The phrase "assembly command" is correct and usable in written English.
It can be used in contexts related to computer programming, particularly when referring to instructions for assembling code or commands in assembly language.
Example: "The assembly command MOV is used to transfer data from one location to another in memory."
Alternatives: "assembly instruction" or "assembly directive".
Furthermore, this encoding gives each assembly command or Asm file token a common length.
This information includes things such as function calls, strings, assembly command sequences and more.
For instance, the assembly command "adc" was assigned the base 10 index 4, which converts to the base 4 value 00010, and is encoded to the gene sequence value "AAACA".
Extracting only the assembly command sequence data from each Asm file also greatly reduced the total size and volume of data being processed by Strand during training and classification resulting in an almost 1/7 decrease in processing time when compared to models produced by using the bytes files as input.
Fig. 6 Sample of assembly commands in an Asm file pure code segment.
Unique assembly commands and tokens collected ranged from 2 to 15 characters in length.
We focused on extracting only the assembly language commands from each pure code segment contained within each Asm file. Figure 6 shows a sample of assembly commands within an Asm file's pure code segment.
For example, creating gene sequence words within Strand of length 50 would ensure that each gene sequence word contained a block of five assembly commands or Asm file tokens.
Each Asm file provided for training or classification contains a metadata manifest which includes details extracted from each malware program's binary content such as function calls, assembly commands, strings, and other relevant executable data elements.
When using the sequence data created from Asm file assembly commands and tokens, we were able to increase accuracy during ten-fold cross-validation while drastically reducing training and classification times to under 45 and 5 mins respectively per fold.
All assembly commands are provided in File S3.
