Skip to content

Latest commit

 

History

History
99 lines (83 loc) · 5.17 KB

cfg-bns-citation-summary.md

File metadata and controls

99 lines (83 loc) · 5.17 KB

Quick Summaries of Papers that Cite CFG-BNS Method

DroidDet: Effective and robust detection of android malware using static analysis along with rotation forest model

  • 2018, 81 citations
  • Extracts 4 key features, viz., permissions, senstitive APIs, monitoring system events, and permission-rate
  • Uses those extracted features to train a rotation forest model
  • Cites background

Automatic malware classification and new malware detection using machine learning

  • 2017, 46 citations
  • Uses gray-scale images, n-gram opcodes and import functions as features and passes those features through a decision making algorithm to classify
  • Also uses Shared Nearest Neighbour (SNN) algorithm to detect new malware
  • Cites method of n-gram usage

HEMD: a highly efficient random forest-based malware detection framework for Android

  • 2018, 26 citations
  • Extracts 4 key features, viz., permissions, senstitive APIs, monitoring system events, and permission-rate
  • Uses those extracted features to train an ensemble random forest model
  • Cites background

Automatic extraction of malicious behaviors

  • 2016, 10 citations
  • Make a model of a program by building its API call graph. Represent malicious behaviours with a malicious API graph
  • Retrieve relevant subgraphs from the API call graph of a target program by using TFIDF and classify with the help of them
  • Cites method

A novel graph-based approach for IoT botnet detection

  • 2020, 11 citations
  • Lightweight static analysis and effective in dealing with multi-architectural device networks
  • Extracts high level features from function-call graphs for each files, makes them into PSI-graphs and then passes through CNN
  • Cites method

SMASH: a malware detection method based on multi-feature ensemble learning

  • 2019, 11 citations
  • Uses multiple features like API call sequence, resistance to evading memory dumps and hardware performance counters
  • Enrich the features with neural networks and then use ensemble learning algorithms to classify
  • Cites method

A novel framework to classify malware in mips architecture-based IoT devices

  • 2019, 6 citations
  • Uses syscalls obtained in MIPS-architecture devices to classify
  • Cites method

Malware detection with convolutional neural network using hardware events

  • 2017, 5 citations
  • Uses hardware features like cache miss rate and branch misprediction rate are collected from Intel based CPUs
  • Such features are then fed to a CNN (finally achieving an AUC of 0.9973)
  • Cites method

Malware detection based on directed multi-edge dataflow graph representation and convolutional neural network

  • 2019, 3 citations
  • Builds MeQDFGs which are a kind of multi-egde data flow graphs (nodes consist of processes, files, registry keys and network sockets)
  • Feeds the MeQDFGs to Graph Convolutional Network
  • Cites method

PEFile analysis: a static approach to ransomware analysis

  • 2019, 2 citations
  • Statically analyses the Portable Executable (PE) files and keeps track of the DLLs used by it to detect whether it's a ransomware
  • Cites method

Detection and classification of obfuscated malware

  • 2016, 1 citation
  • Detects whether a program is obfuscated and then classifies based on file structure, runtime behaviour, instructions and obfuscation
  • Additionally, detects the malware packers or obfuscators used by the program by analysing CFG-based signatures
  • Cites method

Multimodal approach for malware detection

  • 2019, 1 citation
  • Uses multimodal features like power consumption, system logs, network traffic, and code-based static data
  • Classifies with the help of standard supervised ML algorithms like, Random Forest, J48, JRip, PART, Naive Bayes, and SMO
  • Cites method

Anticoncept Drift Method for Malware Detector Based on Generative Adversarial Network

  • 2021, no citations
  • Trains on actual malware samples as well as generated malware samples to better classify derived malware
  • Uses API call sequences
  • Cites method

Malware Detection Method based on Control Flow Analysis

  • 2019, no citations
  • Extracts execution traces from CFGs and then converts opcodes into intermediate code that is then represented by Vector Space Model
  • Uses Naive Bayes Classifier, Support Vector Machines and Random Forest model for classification
  • Cites method

Selecting Prominent API Calls and Labeling Malicious Samples for Effective Malware Family Classification

  • 2019, no citations
  • Extracts API call sequences that are then used as n-gram features (for n = 1, 2)
  • Feature selection methods like PCA and KPCA are employed for the grams after which, SVC, RF and KNN are used to classify the samples
  • Cites method

An Approach To Comparing Control Flow Graphs Based On Basic Block Matching

  • 2020, no citations
  • Compares basic blocks and edges of CFG by matching identical instructions
  • Uses 3 matching criteria: identical instructions of a pair of blocks and identical instructions in all blocks reachable from out and in edges from them
  • Cites method

Opcode Frequency Based Malware Detection Using Hybrid Classifiers

  • 2020, no citations
  • Uses a list of 1-gram opcodes to feed to standard ML models and classify the samples
  • Cites method