Pattern Recognition and Anomaly Detection in Bookkeeping Data
Spring 2019 to 2024
We introduce the Minimum Description Length (MDL) principle in performing tasks of pattern recognition and anomaly detection in bookkeeping data, leveraging the graph structure of double-entry bookkeeping.
- Principal Participants: Pierre Liang, Aluna Wang, Lavender Yang, Leman Akoglu, Christos Faloutsos, Jeremy Lee
- Sponsors: Digital Transformation and Innovation Center sponsored by PwC
Output
- Academic paper: AutoAudit: Mining Time-Evolving and Accounting Graphs
2020
2023
Output
- Academic Monograph: Bookkeeping Graphs: Computational Theory and Applications
- Academic paper: Summarizing Labeled Multi-Graph
- Academic paper: Detecting Anomalous Graphs in Labeled Multi-Graph Databases
- Academic paper: DAMM: Anomaly Detection of Attributed Multi-graphs with Metadata — A Unified Neural Network Approach
- Software codes
Outreach
- Tepper School Center for Intelligent Business Podcast and Web-post
- General talks at Southern Methodist University, Georgetown University, UC-Davis [more to be added]
2024