Parascript: Turning Complex Data into Insights with Intelligent Machine Learning Automation
Parascript is at the forefront of developing and delivering machine learning software for intelligent document processing. The company specializes in automation to reduce manual processes and optimize document classification, data location, extraction, and verification. Parascript also helps prevent fraud and enhances the customer experience by increasing process accuracy and efficiency in areas such as mobile deposit and ATM deposit.
Established 24 years ago, the company’s mission is to automate the interpretation of meaningful, contextual data from the image and document-based information. Parascript’s motto is any source, any data, any document. Long before artificial intelligence (AI) became a buzzword, the company was building AI automation software.
Parascript was the first to solve the cursive handwriting puzzle. Very early on, Apple leveraged the technology of Parascript’s parent company (Paragraph) for the Newton series of personal digital assistants. Later, Microsoft used the company’s technology in its tablets. Since then, Parascript has grown to be the source for applied AI technologies for some of the largest postal services around the world, financial institutions, and other Global 2000 enterprises. The company’s software is used across every continent in over 50 countries.
Today, Parascript is renowned for pattern recognition, machine learning techniques, image analysis technologies and its use in cognitive robotic process automation.
As the company innovates, it leverages its extensive experience and in-depth expertise to build software that solves complex business challenges. Parascript delivers the world’s most advanced document automation solutions in the industry.
A Distinguished Technology Leader
Parascript is led by Dr. Alexander Filatov, President and CEO. He has been with the company since its inception in 1996, and as CEO for over 6 years. Alexander has more than 25 years of research and development expertise in the fields of pattern recognition, image processing, and artificial intelligence. He has a Ph.D. in Applied Mathematics from Moscow State Technical University.
Alexander is the author of many scientific papers and a holder of multiple patents in the field of machine learning. He is known in the industry for having both highly advanced technical expertise and business acumen, leading Parascript to expand its technological capabilities and footprint to address the growing demand for access to information.
Delivering High-Level Automation with Precision
While the company’s expertise with machine learning algorithms can be applied to many different domains, Parascript has chosen to focus primarily on applying its technology to image and document-based information.
One area where Parascript has made significant contributions is in handwriting recognition, which it has applied to payment and postal automation. Another area of contribution is in automating the complex task of data verification (including signature verification) to enable compliance without the traditional significant costs. Parascript’s signature verification technology is also used in Vote By Mail applications in many counties in the United States. These applications become especially important in the current situation when more states would like to switch to 100% Vote By Mail elections even before upcoming presidential elections. Automatic signature verification is necessary to ensure integrity and security of the Vote By Mail elections.
Alexander says, “Importantly, it may be possible to use off-the-shelf machine learning tools to automate work. However, it is a much more difficult task to achieve high levels of automation while simultaneously maintaining extremely high-accuracy output that allows organizations to significantly increase profitability by replacing manual tasks (including data verification) with AI-based processing. It is in this specific area—delivering high automation with high accuracy—where we have contributed the most.”
Every year, Parascript’s software automatically handles tens of billions of postal items, tens of billions of bank checks, medical, insurance and accounting forms in many countries around the world. The manual processing of each document typically costs a few cents. The result is billions of dollars in annual savings for its customers.
Leveraging Machine Learning to Gain Better Insights
While RPA solutions are not inherently disruptive from a technological perspective, RPA acts as a catalyst to reintroduce automation concepts outside of IT, which is driving the need for capabilities such as Parascript’s technology solutions. Machine Learning—especially deep learning—is causing significant disruption across many domains. This is primarily because it can parse and make inferences on large amounts of big data.
One might say that deep neural networks help big data deliver on the promise of developing better and deeper insights that can yield innovation on a variety of levels including business models and new market/product discovery.
Driving Innovation through Industry Collaborations
Alexander believes, the most fundamental aspect of what makes a company innovative is building the right team and hiring truly innovative people. Parascript has a unique team of highly experienced engineers and scientists—over a third with PhDs—many with national and international awards in mathematics and programming.
The company’s innovation is also the result of working in close collaboration with its clients and partners who provide it insight into problems where it is best suited to deliver solutions. He further added, “It is all about applying our expertise in a variety of machine learning techniques to solve specific client challenges. In this respect, innovation is a key criterion for project success.”
Parascript’s partners are business service providers, VARs, system integrators and OEMs. Through innovation, the company remains at the center of their intelligent processing by reliably and precisely automating all their complex document-oriented tasks.
Remarkable Recognition and Achievements
Parascript has recently been recognized by KMWorld in the top “100 Companies that Matter in Knowledge Management” for the fifth consecutive year. KMWorld has also recognized Parascript FormXtra.AI as trend-setting software.
Everest Group recognized the company as a Major Contender in its Intelligent Document Processing Products PEAK Matrix™ Assessment 2019.
Moreover, Deep Analysis advises its clients to consider using Parascript “in high volume, complex format-capture situations to replace and consolidate multiple first-generation capture products.” CIO Applications also recognized the company as one of the “Top 10 Machine Vision and Imaging Technology Solution Providers” in 2018.
Winning Against Business Challenges
As described by Alexander, one challenge that Parascript faces is that organizations are sometimes unwilling to invest in evaluating the accuracy of vendors’ products. In its experience, the company wins the business if an organization conducts a thorough evaluation of the competing technologies.
Parascript is a key part of many systems, and yet, it doesn’t carry the same name recognition of some of its major partners. Ultimately, in the markets the company serves, it ends up getting known through word-of-mouth due to the positive results that it delivers. Lately, Parascript has been much more visible owing to the addition of key partners that work with it side-by-side on client needs, which provides more awareness and visibility about just what the company’s software can do.
Another challenge over the past decade is battling the perception that document automation, which was once called, “document capture,” is no longer necessary due to born-digital documents. However, as the company first saw with enterprise content management, then business process management, and now RPA, intelligent document processing or IDP is always a key enabling capability that drives just about everything else downstream. So, Parascript is enjoying a renaissance in demand for the capabilities that it provides.
A final challenge, which is a problem for any IDP vendor, is the high level of complexity relative to other complementary parts of an automation solution. Alexander asserts, “Typically, the services portion of a project just for configuration and tuning of IDP can be 2 or 3 times that of the other software and it limits adoption.” Parascript is solving that problem with what it calls Smart Learning, which effectively removes a significant amount of cost and complexity from implementations, improves automation level, and shortens time to deployment of the technology. The result is that the capability can scale to smaller volume projects and still deliver good ROI, and for big projects, the benefits are even more significant. This has made the company’s growing list of partners very happy.
In terms of milestones, Parascript has achieved some major successes over the past two decades ranging from postal automation, check processing, signature verification to document automation using Smart Learning. The company’s software increased automation for handwritten addresses from 22% to 96% automation for one of the largest letter and parcel carriers in the world. Parascript software processed bank checks handily beating Bell Labs product performance initially and then improved industry automation levels over the years from 40% to 98% automation. The company has won international competitions for signature verification systems. Now in the document capture space, Parascript has developed deep learning-based, unstructured document classification and reading technology called Smart Learning, which provides the highest quality data results in the industry.
The Future Belongs to Automation
Alexander believes the future belongs to automation across the front and back offices, which will augment, improve or reinvent current tasks that are—even today—often manual, and to advanced analytics. At the core of automation and analytics is data. Parascript software ensures that the data ingestion and extraction are of high quality, which massively improves its clients’ and the industry’s opportunities for success. Quality data is at the very center of every successful digital transformation project and business strategy decision.
“Automation has never been more popular than it is now and is a major initiative for just about any organization. The result is that, while in the past we had to spend a lot of time making the case for automation, working on only the largest projects, organizations are doing that math for themselves and coming to us for help,” he said. And because automation has characteristics that support business cases in both good times and in economic downturns, Parascript is enthusiastic about the business climate over the next several years.
Source: https://www.analyticsinsight.net/parascript-turning-complex-data-insights-intelligent-machine-learning-automation/
