SAN FRANCISCO–(BUSINESS WIRE)– Seal Software heeft vandaag de nieuwe versie van zijn programma voor de analyse van contracten onthuld: Versie 5.0. Met de nieuwe editie verschijnt ook “Analyze This Now” (ATN), een add-in voor Microsoft Word die de zakelijke gebruiker in staat stelt Seals krachtige functies voor data-abstrahering en -beoordeling te benutten binnen de bekende interface van MS Word. Dit verkort de contactcyclus, verkleint risico’s en verlaagt de verwerkingskosten. De nieuwe versie verruimt bovendien het raamwerk voor machinematig leren, dus de technologie van Seal is ook bruikbaar wanneer zakelijke gebruikers systemen willen leren complexe data-uittreksels uit te voeren. Deze uittreksels verschijnen zodoende sneller, zijn accurater en verkorten de weg naar rendement.
Seal Software Unveils New Version of its Contract Discovery and Analytics Solution to Empower Fortune 500 Business Users
SAN FRANCISCO–(BUSINESS WIRE)– Today Seal Software unveiled a new version of its leading contract discovery and analytics solution: Version 5.0. The new version introduces “Analyze This Now” (ATN), an add-in for Microsoft Word that allows business users to leverage Seal’s powerful contract data extraction and review capabilities from within the familiar MS Word interface in order to shorten contact cycle times, reduce risk and lower processing costs. The release also extends Seal’s machine learning framework and interface so Seal’s technology can be used by business users to train the system for complex data extractions, improving extraction speed, accuracy and shortening time-to-value.
“Seal Version 5.0 was developed with business users in mind, allowing them to easily harness the power of Seal without extensive education or resources. We want those closest to the business objectives in procurement, legal, M&A, and sales to use our system without extensive training and data management skills,” said Ulf Zetterberg, co-founder and CEO of Seal Software. “Empowering business users helps lighten the load on legal operations teams, who are often buried in contract processing, improving business performance and impacting the bottom line.”
Analyzing Contract Terms in Real Time
When contracts need to be reviewed, re-papered, or revised, the heavy lifting tends to fall on a very busy legal operations team. Extensive review cycles with business owners and counterparties often result in long lead-times and high costs to get contracts completed. By putting the power of Seal in the hands of business users, the workload on legal ops is greatly reduced, shortening contract cycle times and reducing costs.
Seal’s Analyze This Now (ATN) pulls existing contracts directly into Microsoft Word, converting all documents – including PDF and TIFF images – into MS Word format. From there, business users can review text, update terms, change metrics, and create new versions within their familiar Word interface. With ATN, the Seal platform is running in the background, analyzing the text and showing users the status of clauses and provisions using color-coded shading for clear visibility. Seal clearly indicates if text matches approved legal language or if it is non-standard text, and the degree of variance. It also shows the status of contracts currently under review.
Machine Learning for Business Users
There have always been edge cases, however, where contract data does not exist in clear expressions, when it is buried within subtle links in text across various sections, or is generally hidden in ways more difficult to decipher. With Version 5.0, Seal is now exposing the layer it has used when working with clients on these complex data extractions. This new user interface is called User Driven Machine Learning (UDML), and allows business users to train the system’s machine learning functions – in a way specifically designed for them.
Business users can now provide positive and negative examples of the information they are looking for and Seal will create the optimal set of algorithms and models, applied in its deep learning framework, to optimize that specific data extraction process. As users define the objectives of the data extraction, the system’s dynamic scaling adjusts on-demand to compensate for the various training models.
Seal has always been focused on speed and simplicity for its contract data extractions, needing just a single example from users to extract clauses, dates, people, money and other discrete data points. The UDML advances Seal’s extraction capabilities beyond what is available in the market today, by rounding out Seal’s “four pillars” of data extraction, including full search capabilities, Natural Language Processing and Latent Semantic Indexing, Machine Learning models, and now putting its Deep Learning framework in the hands of end users to optimize complex extractions.
“Our new solution reflects significant work from the Seal product team to put the power of machine learning into non-technical users’ hands,” said Kevin Gidney, co-founder and CTO of Seal Software. “We are now combining our original, single example method with the exposed AI functions, meaning customers can now take full advantage of the Seal platform, as its team of 40+ developers, data scientists, and legal services teams now do for customers. Version 5.0 is also our first major stepping stone in delivering our overall Seal vision, which will dramatically change how contracts will be processed and their data managed over the next 18 months.”
About Seal Software
Seal Software is the leading provider of contract discovery, data extraction, and analytics. With Seal’s machine learning and natural language processing technologies, companies can find contracts of any file type across their networks, quickly understand what risks or opportunities are hidden in their contracts, and place them in a centralized repository. Based in San Francisco, Seal empowers enterprises around the world to maximize revenue opportunities, reduce costs, and mitigate risks associated with contractual documents, systems, and processes. For more information, visit Seal Software at www.seal-software.com.