Data Governance is a big responsibility for every enterprise, especially so for finance companies. Recognising the need for data confidentiality, the honourable Supreme Court of India, RBI and other statutory agencies have mandated that Aadhaar Numbers should be fully or partially masked in documents collected from customers. To be compliant with this stipulation is a big challenge, considering the volumes.
This is where Cerescope D-tect comes to the rescue. D-tect uses its cognitive capabilities to automate the masking of Aadhaar Numbers, giving you a welcome peace of mind. This is truly AI at your service.
The process flow of D-tect Aadhaar Masking has been designed to seamlessly work with your existing systems. It can read from a folder for processing of legacy documents, while API integration is more suitable for masking documents in real time. In both cases, the process flow is simple, as you can see in the representation above.
The versatile reader
D-tect is really good at choosing its target. From the continuous stream of documents it receives, D-tect is able to recognise the ones that need masking, such as Aadhaar Cards and Application Forms. This ability is truly priceless when it comes to putting your Aadhaar Masking system into production.
You know an unpleasant fact - the documents you have are not so nice. They are badly scanned, noisy, titled and flipped. They have different sizes, resolutions and even have multiple documents put in one page. D-tect cuts through this confusion, handles all the noise stoically and gives you the masked documents that make you compliant (and happy!).
Your company has many departments and they do their files in their own way. There are pdfs, jpegs and tiffs. Some are nice e-pdfs but most are scanned and even snapped by mobiles. D-tect can handle many file types - even multi-page tiff images with ease. One less complication for you to handle.
The AI advantage
Cerescope is an AI platform, by virtue of which Cerescope D-tect can apply its cognitive capabilities to deal with document processing. It recognises and detects objects in the documents in a human-like, learning fashion. D-tect does not need templates, which are pain, and do not work anyway.
We built Cerescope D-tect ourselves, from scratch. It contains many ML models that together do the job of processing documents. We do not call any cloud api’s for computer vision (CV), text detection or OCR. This means that your confidential data is not sent anywhere outside your perimeter.
Even though D-tect handles bad data, it still returns a very good accuracy that falls in the correct range for compliance. D-tect not only serves you with the right accuracy, but also keeps on improving itself, thanks to the AI learning capabilities. All this goes a long way to ensure that you do not face any hurdles in being compliant.
Easy to use
When you are implementing a solution as important as a data governance system, you have to make sure it fits the requirements well. D-tect gives you so many options that can be configured easily. These include the inputs and outputs, how to mask (8/12 digits), how to create new files and what to do with the earlier ones. This configurability gives a lot of power in your hands.
Once you welcome D-tect in your IT landscape, it finds its own quiet corner and works well with other systems. It can integrate upstream and downstream with convenient and well documented mechanisms such as REST Api and common-folders. The clear integration methodology means less time to GoLive for you.
Your IT strategies have to be in line with the latest industry practices. Your digital initiatives might mean that a move to the cloud, or security compliances may require sensitive information to be on-premises. Cerescope D-tect can be deployed on-cloud or on-premises with equal ease, enabling you to be on the right side of your organisation’s strategy.