(Jillsac/Shutterstock)
Cognaize, an intelligent document processing company, has announced a major expansion of its technology platform. These new features extend the accuracy of the company’s Financial Insights Engine through a unique knowledge graph, which Cognaize says offers deeper insights and enhanced automation.
Intelligent document processing is technology that automates the extraction, processing, and analysis of data from unstructured and semi-structured documents. This technology combines elements of natural language processing, machine learning, and artificial intelligence to process, understand, and classify information in documents, improving efficiency and accelerating decision-making.
Cognaize says its platform is seeing increased adoption by financial services organizations around the world. According to Grand View Research, the IDP market is expected to reach $11.6 billion by 2030.
Cognaize IDP solutions leverage a combination of AI and humans, or what we call hybrid intelligence. The platform, according to the company, uses sophisticated financial models trained on over 1.3 million financial documents, including loan applications, SEC filings, ESG-related documents, presentations, or fiduciary reports, as well as humans throughout document automation. It combines the “experts in the loop” of process.
Cognaize has developed new features in response to customer needs and demands. A knowledge graph is a conceptual model of the world that maps entities, their properties, and their relationships. The Cognaize Knowledge Graph builds an extensive knowledge network trained on various AI models and validated by human experts for graphical processing of information. According to the company, the Knowledge Graph takes the automation of new document ingestion to an unprecedented level.
“At Cognize, we build a knowledge graph across documents,” said Vahe Andonians, Cognize CTO. Data Nami in an email interview. He gives the example of processing a bond deed whose information such as name, issuer, guarantor, interest rate, principal and current outstanding amount is spread across many different documents. “Information about all these features is scattered across different documents, making it difficult to understand the bond itself from one document,” he said.
This diagram shows hybrid intelligence between AI and humans in the validation loop of the platform. (Source: Cognize)
With Cognaize Knowledge Graph, you can automatically extract additional information about bonds, such as outstanding balances, from 10-K filings (annual reports filed by publicly traded companies with the SEC). The bond contract lists the name of the guarantor but does not detail the outstanding balance. A knowledge graph forms the connections between these various documents to provide the answer to this query.
According to the company, the technology will allow users to ask questions and receive near-instantaneous answers by running AI models ad-hoc directly in the browser. Other features the company points out include his XBRL format ingestion support, including SEC filing, and AI-powered methods for optimally distributing verification tasks to human experts.
Human verification is a key component of the Cognaize platform. Andonians explain: While AI models efficiently process large amounts of information to uncover patterns and make predictions, humans are endowed with domain knowledge, critical thinking, and creativity, Andonians say, and these traits are generated by AI. It is essential to ensure that the insights generated match the real-world context.
Recognize CTO Vahe Andonians
The Cognaize Knowledge Graph is continuously validated by financial professional users and automatically creates training data through this interaction. “As a result, we don’t distinguish between ‘training’ and ‘deploying’, but seeing it as a never-ending loop that can be fully automated or a data scientist can be put in the middle, depending on the client’s decision. ” explains the Andonians.
Andonians believe that the Knowledge Graph is a step towards replicating human intelligence through the construction of complex world models, providing a path to true artificial intelligence.
“In keeping with the rapid pace of technological innovation, we have adopted a hybrid intelligence approach that puts our clients’ financial experts at the heart of the entire document automation process, providing optimal accuracy and precision for the world’s leading innovators in banking and insurance. It guarantees efficiency,” he said. .
Related products:
Harnessing Hybrid Intelligence: Balancing AI Models and Human Expertise for Optimal Performance
A New Era of Natural Language Search for the Enterprise Has Arrived
Overcoming Data Integration Challenges to Unlock Great Potential: The Knowledge Graph 101
