Back to search

University of Nevada - Reno

XAPID: An Explainable AI Framework for Behavior-based Malware Detection via Windows API Calls

Abstract

dc:description.abstract

As the online threat of malware continues to become more complex and pervasive, novel detection strategies are always being developed and expanded upon. In particular, behavior-based detection via analysis of Windows Application Programming Interface (API) calls has sparked considerable debate and development over the past decade. By examining the Windows API calls made by an executable throughout its runtime, researchers can gain insight into behavioral patterns that are characteristic of malicious or benign operations. Current research and detection solutions perform this analysis using artificial intelligence, feeding the API call sequences into various machine learning models to classify executables as malicious or benign. Although these systems can perform classification with high accuracy, researchers have identified a substantial need for Explainable AI (XAI) systems within malware detection. As these AI solutions tend to create black-box systems, in which users cannot be sure exactly what metrics are determining the final classification result, there is a natural lack of trust and an inability to understand the meaning behind the classification. Which API calls and sequences suggest malicious behavior, and what meaning can we extrapolate from the context of those sequences to better understand the malware's behavior as users? In this study, we propose the Explainable API-based Detection (XAPID) framework. This framework provides a method for performing behavior-based malware detection using Windows API calls, implementing explainable AI solutions to help human users derive meaningful behavioral patterns from API call sequences.

Degree

thesis:*
Level thesis:degree_level
Master's Degree
Department thesis:degree_department
Computer Science
Year dc:date.issued
2025

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Hankins, Amber
Advisor dc:contributor.advisor
  • Sengupta, Shamik
Committee members dc:contributor.committeemember
  • LaTourrette, Nancy
  • Park, Sung-Yeon

Subjects

dc:subject × 4

Rights

Language dc:language.iso
en_US, English

Identifiers

dc:identifier.*
Repository record dc:identifier.uri
https://scholarwolf.unr.edu/handle/11714/11379

Chain of custody

source
Harvested from
University of Nevada - Reno
Base URL
scholarwolf.unr.edu/server/oai/request
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
citation

Hankins, Amber. XAPID: An Explainable AI Framework for Behavior-based Malware Detection via Windows API Calls. Master's Degree thesis, 2025. https://scholarwolf.unr.edu/handle/11714/11379