A New Language and Hidden Markov Input/Output Model for Automated Audit Inquiry | Programming Repository
Pushkin Kachroo, Aaron Saiewitz, Robyn Raschke, Shaurya Agarwal, Archie J. Huang
This is the code reference page of the paper "A New Language and Hidden Markov Input/Output Model for Automated Audit Inquiry".
An "Auditor Inquiry - Client Response" data sample is also available in this repository.
The program is built using Dr. Mogeng Yin's Input Output Hidden Markov Model (IOHMM) Python package.
The sample data have 52 communicating dyads simulating the auditor-client communication process.
In this dataset, there are 7 auditor inquiries, coded as I1, I2, ..., I7. There are more than 21 client responses, coded as R1, R2,..., R21, as sometimes a client response was logged as a "partial response". For instance, a response which beared resemblance to response R1 was recorded as a "R1P" response ("P" for "Partial").
Inquiry | Frequency |
---|---|
I7 | 131 |
I1 | 40 |
I2 | 24 |
I5 | 23 |
I4 | 22 |
I3 | 21 |
I7 | 6 |
For more information, please refer to the following:
- P. Kachroo, A. Saiewitz, R. Raschke, S. Agarwal and A. J. Huang, "A New Language and Input–Output Hidden Markov Model for Automated Audit Inquiry" in IEEE Intelligent Systems, vol. 35, no. 6, pp. 39-49, 1 Nov.-Dec. 2020
@article{kachroo2020new,
title={A New Language and Input-Output Hidden Markov Model for Automated Audit Inquiry},
author={Kachroo, Pushkin and Saiewitz, Aaron and Raschke, Robyn and Agarwal, Shaurya and Huang, Archie J.},
journal={IEEE Intelligent Systems},
year={2020},
volume={35},
number={6},
pages={39-49},
doi={10.1109/MIS.2019.2963653}
publisher={IEEE}
}