Wednesday, November 25, 2020
Home Lead Story Intel-Microsoft Collaborated Project Turns Malware into Images

Intel-Microsoft Collaborated Project Turns Malware into Images

Intel and Microsoft joined hands to work on the study

Researchers from Intel and Microsoft have joined forces to study the use of deep learning for malware threat detection in a project that first converts malware into images.

The basis for this study is the observation that if malware samples are turned into grayscale images, the textural and structural patterns can be used to effectively classify them as either benign or malicious, as well as cluster malicious samples into respective threat families, Microsoft said.

The researchers used an approach that they called static malware-as-image network analysis (STAMINA), Jugal Parikh and Marc Marino from Microsoft Threat Protection Intelligence Team wrote in a blog post.

For the first part of the collaboration, the researchers built on Intel’s prior work on deep transfer learning for static malware classification and used a real-world dataset from Microsoft to ascertain the practical value of approaching the malware classification problem as a computer vision task.

Using the dataset from Microsoft, the study showed that the STAMINA approach achieves high accuracy in detecting malware with low false positives.

The results were detailed in a paper titled “STAMINA: Scalable deep learning approach for malware classification”.

intel-logo
The researchers used an approach that they called static malware-as-image network analysis (STAMINA). Wikimedia Commons

Read More: Healthy Eating Habits in Toddlers Reduces Chances Of Heart Realted Risks Later: Study

To establish the practicality of the STAMINA approach, which posits that malware can be classified at scale by performing static analysis on malware codes represented as images, the study covered three main steps: image conversion, transfer learning, and evaluation.

The study was performed on a dataset of 2.2 million PE file hashes provided by Microsoft. This dataset was temporally split into 60:20:20 segments for training, validation, and test sets, respectively.

The joint research encourages the use of deep transfer learning for the purpose of malware classification. (IANS)

STAY CONNECTED

19,120FansLike
362FollowersFollow
1,780FollowersFollow

Most Popular

Self-Care Guidelines To Boost Immunity, With Special Reference To Respiratory Health

With the monsoon winding down and winter moving in, colder weather will arrive soon, making us more prone to catching the flu or common...

Few Points For Pet Parents To Keep In Mind

Every dog has his or her own individual personality. This personality goes beyond the common labels used for each breed, for instance, Labradors are...

The 2021 Grammy Awards Nominations Are All About Women Power

The 2021 Grammy Awards nominations are all about women's power with Beyonce, Taylor Swift, and Dua Lipa leading the way. Beyonce is leading the...

Young People’s Anxiety Levels Doubled During Covid-19 Lockdown

New research adds to a growing body of evidence that young people's anxiety levels doubled during Covid-19 lockdown. The study, published in the British...

4,400 Species Across The Globe Are At Risk Due To Changes In Wildfire Intensity

Changes in wildfire intensity and frequency driven by human activities are putting at risk more than 4,400 species across the globe, says a study....

Scotland Becomes The First Nation To Provide Sanitary Products For Free

Scotland on Tuesday made sanitary products free to all women, becoming the first nation in the world to take such a step against "period...

The Pandemic-Era Incarnation Of Self-Care

These days, with a pandemic raging, this is what life can look like:  out the days in loungewear. Wearing minimal makeup because no one...

ICRA: Restocking Exaggerated Improvement In Oct’s Economic Performance

Restocking activity ahead of the festive season to satisfy pent-up demand had exaggerated the pace of improvement recorded by many lead economic indicators in...

Recent Comments