Experimental Research Data Quality in Materials Science

Wuest, Thorsten; Tinscher, R.; Porzel, Robert; Thoben, Klaus-Dieter

doi:10.5121/ijait.2014.4601

Cited by 6 publications

(2 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The silhouette values for the classification using Genetic -K-means and normal K-means are listed and presented in figure 2,3,4 and 5. The silhouette values [23]is an object consisting of sample data, clustering data, and silhouette criterion values used to evaluate the optimal number of data clusters. Figure 2 and figure 3 show the values for both normal K-means and genetic K-means for Opcode data; the number of clusters is small in genetic Kmeans while it is bigger in normal K-means.…”

Section: Resultsmentioning

confidence: 99%

Automated Malware Detection Based on HMM and Genetic K-means

Alqurashi¹,

Batarfi²

2018

IJICR

View full text Add to dashboard Cite

With the increased use of the Internet and application-based networks, malware detection is a serious challenge. The signature-based detection technique has been widely used as the main method of detecting malware, but with obfuscation techniques, it has failed to detect modern malware. Recent research has proven that a Hidden Markov model (HMM) is useful for malware detection using features that reflect the malware behavior. The motivation in this work is to enhance the working strategy of malware detection. In this study, the related problem of malware clustering based on HMM is considered. In meeting this goal, this study has proposed a system of testing the malware behavior based on HMM scores, which have been extracted from the learning model on application programming interface (API) call sequences and Operational code (opcode) sequence datasets as malware behavior. API call sequences that extract dynamically and opcode sequences that extract statically are used; they are compared to see which behavior is better for malware detection. Then genetic operators are used in enhancing normal K-means working with the HMM. The proposed genetic K-means is used as a classification algorithm to cluster new behaviors based on the scoring from the HMM. Next, the enhancement results are compared to the normal Kmeans classification based on the HMM, evaluating the proposed optimization technique. Therefore, this study is considered to be an optimization enhancement and an evaluation study among normal K-means with the HMM in malware detection, proposing genetic K-means with HMM in malware detection. The results obtained from the experiments demonstrate that the objectives are successfully completed with an average of high detection rate about 87.68%.

show abstract

Section: Resultsmentioning

confidence: 99%

Automated Malware Detection Based on HMM and Genetic K-means

Alqurashi¹,

Batarfi²

2018

IJICR

View full text Add to dashboard Cite

show abstract

Section: Resultsmentioning

confidence: 99%

Untitled

2018

IJICR

View full text Add to dashboard Cite

With the increased use of the Internet and application-based networks, malware detection is a serious challenge. The signature-based detection technique has been widely used as the main method of detecting malware, but with obfuscation techniques, it has failed to detect modern malware. Recent research has proven that a Hidden Markov model (HMM) is useful for malware detection using features that reflect the malware behavior. The motivation in this work is to enhance the working strategy of malware detection. In this study, the related problem of malware clustering based on HMM is considered.In meeting this goal, this study has proposed a system of testing the malware behavior based on HMM scores, which have been extracted from the learning model on application programming interface (API) call sequences and Operational code (opcode) sequence datasets as malware behavior. API call sequences that extract dynamically and opcode sequences that extract statically are used; they are compared to see which behavior is better for malware detection. Then genetic operators are used in enhancing normal K-means working with the HMM. The proposed genetic K-means is used as a classification algorithm to cluster new behaviors based on the scoring from the HMM. Next, the enhancement results are compared to the normal Kmeans classification based on the HMM, evaluating the proposed optimization technique.Therefore, this study is considered to be an optimization enhancement and an evaluation study among normal K-means with the HMM in malware detection, proposing genetic K-means with HMM in malware detection. The results obtained from the experiments demonstrate that the objectives are successfully completed with an average of high detection rate about 87.68%.

show abstract