MALWARE DETECTION USING MACHINE LEARNING

MALWARE DETECTION USING MACHINE LEARNING

¹ Mrs M.Indumathy, ²  Mr R.Vinoth Kumar³ K.Karl Amarthya, ⁴ Praveen Raja

¹ Asst. Professor, faculty of Information Technology, Rajiv Gandhi college of Engineering and Technology, Kirumampakam 607403.

² Asst. Professor, faculty of Information Technology, Rajiv Gandhi college of Engineering and Technology, Kirumampakam 607403.

^3,4Department of Information Technology, Rajiv Gandhi college of Engineering and Technology, Kirumampakam 607403.

ABSTRACT:

Current  antivirus  software’s  are  effective  against  known  viruses,  if  a malware  with  new  signature  is  introduced  then  it  will  be  difficult  to  detect

that  it  is  malicious.  Signature-based  detection  is  not  that  effective  duringzero-day  attacks.  Till  the  signature  is  created  for  new  (unseen) malware, distributed  to  the  systems  and  added  to  the  anti-malware  database,  the systems  can  be  exploited  by  that  malware.  Research  shows  that  over  the  last  decade,  malware  has  been  growing  exponentially,  causing  substantial financial  losses  to  various  organizations.  Different  anti-malware  companies have  been  proposing  solutions  to  defend  attacks  from  these  malware.  The velocity,  volume,  and  the  complexity  of  malware  are  posing  new  challenges to  the  anti-malware  community.  Current  state-of-the-art  research  shows that  recently,  researchers  and  anti-virus  organizations  started  applying machine  learning  and  deep  learning  methods  for  malware  analysis  and detection.  Machine  learning  methods  can  be  used  to  create  more effective anti  malware  software  which  is  capable  of  detecting  previously  unknown malware,  zero-day  attack  etc.  We  propose  an  approach  that Various machine  learning  methods  such  as  Support  Vector  Machine  (SVM),Decision  tree,  Random  Forest  and  XG  Boost  will  be  used

Keywords:  Malware,  Machine  Learning,  Algorithms