APIMDS: API Call Sequence\u2013Based Malware Detection Dataset

This dataset was constructed to support research on malware detection through dynamic API call sequence analysis. It includes API call traces extracted from 23,080 malware samples, along with benign program traces, generated in a controlled virtual machine environment using API hooking techniques. The dataset reflects realistic execution behaviors such as DLL injection, IAT hooking, antidebugging routines, and screen-capture actions. The API calls were collected using a Detours-based hooking system on Windows XP SP3, and categorized into 26 functional API groups for further analysis. Researchers can utilize this dataset to study sequence alignment, behavioral similarity, and dynamic malware profiling.