng & Comp Sci ISSN: 2502-4752  471

and has seen its advantage [9], [10] but on the other hand, others find it complicated, challenging, and costly
therefore recommended to be better suited for financial and other sectors where mining is present [11].
Several studies successfully created verification apps using QR. One constructed a mobile app for
securing issued degree documents which encrypts students’ information from the database and saved it on a
server, afterward creating a QR code to be printed on the document which is then used for validation
purposes [12]. Another study combines QR codes, digital signatures, and hashing, in a smartphone
application [13]. On the other hand, one study was able to integrate secured communication via transport
layer security and hashing [14]. Another also developed an android application that used QR codes in the
identification of objects along with hash [15]. However, for all studies the QR code can only be read using
the created mobile application which is a hassle because it needs to be installed separately.
Hashing has been proven to ensure data integrity, so its wide usage has been evident. In one study, a
hash function was used to create an efficient way to secure the personally identifiable information (PII) of a
user in a QR code which is good because data integrity is assured but the disadvantage is that secure hash
algorithm 1 (SHA-1) was used as the hashing algorithm and is already known to be weak [16]. Blockchain
was successfully integrated with QR in one more study but failed to add additional security on transit [17].
Previous research applied QR code technology in verifying the authenticity of documents using a web
application that doesn’t require additional installation on the part of the user [18] but failed to consider data
security by transmitting data in plaintext and thus is prone to hacking. To address this data security weakness,
other studies make use of encryption schemes for security [19]–[21] against hacking.
The contribution of this paper is the improvement in the design and development of a document
integrity verification using QR code by incorporating modified SHA-1, a better hashing algorithm than the
weak SHA-1 to emphasize data integrity, and inclusion of modified blowfish encryption algorithm, a modern
algorithm to encrypt the confidential data embedded as QR code in the document as the security measure.
Both data integrity and data confidentiality are assured by incorporating both hashing and encryption. The
software does not need to be installed separately unlike in previous studies and can be used on both Android
and Apple smartphones to verify certificates. The app will make use of the phone's embedded camera. The
specific objectives of this study are to: i) develop a more secured document integrity verification software
using QR code with modified SHA-1 for hashing and integration of modified blowfish algorithm for
encryption, ii) evaluate the acceptability of the developed software via use case tests, and iii) analyze the
performance of the system in verifying documents by computing the error rate during the alpha testing.

2. METHOD
2.1. Research design
This study utilizes the design and development approach. The software was created in Visual Studio
Community Edition in an Intel(R) Core (TM) i5-10210U processor with CPU @1.60GHz 2.11 GHz and 8.00
GB RAM running Windows 10 Pro. The web server must run on internet information services (IIS) version
10.0.19401.1. The developers applied the rapid application development (RAD) software development
methodology. The process flow of the study is shown in Figure 1. The document integrity verification is
subdivided into three sections: hashing, encryption and generation of QR code, certificate management, and
certificate verification.

Figure 1. The process flow of the document integrity verification

Document verification using quick response code with modified secure … (Rogel Ladia Quilala)
472  ISSN: 2502-4752

In the hashing, encryption, and generation of QR Code section, first, the system administrator inputs
the unique certificate identification (ID) of the document and message M which is equivalent to the name to
which the document belongs. Then, the modified SHA-1 was applied to ID and message M to produce 192
bits hash value which is then saved to the database. Then, the modified blowfish algorithm (MBA)
encryption will encrypt the hash and message M. After that, the QR code generator generates the QR code
from the encrypted hash value. The QR code generated was to be printed on the documents.
In the certificate management section, the list of names, created QR codes, and images for the
design of the document that will act as the certificate template, must be provided before the printing of the
certificate commences. The documents were set using letter-size (8.5”×11”) paper. The arrangement such as
the draft of the background, location of the name, and QR code position in the certificate template is done on
the developed windows application. Once done, the QR code is now embedded in the certificate template and
can now be printed based on the list of names provided.
In the certificate verification section, after printing, anybody who wants to verify the document must
use their smartphone with a camera to scan the QR. The operating system of the phone does not matter, it
only requires a camera. However, the user’s smartphone wireless fidelity (Wi-Fi) should be connected to the
same network as that of the server. After capturing the QR code using the smartphone camera, the verify
button needs to be clicked to let the web application send the encrypted hash value to the web server for data
integrity verification. The server receives the information stored in the QR code, decrypts the code, and then
it will search if the hash value exists in the database. If the hash value is found, the system retrieves the
unique certificate ID of the document from the database. The system verifies if the generated encrypted hash
value is the same as that saved on the server. If the values are equal, that means it is successfully verified,
therefore the message is said to be authentic. If the hash does not exist in the server, the QR code message M
was modified. The verification system prompts the message fetch from the server for visual assessment and
compared the information from the printed document.

2.2. Use case testing


The study adopted alpha testing and will make use of sample documents (certifications) as the
source of input. It will take note of message M (name on the document) and document ID to generate the
hash value using modified SHA-1. Next, the generated hash will be encrypted using the modified blowfish
algorithm, sent to the server, and decrypted when necessary.
The software is to be tested to validate as per compliance with customer requirements and will make
use of the blackbox testing technique specifically the use case testing type [22]. This testing was selected to
check if all parts of the system are working as intended and are acceptable to the user for quicker test case
development, even without the knowledge or access to the code. The target of evaluation is categorized into
login, hashing, encryption and decryption, QR code generation, certificate management, and certificate
verification as reflected in Table 1. The test case objectives were also stated as well as the equivalent
functional requirement. These test cases will be checked against their expected result and will be marked
passed or failed depending on the behavior of the software.

Table 1. Test cases


Test case Test case objective Functional requirement
Login Module Test the login functionality with different Users shall be able to login into the system using the
sets of data correct username and password credentials
Hashing Test the hashing module using different The system shall be able to generate a hash value using
Module certificate ID and message (name) the Modified SHA-1 algorithm out of the certificate ID
and message (name).
Encryption Test if a series of hash values can be The system shall be able to encrypt the generated hash
and encrypted and decrypted without errors value using the Modified Blowfish algorithm before
Decryption transmitting and decrypting upon saving on the server
Module
QR Code Test if QR codes are correctly generated The system must be able to generate the QR code from
Generation from sets of encrypted hash values the encrypted hash value
Module
Certificate Test that QR codes can be placed on The user must be able to embed the QR code in the
Management certificate templates and can be printed certificate template.
Module The user must be able to print the certificates with the
attached QR Code
Certificate Test if printed certificates display the The user must be able to verify the authenticity of the
Verification correct message (name) on printed certificate using the QR code
Module certificates

Indonesian J Elec Eng & Comp Sci, Vol. 28, No. 1, October 2022: 470-479
Indonesian J Elec Eng & Comp Sci ISSN: 2502-4752  473

2.3. Performance analysis


After the test case evaluation of the use cases, the error rate will be analyzed and tested using thirty
(30) documents with correct entries. Names were generated using a name test data generator tool [23].
The formula for error rate and accuracy will be computed as (1)-(2).

|𝑜𝑏𝑠𝑒𝑟𝑣𝑒𝑑 𝑣𝑎𝑙𝑢𝑒−𝑎𝑐𝑡𝑢𝑎𝑙 𝑣𝑎𝑙𝑢𝑒|


𝐸𝑟𝑟𝑜𝑟 𝑟𝑎𝑡𝑒 = × 100 (1)
𝑎𝑐𝑡𝑢𝑎𝑙 𝑣𝑎𝑙𝑢𝑒

𝐴𝑐𝑐𝑢𝑟𝑎𝑐𝑦 = 100% − 𝐸𝑟𝑟𝑜𝑟 𝑟𝑎𝑡𝑒 (2)

3. RESULTS AND DISCUSSION


In this section, it is explained the results of research and at the same time is given
the comprehensive discussion. Results can be presented in figures, graphs, tables and others that make
the reader understand easily [24], [25]. The discussion can be made in several sub-sections.

3.1. Development of the document integrity verification using QR code with modified SHA-1 and
modified blowfish
Before the user can access the system, a login form is created. To verify the authenticity of the user,
only authorized username and password credentials are allowed. Figure 2 shows the login screen. Figure 3
demonstrates the creation of the hash value using the certificate ID and the message. The message to be
embedded in the QR code is the name of the student to which the certificate belongs. A modified SHA-1
algorithm [26] was applied to create the hash value. After the hash value is created, the hash will be
encrypted using modified blowfish algorithm [27] for added security during the transmission of data to the
server. This will ensure the safe passage of data that will prevent hacking. After encryption, the QR code will
now be generated as shown.

Figure 2. Login screen

Figure 3. Hashing, encryption, and QR code generation

Document verification using quick response code with modified secure … (Rogel Ladia Quilala)
474  ISSN: 2502-4752

Figure 4. Certificate template with encrypted QR code

Figure 4 shows a sample certificate template where the encrypted QR code was embedded. This
certificate may be printed and contains different QR codes from the Certificate ID and names of all attendees
of seminars or conferences. Figure 5 displays how the user will see the QR code when viewed on a cellphone
camera. The IP address of the server will be displayed and will be directed to the certificate verification
module.

Figure 5. QR code as seen on cellphone camera

Figure 6 illustrates a sample verification message where the name of the holder of the certificate is
displayed. This will then be counter-verified to the name listed on the printed certificate. If the names
matched, the certificate is said to be untampered and valid.

Indonesian J Elec Eng & Comp Sci, Vol. 28, No. 1, October 2022: 470-479
Indonesian J Elec Eng & Comp Sci ISSN: 2502-4752  475

Figure 6. Certificate verification

3.2. Use Case Testing


Table 2 displays the test case created for the login module. Four (4) test data were used, and a
sample screenshot was inserted to show the result of the test. The module behaves as expected therefore the
login module passed the use case test. Table 3 shows the test case created for the Hashing Module. Three (3)
sets of certificate ID and name combinations were used. Sample screenshots were attached to show the result
of the hashing.

Table 2. Login module test case


Test case objective Test data Expected result Actual Sample screenshot Remarks
result
Test the login 1. Valid username Users shall be able Same Passed
functionality with and password to login
different sets of data successfully

2. A valid Users should not Same Passed


username and be able to log in
invalid
Users should not Same Passed
password
be able to log in
3. Invalid
username and Users should not Same Passed
valid password be able to log in
4. Invalid
username and
invalid
password

Document verification using quick response code with modified secure … (Rogel Ladia Quilala)
476  ISSN: 2502-4752

Table 3. Hashing module test case


Test case Test data Expected result Actual Sample screenshot Remarks
objective result
Test the CCS032222-1 5c7a1db3bf Same Passed
hashing Lina Upton 3c20214469
module f04d4f43ca
using a40a11447df
different f742aa4
certificat
e ID and
message
(name)

CCS032222-2 f347b7822b Same Passed


Jose Davis 537f1c74f3
333be58877
10d6759342b
6f5b9b5

CCS032222-3 7d0b577323 Same Passed


Giovani Cole 382f701340
850a93466b
ae696e7729b
4f88920

Table 4 shows the test case created for the Encryption and Decryption Module, the QR Code
Generation, and the Certificate Verification test cases. The three (3) names and ID combinations used in the
hashing module were re-used here. From the encrypted hash, the QR code is generated and placed in the
certificate template. To check the process of decryption, the generated QR from the name was scanned and
the decrypted name was displayed. As shown in the table, the same name appears in the certificate and the
displayed decrypted hash except for test data four (4). Test data for sample four (4) was intentionally
modified to showcase where a copy-pasted QR code will still show the original hash value from the initial
certificate ID and name combination. This only signifies that modification or editing of a name without
updating the QR code will be detected and thus can be marked as a fraudulent certificate. Sample screenshots
were attached to check the comparison. For all input data, the output data is the same indicating that the test
case was a success. Table 5 displays the Certificate Management module for the three (3) sets of data. These
certificates can be printed after the placement of QR codes on the certificate template.

Indonesian J Elec Eng & Comp Sci, Vol. 28, No. 1, October 2022: 470-479
Indonesian J Elec Eng & Comp Sci ISSN: 2502-4752  477

Table 4. Encryption and decryption, QR code generation, and certificate verification module test case
Test Sample Screenshot
Data and
Test Case Actual
Expected Decrypted Name QR Code Remarks
Objective Result
Result

Test if a 1. Lina Same Passed


series of Upton
hash
values can
be
encrypted
and
decrypted
without
errors

Test if the 2. Jose Same Passed


QR code Davis
is
correctly
generated
from sets
of
encrypted
hash
values
Test if 3. Giovanni Same Passed
printed Cole
certificates
display the
correct
message
(name) on
printed
certificates
4. Test Same
Data is Passed
John
Doe, the
decrypte
d value
should
be Lina
Upton

Table 5. Certificate management module test case module


Test Data
Test Case and Actual
Sample Screenshot Remarks
Objective Expected Results
Result
Test that Certificates Same Passed
QR codes for Lina
can be Upton,
placed on Jose Davis,
certificate and
templates Giovanni
and can be Cole
printed

Document verification using quick response code with modified secure … (Rogel Ladia Quilala)
478  ISSN: 2502-4752

3.3. Error rate computation


The observed value here is the number of correct certificates which is 29 and the actual value refers
to the number of samples used, in this case, 30 certificates. Out of the 30 samples, one was found incorrect
leading to an error rate of 3.33%. Computing for accuracy, the document verification system yields a 96.67%
accuracy. The error was further investigated after a certificate does not exist prompted:
|29−30|
𝐸𝑟𝑟𝑜𝑟 𝑟𝑎𝑡𝑒 = × 100 % = 3.33 %
30

𝐴𝑐𝑐𝑢𝑟𝑎𝑐𝑦 = 100% − 3.33% = 96.67%

Content of the QRCODE that should be generated:


http://191.192.193.79/?dOFzzTED3yWTMI67YSH8rmX2u31KIfyaZib1B/od0wTIGoql7x2g9mtkhKwHOnS
nW3IcttxBDN/tt/YNMnyBTw==,Devyn Gleichner

Content of the QRCODE after generation:


http://191.192.193.79/?dOFzzTED3yWTMI67YSH8rmX2u31KIfyaZib1B/od0wTIGoql7x2g9mtkhKwHOnS
nW3IcttxBD^/tt/YNMnyBTw==,Devyn Gleichner

The capital letter N has been changed to the caret symbol ^.

4. CONCLUSION
A more secured document integrity verification using QR code was designed and developed by
successfully incorporating a better hashing algorithm-modified SHA-1 and integrating a modern encryption
algorithm–modified blowfish algorithm. By integrating both, data integrity and data confidentiality is assured
as compared to previous research. The verification works even if no additional software is installed by using
the smartphone’s built-in camera making it better than other verification software. The developed software
has been proven to satisfy all user requirements and is deemed to be acceptable based on the expected and
actual results using the test data. The software is also deemed highly accurate with an error rate of 3.33%.
However, to achieve 100% accuracy, the QR code generator dll needs to be further investigated and tested to
much larger sample size. Exploration of the embedding of the simplest and most economical blockchain
technology for an educational institution may also be further studied.

ACKNOWLEDGEMENTS
The authors would like to express their sincerest thanks to Tarlac State University for their support
during the conduct of this research.

REFERENCES
[1] J. G. Dongre, “Education Degree Fraud Detection and Student Certificate Verification using Blockchain,” Int. J. Eng. Res., vol.
V9, no. 07, pp. 287–289, Jul. 2020, doi: 10.17577/IJERTV9IS070156.
[2] N. Malsa, V. Vyas, J. Gautam, R. N. Shaw, and A. Ghosh, “Framework and smart contract for blockchain enabled certificate
verification system using robotics,” Stud. Comput. Intell., vol. 960, no. July, pp. 125–138, 2021, doi: 10.1007/978-981-16-0598-
7_10.
[3] H. Indriyawati, T. Winarti, and V. Vydia, “Web-based document certification system with advanced encryption standard digital
signature,” Indones. J. Electr. Eng. Comput. Sci., vol. 22, no. 1, pp. 516–521, 2021, doi: 10.11591/ijeecs.v22.i1.pp516-521.
[4] M. Hasson, A. A. Yassin, A. J. Yassin, A. M. Rashid, A. A. Yaseen, and H. Alasadi, “Password authentication scheme based on
smart card and QR code,” Indones. J. Electr. Eng. Comput. Sci., vol. 23, no. 1, pp. 140–149, 2021, doi:
10.11591/ijeecs.v23.i1.pp140-149.
[5] C. M. Li, P. Hu, and W. C. Lau, “AuthPaper: Protecting paper-based documents and credentials using Authenticated 2D
barcodes,” IEEE Int. Conf. Commun., vol. 2015-Septe, no. January 2013, pp. 7400–7406, 2015, doi: 10.1109/ICC.2015.7249509.
[6] T. Mantoro, D. D. Permadi, and A. Abubakar, “Stegano-image as a digital signature to improve security authentication system in
mobile computing,” in 2016 International Conference on Informatics and Computing (ICIC), 2016, no. Icic, pp. 158–163, doi:
10.1109/IAC.2016.7905708.
[7] Q. Aini, U. Rahardja, M. R. Tangkaw, N. P. L. Santoso, and A. Khoirunisa, “Embedding a Blockchain Technology Pattern Into
the QR Code for an Authentication Certificate,” J. Online Inform., vol. 5, no. 2, p. 239, 2020, doi: 10.15575/join.v5i2.583.
[8] A. M. Ali and A. K. Farhan, “Enhancement of QR Code Capacity by Encrypted Lossless Compression Technology for
Verification of Secure E-Document,” IEEE Access, vol. 8, pp. 27448–27458, 2020, doi: 10.1109/ACCESS.2020.2971779.
[9] K. Kumutha and S. Jayalakshmi, “The Impact of the Blockchain on Academic Certificate Verification System-Review,” EAI
Endorsed Trans. Energy Web, vol. 8, no. 36, pp. 1–8, 2021, doi: 10.4108/eai.29-4-2021.169426.
[10] K. Dhyani, J. Mishra, S. Paladhi, and I. S. Thaseen, “A Blockchain-Based Document Verification System for Employers,” in
Algorithms for Intelligent Systems, 2022, no. March, pp. 123–137, doi: 10.1007/978-981-16-7182-1_11.

Indonesian J Elec Eng & Comp Sci, Vol. 28, No. 1, October 2022: 470-479
Indonesian J Elec Eng & Comp Sci ISSN: 2502-4752  479

[11] C. Shaik, “Preventing forged and fabricated academic credentials using cryptography,” Int. J. Comput. Sci. Eng. Appl., vol. 11,
no. 1, pp. 11–20, 2021, doi: 10.5121/ijcsea.2021.11102.
[12] Z. Yahya et al., “A New Academic Certificate Authentication Using Leading Edge Technology,” in Proceedings of the 2017
International Conference on E-commerce, E-Business and E-Government - ICEEG 2017, 2017, pp. 82–85, doi:
10.1145/3108421.3108428.
[13] A. Singhal and R. S. Pavithr, “Degree Certificate Authentication using QR Code and Smartphone,” Int. J. Comput. Appl., vol.
120, no. 16, pp. 38–43, Jun. 2015, doi: 10.5120/21315-4303.
[14] A. Wibiyanto and I. Afrianto, “QR code and transport layer security for licensing documents verification,” IOP Conf. Ser. Mater.
Sci. Eng., vol. 407, no. 1, 2018, doi: 10.1088/1757-899X/407/1/012069.
[15] D. Jagodic, D. Vujicic, and S. Randic, “Android system for identification of objects based on QR code,” 2015 23rd Telecommun.
Forum, TELFOR 2015, vol. 7, pp. 922–925, 2016, doi: 10.1109/TELFOR.2015.7377616.
[16] G. K. Hong and S. Sinha, “Tracking Vulnerable People Using Body Worn QR Code,” 2018.
[17] M. U. Abdullahi, G. I. O. Aimufua, and A. Aminu, “Certificate Generation and Verification System Using Blockchain
Technology and Quick Response Code,” IOSR J. Comput. Eng. ·, vol. 24, no. May, pp. 37–47, 2022, doi: 10.9790/0661-
2401023747.
[18] R. L. Quilala, A. M. Sison, and R. P. Medina, “QR Code Integrity Verification Based on Modified SHA-1 Algorithm,” Indones. J.
Electr. Eng. Informatics, vol. 6, no. 4, pp. 385–392, 2018, doi: 10.11591/ijeei.v6i1.494.
[19] D. Agnihotri, S. Ahmed, D. Darekar, C. Gadkari, S. Jaikar, and M. Pawar, “A Secure Document Archive Implemented using
Multiple Encryption,” Proc. - Int. Conf. Smart Electron. Commun. ICOSEC 2020, no. Icosec, pp. 765–770, 2020, doi:
10.1109/ICOSEC49089.2020.9215302.
[20] F. Ahmad and L.-M. Cheng, “Paper Document Authentication Using Print-Scan Resistant Image Hashing and Public-Key
Cryptography,” in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture
Notes in Bioinformatics), 2019, vol. 11611 LNCS, pp. 157–165, doi: 10.1007/978-3-030-24907-6_13.
[21] M. S. Ahamed and H. A. Mustafa, “A Secure QR Code System for Sharing Personal Confidential Information,” 5th Int. Conf.
Comput. Commun. Chem. Mater. Electron. Eng. IC4ME2 2019, no. November, 2019, doi: 10.1109/IC4ME247184.2019.9036521.
[22] M. A. Umar, “Comprehensive study of software testing: Categories , levels , techniques , and types,” Int. J. Adv. Res. Ideas Innov.
Technol., vol. 5, no. 6, pp. 32–40, 2019, doi: 10.36227/techrxiv.12578714.
[23] Coders Tool, “Name Test Data Generator Tool,” 2022. [Online]. Available: https://www.coderstool.com/fake-test-identity-
data?fbclid=IwAR1ppNxa8iIKq5Hx6pygbl_2vzndCvkrZjEELSuvPAtz4nx76wKS3vi7M. [Accessed: 03-Mar.-2022].
[24] J. Sadowski, “When data is capital: Datafication, accumulation, and extraction,” Big Data Soc., vol. 6, no. 1, pp. 1–12, 2019, doi:
10.1177/2053951718820549.
[25] J. R. Saura, B. R. Herraez, and A. R. Menendez, “Comparing a traditional approach for financial brand communication analysis
with a big data analytics technique,” IEEE Access, vol. 7, pp. 37100–37108, 2019, doi: 10.1109/ACCESS.2019.2905301.
[26] R. L. Quilala, A. M. Sison, and R. P. Medina, “Modified SHA-1 Algorithm,” vol. 11, no. 3, pp. 1027–1034, 2018, doi:
10.11591/ijeecs.v11.i3.pp1027-1034.
[27] T. F. G. Quilala, A. M. Sison, and R. P. Medina, “Modified Blowfish Algorithm,” vol. 12, no. 1, pp. 38–45, 2018, doi:
10.11591/ijeecs.v12.i1.pp38-45.

BIOGRAPHIES OF AUTHORS

Dr. Rogel Ladia Quilala is an Assistant Professor of ICT at the Tarlac State
University-College of Computer Studies Tarlac City Philippines. He received his Doctor in
Information Technology (DIT) degree from Technological Institute of the Philippines (TIP),
Cubao Quezon City Philippines. His research areas are security and data mining. He is the
Chairperson of the Master in Information Technology Department. He has 18 years of
academic experience teaching ICT courses at the tertiary level. During this time, he had a stint
as an exchange professor in IT at YeungJin College in South Korea. He can be contacted at
email: [email protected].

Dr. Theda Flare G. Quilala is currently an Associate Professor in the College of


Computer Studies at Tarlac State University, Tarlac City, Philippines. A Doctor of
Information Technology graduate at Technological Institute of the Philippines and a CHED
K-12 scholar. Research interest includes security, data mining, and algorithms. She is
currently the Director of the Admission and Registration Office. She can be reached via email
at [email protected].

Document verification using quick response code with modified secure … (Rogel Ladia Quilala)

You might also like