MARC details
| 000 -LEADER |
|---|
| fixed length control field |
04399ntm a2200313 i 4500 |
| 003 - CONTROL NUMBER IDENTIFIER |
|---|
| control field |
MY-KuUP |
| 005 - DATE AND TIME OF LATEST TRANSACTION |
|---|
| control field |
20260407154021.0 |
| 006 - FIXED-LENGTH DATA ELEMENTS--ADDITIONAL MATERIAL CHARACTERISTICS |
|---|
| fixed length control field |
t||||fr|||| 000 0 |
| 007 - PHYSICAL DESCRIPTION FIXED FIELD--GENERAL INFORMATION |
|---|
| fixed length control field |
ta |
| 008 - FIXED-LENGTH DATA ELEMENTS--GENERAL INFORMATION |
|---|
| fixed length control field |
260407t20252025my a|||fr|||| 000 0 eng d |
| 020 ## - INTERNATIONAL STANDARD BOOK NUMBER |
|---|
| International Standard Book Number |
THE0010446 (Local) |
| Qualifying information |
Hardback |
| 040 ## - CATALOGING SOURCE |
|---|
| Original cataloging agency |
UMPSA |
| Language of cataloging |
eng |
| Transcribing agency |
UMPSA |
| Description conventions |
rda |
| 090 ## - LOCALLY ASSIGNED LC-TYPE CALL NUMBER (OCLC); LOCAL CALL NUMBER (RLIN) |
|---|
| Classification number (OCLC) (R) ; Classification number, CALL (RLIN) (NR) |
FKOM .K43 2025 r Thesis |
| 245 10 - TITLE STATEMENT |
|---|
| Title |
CAGDEEP : |
| Remainder of title |
mobile malware static analysis using force atlas 2 with strong gravity call graph and deep learning / |
| Statement of responsibility, etc. |
Nur Khairani Kamarudin |
| 264 #1 - PRODUCTION, PUBLICATION, DISTRIBUTION, MANUFACTURE, AND COPYRIGHT NOTICE |
|---|
| Place of production, publication, distribution, manufacture |
Kuantan, Pahang : _ |
| Name of producer, publisher, distributor, manufacturer |
UMPSA |
| Date of production, publication, distribution, manufacture, or copyright notice |
2025 |
| 264 #4 - PRODUCTION, PUBLICATION, DISTRIBUTION, MANUFACTURE, AND COPYRIGHT NOTICE |
|---|
| Date of production, publication, distribution, manufacture, or copyright notice |
© 2025 |
| 300 ## - PHYSICAL DESCRIPTION |
|---|
| Extent |
xviii, 237 pages : |
| Other physical details |
illustration (some colour) ; |
| Dimensions |
30 cm. + |
| Accompanying material |
1 CD-ROM. |
| 336 ## - CONTENT TYPE |
|---|
| Source |
rdacontent |
| Content type term |
text |
| 337 ## - MEDIA TYPE |
|---|
| Source |
rdamedia |
| Media type term |
unmediated |
| 338 ## - CARRIER TYPE |
|---|
| Source |
rdacarrier |
| Carrier type term |
volume |
| 347 ## - DIGITAL FILE CHARACTERISTICS |
|---|
| Source |
rda |
| File type |
text file |
| Encoding format |
PDF |
| 500 ## - GENERAL NOTE |
|---|
| General note |
Faculty of Computing |
| 502 ## - DISSERTATION NOTE |
|---|
| Dissertation note |
Thesis (Doctor of Philosophy) -- Universiti Malaysia Pahang Al Sultan Abdullah - 2025 |
| 504 ## - BIBLIOGRAPHY, ETC. NOTE |
|---|
| Bibliography, etc. note |
Include bibliographical reference |
| 520 3# - SUMMARY, ETC. |
|---|
| Summary, etc. |
The increasing number of Android users, mobile malware has seen an explosive growth phase, brought along great security threats and highlighting the need for more effective malware detection techniques. Mobile malware poses serious security risks such as data theft, financial fraud, unauthorized surveillance, and disruption of device functionality, which provide researchers with motivation and practical approaches for developing solutions. Existing static analysis methods, largely implemented through syntactic feature extractions manually, are proving to be insufficient as they typically neglect to model the key semantic and behavioural properties of malware. The behavioural properties of malware include patterns of API usage, call sequences, and inter-procedural interactions that reflect how malicious code operates beyond its shallow syntax. From these behavioural features, critical elements and attributes can be extracted, which later allow researchers to measure malware’s efficiency and effectiveness in evading detection. While graph-based approaches attempt to model program behaviour, they often convert complex structural data into numerical values, losing critical contextual information regarding the relationships between function calls and execution patterns. Moreover, they predominantly focus on extracting specific categories of syntactic features, often neglecting the deeper semantic and behavioural properties of malware. This selective approach restricts the exploration of a wider range of features, hindering the ability to comprehensively and accurately model malicious behaviours. CAGDeep is a study designed to overcome these challenges by examining the full range of semantic features embedded in malware samples using behavioural static analysis. It uses behavioural static analysis to capture the pattern and then model the wide range of semantic information embedded in malware samples. This approach leverages Force Atlas 2 Call Graphs with strong gravity parameters, which provide highly detailed visual representations of application behaviour. These graphs illustrate the intricate interactions within a program's call hierarchy, capturing the dynamic and complex relationships between its components. The call graphs are transformed into greyscale node-to-node representations, allowing convolutional neural networks (CNNs) to effectively learn and identify intricate spatial and structural relationships present in the graph. By employing this innovative methodology, CAGDeep ensures that even subtle and hidden patterns of malicious activity are systematically captured with high precision. In our empirical comparison of leading Android malware detection methods, CAGDeep was evaluated using the publicly available Android Malware Dataset (AMD), Malgenome, Drebin and AndroZoo dataset, comprising more than 20,000 malware and benign samples. CAGDeep achieved a classification accuracy of 99% which is higher than state-of-the-art models yielding the industry benchmark. The high accuracy proves that the CAGDeep method can expose more information regarding malicious behaviours. This may have significant implications for the advancement of strategies to improve malware detection capabilities and cyber-security measures. |
| 610 20 - SUBJECT ADDED ENTRY--CORPORATE NAME |
|---|
| Corporate name or jurisdiction name as entry element |
Faculty of Computing |
| General subdivision |
Dissertations |
| 650 #0 - SUBJECT ADDED ENTRY--TOPICAL TERM |
|---|
| Topical term or geographic name entry element |
Universities and colleges |
| General subdivision |
Dissertations |
| 942 ## - ADDED ENTRY ELEMENTS (KOHA) |
|---|
| Source of classification or shelving scheme |
Library of Congress Classification |
| Koha item type |
Thesis |