GOVERNMENT
Uncovering fraud with a unified view across many data sources
To find evidence of fraud, investigators may need to extract and compare information from multiple internal sources such as email and file shares, as well as public intelligence sources such as social media.
How Environ Assists
-
Giving you a single window across all these sources
-
Automatically extracting and correlating connections between people, objects, locations and events (POLE)
-
Quickly making the relevant facts and documents available for review and further investigation
CONNECTING THE DOTS
Uncovering fraud is a game of connecting the dots. In some industries such as financial trading, statistical analysis can flag suspicious activity. Textual analysis may provide clues to the pressure on the individual, the opportunity to commit fraud and the ability to rationalise the crime – the classic fraud triangle. However, finding proof is harder.
​
The clues may lie in emails and attachments, documents on file shares, written correspondence, phone call recordings and voicemails, and external sources such as social media and forum sites. They may only become apparent by comparing the claims people make against indications of real-world activity such as building access logs, records in business systems or patterns
of behaviour.
​
It would be extremely time consuming, and in many cases impossible, for a fraud investigator to manually correlate company names, people, email addresses, telephone numbers, IP addresses and many other potential connections. You need to integrate all this information into a single view.
ADVANCED INVESTIGATIVE SOFTWARE
Environ’s advanced investigation and digital forensics software is engineered to triage, process, analyse and bring to the surface critical evidence across entire data sets. It has unmatched ability to extract relevant text and metadata from enterprise content stores, email servers and archives, files shares, cloud services and other potential information sources. It can also perform optical character recognition during processing to extract searchable text from images and scanned documents.
​
The Environ ‘named entities’ model automatically extracts key intelligence items such as individual and company names, email addresses, sums of money, credit card numbers and countries. What’s more, you can construct your own named entities using regular expressions to extract any type of information that follows a specific pattern, such as customer, policy and contract numbers.
Environ’s advanced investigation and digital forensics software is engineered to triage, process, analyse and bring to the surface critical evidence across entire data sets.
ONE WINDOW INTO THE EVIDENCE
Using Nuix software, you can ingest relevant data from internal content repositories – including email systems, mobile devices and file shares – as well as open source intelligence from sites such as social media and forums.
This gives you a powerful arsenal of fraud analysis techniques:
​
-
Search and tag. Using lists of known fraud terms, you can run automated search-and-tag operations during processing to highlight potentially relevant documents at the outset of the investigation.
​
-
Search and filter. You can perform simple keyword searches or complex Boolean, wildcard, proximity, fuzzy and phrase searches across all these data sources at once. You can filter the data by date range, file type, custodian or any other metadata field.
​
-
Near duplicates. Finding duplicate and near-duplicate text can help you identify counterfeit documents and see how people shared, modified and reused those documents over time.
​
-
Rubbery figures. With sums of money from multiple sources, you can use techniques such as round sum or duplicate payments, Benford’s law and timeline analysis to identify suspicious figures or transactions.
​
-
Location data. Nuix can extract location data from digital photos, mobile devices, IP addresses and other geotagged files, and automatically display them on a map. You can automatically tag items connected with countries on the global hotlist for bribery and corruption.
​
-
Combined analytics. Combining multiple analytics can rapidly identify the most suspicious documents, such as a list of emails containing known fraud keywords and sums of money sent to Gmail addresses in high-risk countries.
​
-
Automated link analysis. Nuix really saves you time by automatically finding the connections within the intelligence you extract. The ‘link analysis’ visualisation (figure 1) shows the connections between people and the intelligence items they share. The Nuix Context interface (figure 2) makes it easy to explore connections and drill down to fine details. Nuix Insight Analytics & Intelligence (figure 3) examines all the available evidence through the POLE framework.
​
​
These powerful analytics can uncover connections, even at multiple levels of removal, that any human being would struggle to spot.
MAKING THE FINAL DECISION
Having identified proof of fraud, or potential signs that warrant further review, you can now tag the relevant documents and start off a workflow for managers, the legal team or subject matter experts to log in and review the evidence. These reviewers can examine the documents relevant to their areas of expertise, for example spreadsheets for the financial team. They can perform additional searches or analytics, add their own tags or redactions and generate reports before making a final decision.
