Skip to content
/ Hmm Public

Hidden Markov Model code developed from scratch. Configured to run on Linux BlueHive Supercomputing Cluster.

Notifications You must be signed in to change notification settings

ROC-HCI/Hmm

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

26 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Hidden Markov Models (HMM)

Hidden Markov Model code developed from scratch to be used to identify deceptive behavior from the ROC-HCI deception dataset.

Authors: Taylan Sen and Matt Levin

Main Files/Folders

  • hmm.py - Implements and test a hidden Markov model
  • truth_bluff.py - Dual-HMM classification on a dataset using cross-validation
  • lstm_hmm.py - Modified HMM to avoid exponential decay of prolonged hidden states
  • runner.sh - Used to deploy code on a BlueHive computing cluster (SLURM system)
  • analysis - Jupyter notebooks to analyze results and code to apply confidence-based thresholds to results
  • sh - Shell scripts for dispatching jobs to obtain the dataset and other utilities

About

Hidden Markov Model code developed from scratch. Configured to run on Linux BlueHive Supercomputing Cluster.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published