Meta-Feature Classification to Explore Automatic Detection of Malware Using Segmentation Method
Chandra Sekhar Vasamsetty1, Siva Sankar Chandu2, Janakidevi Maddala3
1Chandra Sekhar Vasamasetty, Department of Computer Science Engineering, SRKR Engineering College, Bhimavaram, India.
2Siva Sankar Chandu, Department of Computer Science Engineering, SRKR Engineering College, Bhimavaram, India.
3Janakidevi Maddala, Department of Computer Science Engineering, SRKR Engineering College, Bhimavaram, India,
Manuscript received on 03 August 2019 | Revised Manuscript received on 09 August 2019 | Manuscript published on 30 August 2019 | PP: 3458-3462 | Volume-8 Issue-10, August 2019 | Retrieval Number: J97190881019/19©BEIESP | DOI: 10.35940/ijitee.J9719.0881019
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Paper Anti-malware software producers are persistently tested to recognize and counter new malware as it is discharged into nature. An emotional increment in malware generation as of late has rendered the ordinary technique for physically deciding a mark for each new malware test unsound. This paper introduces a versatile, mechanized methodology for identifying and arranging malware by utilizing design acknowledgment calculations and measurable techniques at different phases of the malware examination life cycle with Meta highlights. By utilizing a regular fragment examination, Mal-ID can dispose of malware parts that start from kindhearted code. What’s more, Mal-ID uses another sort of highlight, named Meta-include, to more readily catch the properties of the dissected portions. In this paper, we introduce Ensemble Classifier technique to handle malware uncovering based on meta features. Our system consolidates the static highlights of capacity length and printable string data separated from malware tests into a solitary test which gives order results superior to those accomplished by utilizing either include independently. In our testing, we information includes data from near 1400 unloaded malware tests to various diverse grouping calculations. Utilizing k-overlap cross approval on the malware, which incorporates Trojans and infections, alongside 151 clean records, we accomplish a general characterization exactness of over 98%.
Keywords: Malware Detection, Classifier, Security, Ensemble Learning, Meta-Feature Classification and Communication Infrastructure.
Scope of the Article: Classification