ahlashkari/androidapplyzer

Static and Dynamic Android App analyzer

androidapplyzer's Issues

Question about the static dataset

Hi Team,

Thanks for sharing the CIC datasets and the scripts. May I check with you do the static analysis CSV datasets (malicious and benign) contain any column features? Originally planned to do some top-ranking feature analysis (analytics) so wondering if your team has any recommendation for us to access the features for the raw data apk. Thanks for any advice.

Questions regarding dataset

Good afternoon. I have some questions regarding the actual CCCS-CIC-AndMal-2020 dataset. I tried emailing but couldn't get in contact with anyone.

Regarding the static portion of the dataset:

1. The BenignCSVs files (Ben0 - Ben4) have a total of 162,181 entries. I'm a bit confused if I'm missing the other ~40k samples that are mentioned in the writeup.

2. In this git (AndroidAppLyzer), the Unique_Lists contains 60,948 entries. In the DIDroid: Android Malware Classification and Characterization, it says there were 9504 features, that were reduced to 2465 and then 2237. Is there a list of the original 9504 features anywhere? Is there a list of the final 2237 that were chosen to be included?

3. The provided MD5 hashes don't match

   • BenignCSVs.md5 is e40f78f9ec8a4b26ca914c28c40f6ad1
   • Actual MD5 of BenignCSVs.zip is 8fd36b3c490423cb53b575c7bbe2f1a0
   • MaliciousApps-LabelledHash.md5 is a907619fbfe864f7893606ac4409c2fc
   • Actual MD5 of MaliciousApps-LabelledHash.zip is 9c9a3fa772164517ed453d79e5b73b1c
   • Malicious-CSVs.md5 is 4116bd27bf8397f770fe3f29617acb53
   • Actual MD5 of Malicious-CSVs.zip is 2d44e1cac47a797d61abc6f7b91b9ae8

Can you provide any insight on these questions? Apologies if this isn't the right place to ask.

What is layer2.cvs in the CNN_model.py?

Hello, I'm confused about what this file stands for, where does it come from?