Описание Cortical.io Contract Intelligence
Cortical.io delivers AI-based Natural Language Understanding (NLU) solutions which are quicker and easier to implement and more capable than current approaches. The company’s patented method, inspired by neuroscience, enables enterprises to more effectively search, extract, annotate and analyze key information from any kind of unstructured text.
Cortical.io Contract Intelligence analyzes relevant information from large quantities of documents quickly and with an accuracy that is difficult to achieve at scale with manual labor or with other automation tools. The solution offers a powerful, meaning-based tool to extract and classify key information, search within individual documents or over the entire database, as well as to compare documents against a template or previous versions.
Key differentiators include:
- Meaning-based extractions, search and comparison: The system understands concepts, instead of just keywords, for example descriptions of dates (“not later than ten business days after demand therefore”) or amounts (“equal to three percent (3%) of the shareholders’ equity of XY corporation”)
- Built for subject matter experts: SMEs define extraction targets and inferences, upload documents and start the training system to automatically perform extractions without the need or cost of an AI expert
- Little training material required: The system can be trained by SMEs starting with as few as 50 documents. Once the SMEs have defined custom extraction targets and annotated a few documents, they can train the system to automatically perform the extractions. SMEs also have the ability to review results and can continue to fine tune the system as part of a continuous learning process.
- Sophisticated table extraction: The solution is able to parse and extract information from tables regardless of the row/column format in the PDF document
- Built-in OCR capabilities: The system detects scanned pdf files and converts them into machine-readable files capable of being annotated
Hide