Finding fraudsters using genomic analysis
The Swiss startup OrphAnalytics uses genomics to detect academic fraud and authenticate documents.
Founded in the Swiss canton of Valais in 2014, OrphAnalytics applies a genomic algorithm to analyse documents for evidence of academic plagiarism and ghostwriting. The startup’s CEO Claude-Alain Roten explains how applications of their software is now extending to legal investigations, including financial fraud and anonymous harassment.
Technologist: How does OrphAnalytics’ technology work?
Claude-Alain Roten: The technology enables us to capture the syntax of a document. A document is made up of semantic elements, i.e. words, which are themselves structured into turns of phrase by the author. At this level, syntax is essential as it indicates the author’s style choices. Our method determines authorship by comparing the writing style in the document with that of other documents by the same author. This only requires a few pages.
T: In addition to detecting academic fraud, you also apply the technology to legal situations. In what specific cases?
CAR: Whether an investigation concerns criminal acts or blackmail, we can analyse documents and identify the author from a limited number of suspects. We also used the technology on key documents in the Ford-Kavanaugh case in the United States.
Brett Kavanaugh, the recently appointed judge to the Supreme Court, was accused by Christine Blasey Ford, a psychology professor, of carrying out sexual assault 36 years ago. Our experts compared the stylistic signature of two key documents written by Ford with the transcription of her answers delivered orally during the hearing. Our style analysis confirmed her authorship of the key documents in the case. These results prove that Christine Blasey Ford is the true author of the written statements, therefore corroborating their authenticity.
Claude-Alain Roten, CEO
Whether an investigation concerns criminal acts or blackmail, OrphAnalytics can analyse documents and identify the author from a limited number of suspects.
CAR: Our technology can also be applied in suspected scams. For instance, an international court recently used our technology to investigate an attempt to extort money by manipulating contracts worth several hundred million euros. We are currently working with the Institut de Police Scientifique at the University of Lausanne for this reason.
T: You analyse written and handwritten documents, but does the technology work with transcriptions of speech?
CAR: The Ford-Kavanaugh case in the United States gave us the opportunity to move into a new phase – dealing with legal issues by comparing written styles with oral statements. People create a very personal signature in the way they use language in both written statements and oral communication: it’s like a fingerprint. This means we can analyse transcriptions, which is especially useful in cases of anonymous phone harassment.
T: What are the next phases of development?
CAR: We are applying our innovative technology to a number of fields, such as music and digital information. We’ve optimised our algorithms to determine identity based on just a few hundred characters. That’s ideal for social media. We are always trying to apply our cross-disciplinary technique to new areas. For now we have 10 clients – two in Europe and the rest in Switzerland.
The European Central Bank is offering a payment system to compete with digital services and credit card companies. European start-ups could benefit from it.
Artisense develops a 3D-vision technology for autonomous cars and robots to navigate effectively.