ZEXPR, A premier IT boutique located in southern town Nis in the Republic of Serbia named BAD SISTEMS isn’t as BAD as the name would imply. The company opened its doors in spring 2016. And, instead of taking paths like many other IT companies in Eastern Europe, the
founding partners decided to build their own products. Now, as usual, as this sounds for other developed markets, this is indeed a risky decision to make for a Serbian company, as every dime counts. The founding members sponsored the development of six products from their own pockets and through past years BAD SISTEMS has been building domain knowledge in different industries, and it seems there is new product on shelf in this premier IT boutique.
Let’s talk about the problem. Did you know that over 2 million PDF documents were sent as an attachment via Microsoft Outlook in 2018? We can safely assume that over 70% of these documents represent a great source of information for any company; it can be anything from simple invoices to financial reports being sent almost every day.
We are not even going to mention those insurance forms which are handwritten populated. A standard GO-TO solution for this problem would be a Manual Data Entry service which has become a service of its own, offered by many companies and even freelancers. Research shows that the average price for data entry in the United States is around $20 per hour.
This is a sizeable operational cost for companies. Large institutions like Banks have tried to solve this issue by investing millions in OCR (Optical Character Reading) or RegEx (Regular Expression) automated solutions. These solutions work fine for static documents, which do not change over time for a given sector. Hence, an agile solution is required to adapt to changes overtime to make it truly scalable.
Amidst these solutions and manual data entry, there has been a grey area of combining these two. Human interaction would help steer the ship in the right direction to understand the document formats. Of course, AI (Artificial Intelligence) and ML (Machine Learning) can help automate some of the repetitive workloads, but human comprehension is still very needed. Thus, we arrive at the beautiful country of Serbia, a country of incredible landscapes and even better food, but it also appears to be hiding its digital gems.
Also Read: Four Factors to Consider when hiring an MSSP
GIN – Genuine Intelligent Notation is an Engine made by BAD SISTEMS which uses AI to transform your data and make it more comprehensive, allowing you to eliminate data clutter once and for all. GIN is a natural language processing (NLP) service that uses machine learning to find insights and relationships in a text.
The service identifies the language of the text, extracts key phrases, places, people, or brands, understands the sentiment in text, and automatically organizes a collection of text files by topic.
Imagine documents, such as mortgage applications, at financial services organizations contain sensitive information that must be analyzed precisely. The slightest errors could result in the wrongful denial of a loan or a costly overpayment of interest. With Papergin, financial organizations can extract text, forms, and tables from documents.
A tremendous benefit of PAPERGIN is its ability to understand the correlation between information in between the pages, where specific information on page one could represent table on page four or five. What is truly remarkable is that a company that would use Papergin would not have to change its business process or adapt its current infrastructure to Papergin.
It is as easy as choosing a source of documents (AWS S3, Dropbox, Email or even a network folder) and defining your points of interest which will be available at the chosen destination (Excel file, API or other integrations). It is as simple as that. Oh, and it gets better; pricing is determined per processed page, leading to a significant cut in operational expenses. The saved money can be used for something more crucial than manually processing documents.
PAPERGIN is a cloud solution (SaaS). It can be self-hosted (enterprise-level) if data privacy is essential, allowing you to adapt to any circumstance or situation. One could say that all of this is already overwhelming, but no, it goes even one step further.
In recent days BAD SISTEMS announced a partnership with Accern , a no-code AI platform that enables financial organizations to easily create AI models to uncover risk and investment insights from a vast amount of data.
This partnership strives to unlock the full capacity of any given data by ingesting and extracting disordered data and turning it into a meaningful, analyzed, structured and coherent document.
For more such updates follow us on Google News ITsecuritywire News.
