Unlocking unstructured data

Semantic Extract is a proven and adopted technology that applies proprietary AI techniques, Machine Learning, advanced semantics and NLP, to automatically extract target data from unstructured documents. This flexible and intelligent technology can scale across an organization providing businesses with improved operational efficiencies and a rapid ROI.

Request a demo

Key benefits

  • Improves Accuracy

    A great deal of resources are spent manually inputting data into downstream systems. Semantic Extract allows businesses to fully automate these processes whilst reducing human error.

  • Increases Accessibility

    Semantic Evolution makes extracted data available in the preferred format, place and UI. By empowering businesses with our UI, users have the ability to configure for new document types and enrich data models with no coding or scripting required.

  • Enhances Efficiency

    Users have full control to design and implement their desired workflow within Semantic Extract. By automating the data extraction process and workflow, businesses experience a quicker time to market, increased cost savings and a rapid ROI.

  • Strengthens Data Integrity

    Semantic Extract produces a full audit trail, allowing users to visualize and identify the extracted, normalized and validated data from the source document, giving businesses the confidence that data remains uncompromised.

Capabilities

The diagram below represents how Semantic Evolution can support users in automatically extracting target data from unstructured documents and converting it into actionable information.

Full Audit Trail

The system produces a Full Audit Trail allowing you to visualize and identify the data extracted from the source document.

Unstructured data

  • PDF

    Unstructured Data

    Semantic Extract supports document formats such as PDF, CSV, TIFF, JPEG, Word, Scan, Email, HTML and Excel.

  • Word

    Unstructured Data

    Semantic Extract supports document formats such as PDF, CSV, TIFF, JPEG, Word, Scan, Email, HTML and Excel.

  • Jpeg

    Unstructured Data

    Semantic Extract supports document formats such as PDF, CSV, TIFF, JPEG, Word, Scan, Email, HTML and Excel.

  • Scan

    Unstructured Data

    Semantic Extract supports document formats such as PDF, CSV, TIFF, JPEG, Word, Scan, Email, HTML and Excel.

  • Excel

    Unstructured Data

    Semantic Extract supports document formats such as PDF, CSV, TIFF, JPEG, Word, Scan, Email, HTML and Excel.

  • Html

    Unstructured Data

    Semantic Extract supports document formats such as PDF, CSV, TIFF, JPEG, Word, Scan, Email, HTML and Excel.

  • Email

    Unstructured Data

    Semantic Extract supports document formats such as PDF, CSV, TIFF, JPEG, Word, Scan, Email, HTML and Excel.

Ingestion

  • SE Extract Job

    SE Extract Job

    Extract documents in bulk continuously as they are received in users’ environment or FTP locations.

  • SE Web Agent

    SE Web Agent

    Automates configurable web scraping and downloading tasks.

  • SE Email Connector

    SE Email Connector

    Connects to e-mail exchange services and automates e-mail download services.

  • API

    API

    Enables users to swiftly modify and access information without entering the Graphical User Interface.

Parsing

  • OCR Engine

    OCR Engine

    Integrates with all major OCR solutions and can identify the OCR solution that produces the most accurate results for a given set of scanned or image based documents.

  • User Driven Data Model

    User Driven Interface

    Intuitive User Interface offers users the ability to define and construct data models with no coding or scripting skills.

  • Semantic Extract

    Semantic Extract automatically extracts target data from unstructured documents and converts it into actionable information.

  • AI Techniques Metadata

    AI Techniques

    SVMs, neural networks and other AI techniques to enrich document structure via line, contours, tables, and other document attributes detection.

  • Intelligent Business Rules Engine

    Business Rule Engine

    Allows users to capture implicit values and validate them using business logic.

Structured Data

  • SQL

    Structured Data

    Customizable output format for easy integration into existing workflows.

  • XML

    Structured Data

    Customizable output format for easy integration into existing workflows.

  • Excel

    Structured Data

    Customizable output format for easy integration into existing workflows.

  • API

    Structured Data

    Customizable output format for easy integration into existing workflows.

Add–ons

  • Workflow

    Workflow

    Customizable Workflow for exception management.

  • Auto Calibration

    Auto Calibration

    Allows for quick and efficient calibration by using parser AI to suggest calibration methods.

  • QA Expected Data

    QA Expected Data

    Data extraction process can be tracked and conflicts can be easily identified in one window.

  • Test Set

    Test Set

    Intuitive regression test set allows users to track metadata changes and its effects.

  • User Roles

    User Rights & Roles

    Allows administrators to control the functionality available to users when calibrating the parser.

Trusted by global institutions

  • IHS Markit
  • European Data Warehouse
  • Handshakes
  • FIA TECH
  • Thomson Reuters
Semantic Extract

About us

Semantic Evolution, headquartered in London, has a global reach with offices in New York and Singapore.

Leading the way in artificial intelligence, Semantic Evolution focuses on Intelligent Data Extraction.

Many businesses are required to interrogate, extract and organize data as core processes. About 80% of this data is unstructured, meaning it is buried in documents and hard to access.

Semantic Evolution helps firms address all parsing needs and transforms data into actionable information. A proven concept, the technology has been adopted by firms globally.

By using Semantic Evolution to automate the data extraction process, companies have experienced efficiency gains, data coverage improvements and reduction in processing times. Semantic Evolution provides opportunities to improve data quality, free up resources and save costs to deliver a rapid ROI for any industry.

Find us on Linkedin

Edouard Chalopin

15 years of broad exposure to financial services gained at CMA in London and then at Markit in Asia. As a sales leader, strategic account manager and business builder, Edouard has developed new markets and penetrated key accounts providing innovative solutions. Graduated from Northeastern University with a Bachelor of Science Degree in International Business (BBA) and from Reims Management School with a Diplôme d’Etudes Supérieures Européennes de Management.

Marek Chovanec

With his deep academic background in Ontologies and Semantic Networks Marek went on to spend 15 years developing commercial software. Marek spent most of this time applying new programming approaches such as Expert Systems, Neural Networks, Information retrieval and Agent Systems to create novel, scalable and successful real world applications. After 5 years leading the Data Parsing team at CMA (part of CME Group), Marek co-founded Semantic Evolution to explore new ideas in data structuring and analysis.

Stephen Madle

15 years experience in IT within the financial sector, building teams and products that have provided innovative solutions to investment banks, hedge funds and AAA rated structured credit vehicles. Stephen was the creator and co-founder of QuoteVision, the first real-time message parsing technology for the financial markets. CMA QuoteVision was acquired by CME Group in 2008. Graduate of Cambridge University and holds a Masters degree in Engineering.

Tom McNerney

With over 25 years of experience in financial markets and financial data, Tom entered the finance industry where he spent 10 years in the exotic derivatives business as a quantitative modeller and trader at SG Warburg and Barclays. This was followed by a stint as Head of Market Risk Management for Europe and Asia for Toronto Dominion Bank. In 2002 Tom co-founded Markit Ltd, a financial data and services company. Tom was instrumental in particular in building the quantitative foundations and operations of Markit's credit derivatives data and derivatives valuation businesses. Markit floated on NASDAQ in 2014 at a valuation in excess of $4bn.

David Brierwood

David holds a wealth of experience in investment data analytics and solutions, most recently in his role as Chief Operating Officer of MSCI, which he held from 2006 to February 2014. He also held various senior roles at Morgan Stanley spanning more than twenty years, including COO of the company’s equity division and COO of its institutional and retail securities groups. David is a Crown Representative in the Cabinet Office of Her Majesty’s Government, advising on procurement and the management of strategic suppliers.

Kevin MacDonald

Kevin is the former CTO and co-founder of FCS, the developer of the Wall Street Office platform. He earned his B.S. in Electrical Engineering from the University of California, San Diego and has 23 years of experience in the syndicated loan market. Under his leadership, Wall Street Office evolved to become the industry standard for loan portfolio management and accounting. Kevin oversaw the technology effort from the founding of FCS until 2006. In 2007, Kevin co-founded Black Mountain Systems, which has become the industry standard front office solution for syndicated loan managers and private debt lenders.

Parsed documents

Select a document type to see an example of how some of our commonly processed document types could be parsed:

Annual Report Financial Statement Bond Prospectus Tax Form

Other documents we process include:

Broker reports, Pricing sources, CSA, ISDA, Legal Doc, Term Sheets, Trade Confirmations, Compliance, KYC, Tax Forms, Regulatory filings, Bond Prospectuses, Credit Agreements, Loan Prospectus, Structured Finance Prospectuses, Structured Warrants, Contract notes, Fund Orders, Transfer Agencies, Invoices, SSI, Agricultural Data

Contact us

Sematic Evolution's headquarters is in London. For more information about our products and services, please email us at the following address

contact@semantic-evolution.com
  • Singapore
    Semantic Evolution Asia Pte. Ltd.
    16 Raffles Quay #33-03
    Hong Leong Building
    Singapore
    048581
  • London
    Semantic Evolution Ltd
    75 King William Street
    London
    EC4N 7BE
  • New York
    Semantic Evolution Inc
    1204 Broadway
    New York
    NY 10001