An Effective Feature Selection Scheme for Android ICC-Based Malware Detection Using the Gap of the Appearance Ratio

Kyohei OSUGE; Hiroya KATO; Shuichiro HARUTA; Iwao SASASE

doi:10.1587/transinf.2018EDP7301

An Effective Feature Selection Scheme for Android ICC-Based Malware Detection Using the Gap of the Appearance Ratio

Kyohei OSUGE, Hiroya KATO, Shuichiro HARUTA, Iwao SASASE

Full Text Views

0

Cite this

Summary :

Android malwares are rapidly becoming a potential threat to users. Among several Android malware detection schemes, the scheme using Inter-Component Communication (ICC) is gathering attention. That scheme extracts numerous ICC-related features to detect malwares by machine learning. In order to mitigate the degradation of detection performance caused by redundant features, Correlation-based Feature Selection (CFS) is applied to feature before machine learning. CFS selects useful features for detection in accordance with the theory that a good feature subset has little correlation with mutual features. However, CFS may remove useful ICC-related features because of strong correlation between them. In this paper, we propose an effective feature selection scheme for Android ICC-based malware detection using the gap of the appearance ratio. We argue that the features frequently appearing in either benign apps or malwares are useful for malware detection, even if they are strongly correlated with each other. To select useful features based on our argument, we introduce the proportion of the appearance ratio of a feature between benign apps and malwares. Since the proportion can represent whether a feature frequently appears in either benign apps or malwares, this metric is useful for feature selection based on our argument. Unfortunately, the proportion is ineffective when a feature appears only once in all apps. Thus, we also introduce the difference of the appearance ratio of a feature between benign apps and malwares. Since the difference simply represents the gap of the appearance ratio, we can select useful features by using this metric when such a situation occurs. By computer simulation with real dataset, we demonstrate our scheme improves detection accuracy by selecting the useful features discarded in the previous scheme.

Publication: IEICE TRANSACTIONS on Information Vol.E102-D No.6 pp.1136-1144

Publication Date: 2019/06/01

Publicized: 2019/03/12

Online ISSN: 1745-1361

DOI: 10.1587/transinf.2018EDP7301

Type of Manuscript: PAPER

Category: Dependable Computing

Authors

Kyohei OSUGE
  Keio University
Hiroya KATO
  Keio University
Shuichiro HARUTA
  Keio University
Iwao SASASE
  Keio University

Keyword

android, malware detection, ICC, feature selection

Cite this

Copy

Kyohei OSUGE, Hiroya KATO, Shuichiro HARUTA, Iwao SASASE, "An Effective Feature Selection Scheme for Android ICC-Based Malware Detection Using the Gap of the Appearance Ratio" in IEICE TRANSACTIONS on Information, vol. E102-D, no. 6, pp. 1136-1144, June 2019, doi: 10.1587/transinf.2018EDP7301.
Abstract: Android malwares are rapidly becoming a potential threat to users. Among several Android malware detection schemes, the scheme using Inter-Component Communication (ICC) is gathering attention. That scheme extracts numerous ICC-related features to detect malwares by machine learning. In order to mitigate the degradation of detection performance caused by redundant features, Correlation-based Feature Selection (CFS) is applied to feature before machine learning. CFS selects useful features for detection in accordance with the theory that a good feature subset has little correlation with mutual features. However, CFS may remove useful ICC-related features because of strong correlation between them. In this paper, we propose an effective feature selection scheme for Android ICC-based malware detection using the gap of the appearance ratio. We argue that the features frequently appearing in either benign apps or malwares are useful for malware detection, even if they are strongly correlated with each other. To select useful features based on our argument, we introduce the proportion of the appearance ratio of a feature between benign apps and malwares. Since the proportion can represent whether a feature frequently appears in either benign apps or malwares, this metric is useful for feature selection based on our argument. Unfortunately, the proportion is ineffective when a feature appears only once in all apps. Thus, we also introduce the difference of the appearance ratio of a feature between benign apps and malwares. Since the difference simply represents the gap of the appearance ratio, we can select useful features by using this metric when such a situation occurs. By computer simulation with real dataset, we demonstrate our scheme improves detection accuracy by selecting the useful features discarded in the previous scheme.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2018EDP7301/_p

Copy

@ARTICLE{e102-d_6_1136,
author={Kyohei OSUGE, Hiroya KATO, Shuichiro HARUTA, Iwao SASASE, },
journal={IEICE TRANSACTIONS on Information},
title={An Effective Feature Selection Scheme for Android ICC-Based Malware Detection Using the Gap of the Appearance Ratio},
year={2019},
volume={E102-D},
number={6},
pages={1136-1144},
abstract={Android malwares are rapidly becoming a potential threat to users. Among several Android malware detection schemes, the scheme using Inter-Component Communication (ICC) is gathering attention. That scheme extracts numerous ICC-related features to detect malwares by machine learning. In order to mitigate the degradation of detection performance caused by redundant features, Correlation-based Feature Selection (CFS) is applied to feature before machine learning. CFS selects useful features for detection in accordance with the theory that a good feature subset has little correlation with mutual features. However, CFS may remove useful ICC-related features because of strong correlation between them. In this paper, we propose an effective feature selection scheme for Android ICC-based malware detection using the gap of the appearance ratio. We argue that the features frequently appearing in either benign apps or malwares are useful for malware detection, even if they are strongly correlated with each other. To select useful features based on our argument, we introduce the proportion of the appearance ratio of a feature between benign apps and malwares. Since the proportion can represent whether a feature frequently appears in either benign apps or malwares, this metric is useful for feature selection based on our argument. Unfortunately, the proportion is ineffective when a feature appears only once in all apps. Thus, we also introduce the difference of the appearance ratio of a feature between benign apps and malwares. Since the difference simply represents the gap of the appearance ratio, we can select useful features by using this metric when such a situation occurs. By computer simulation with real dataset, we demonstrate our scheme improves detection accuracy by selecting the useful features discarded in the previous scheme.},
keywords={},
doi={10.1587/transinf.2018EDP7301},
ISSN={1745-1361},
month={June},}

Copy

TY - JOUR
TI - An Effective Feature Selection Scheme for Android ICC-Based Malware Detection Using the Gap of the Appearance Ratio
T2 - IEICE TRANSACTIONS on Information
SP - 1136
EP - 1144
AU - Kyohei OSUGE
AU - Hiroya KATO
AU - Shuichiro HARUTA
AU - Iwao SASASE
PY - 2019
DO - 10.1587/transinf.2018EDP7301
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E102-D
IS - 6
JA - IEICE TRANSACTIONS on Information
Y1 - June 2019
AB - Android malwares are rapidly becoming a potential threat to users. Among several Android malware detection schemes, the scheme using Inter-Component Communication (ICC) is gathering attention. That scheme extracts numerous ICC-related features to detect malwares by machine learning. In order to mitigate the degradation of detection performance caused by redundant features, Correlation-based Feature Selection (CFS) is applied to feature before machine learning. CFS selects useful features for detection in accordance with the theory that a good feature subset has little correlation with mutual features. However, CFS may remove useful ICC-related features because of strong correlation between them. In this paper, we propose an effective feature selection scheme for Android ICC-based malware detection using the gap of the appearance ratio. We argue that the features frequently appearing in either benign apps or malwares are useful for malware detection, even if they are strongly correlated with each other. To select useful features based on our argument, we introduce the proportion of the appearance ratio of a feature between benign apps and malwares. Since the proportion can represent whether a feature frequently appears in either benign apps or malwares, this metric is useful for feature selection based on our argument. Unfortunately, the proportion is ineffective when a feature appears only once in all apps. Thus, we also introduce the difference of the appearance ratio of a feature between benign apps and malwares. Since the difference simply represents the gap of the appearance ratio, we can select useful features by using this metric when such a situation occurs. By computer simulation with real dataset, we demonstrate our scheme improves detection accuracy by selecting the useful features discarded in the previous scheme.
ER -