Intrusion Detection Method Based on MapReduce for Evolutionary Feature Selection in Mobile Cloud Computing

Emmanuel Mugabo ； Qiu-Yu Zhang ； Aristide Ngaboyindekwe ； Vincent de Paul Niyigena Kwizera ； Victus Elikplim Lumorvie

《International Journal of Network Security》 23卷1期 (2021/01) Pp. 106-115

https://doi.org/10.6633/IJNS.202101_23(1).13

摘要

In the last few years, mobile cloud computing (MCC) has developed rapidly and become one of the most useful in different disciplines to communicate and exchange a lot of sensitive information within many types of mobile terminals, and computer devices. The increase of MCC usage in daily life has expanded from ten to thousands of data day by day. Meanwhile, processing and analyzing those huge amounts of data have been a major issue in data mining and machine learning techniques, wherein many traditional methods were not able to cope with a large number of instances and features found in big datasets. To overcome these drawbacks, an intrusion detection method for MCC based on MapReduce for evolutionary feature selection was proposed. In the proposed method, a MapReduce feature selection based on evolutionary computing is used to obtain a small and useful number of instances and features from big datasets. To evaluate the performance of our proposed method, the popular KDD Cup 99 dataset is used and the random forest classifier is used for classifying the normal and abnormal activities in MCC. The experimental results show that our proposed method can detect the intrusions in MCC with high accuracy, detection rate and low false positive rate.

關鍵字

Big Dataset ； Intrusion Detection ； MapReduce for Evolutionary Feature Selection ； Mobile Cloud Computing ； Random Forest Classifier