The Automatic Accountant: Difference between revisions

From Computer Laboratory Group Design Projects
Jump to navigationJump to search
No edit summary
No edit summary
Line 1: Line 1:
Client: Mark Parsons at [[Cambridge Enterprise]] would be interested for 2022
Client: Mark Parsons, [[Cambridge Enterprise]] <Mark.Parsons@enterprise.cam.ac.uk>


The Automatic Statistician is a system created by Zoubin Gharamani and colleagues that looks for interesting patterns in data sets (by fitting Gaussian process models), and then automatically generates a research paper, including charts and natural language texts, describing the patterns that it finds. Financial accounting data is relatively constrained in its format, so it should be more straightforward to create an AI system that automatically generates professional-looking reports on a company’s performance, including some audit capabilities that work by recognising statistical anomalies.
The Automatic Statistician is a system created by Zoubin Ghahramani and colleagues that looks for interesting patterns in data sets (by fitting Gaussian process models), and then automatically generates a research paper, including charts and natural language texts, describing the patterns that it finds. Financial accounting data is relatively constrained in its format, so it should be more straightforward to create an AI system that automatically generates professional-looking reports on a company’s performance, including some audit capabilities that work by recognising statistical anomalies. You will need to familiarise yourself with the research from www.automaticstatistician.com, and work with your client to design an original accounting application.

Revision as of 10:00, 26 October 2021

Client: Mark Parsons, Cambridge Enterprise <Mark.Parsons@enterprise.cam.ac.uk>

The Automatic Statistician is a system created by Zoubin Ghahramani and colleagues that looks for interesting patterns in data sets (by fitting Gaussian process models), and then automatically generates a research paper, including charts and natural language texts, describing the patterns that it finds. Financial accounting data is relatively constrained in its format, so it should be more straightforward to create an AI system that automatically generates professional-looking reports on a company’s performance, including some audit capabilities that work by recognising statistical anomalies. You will need to familiarise yourself with the research from www.automaticstatistician.com, and work with your client to design an original accounting application.