Integrate artificial intelligence and NLP into your research

We can help you expand your business research capabilities by using a mix of large and small language models combined with traditional Natural Language Processing (NLP) and other sophisticated coding techniques to convert qualitative business disclosures into quantitative data.

Contact Us
Leverage our custom NLP system

We have built a powerful Natural Language Processing (NLP) system that lets us extract the largest possible quantity of high-quality textual data to perform a wide variety of standard and custom calculations.

Solutions for your specific project
When you meet with us, we'll begin by discussing your needs, help you brainstorm possible approaches, and discuss feasibility
Extract a variety of sources
We augment AI models with hand-tuned annotations to extract the largest possible number of high-quality financial disclosures from many sources
Use existing metrics, or design your own
We have expertise calculating a wide variety of common measurements from financial disclosures, or we can help you design a new one to suit your needs
We provide insights, not black boxes
We will share with you a detailed description of important decisions made when creating your dataset so you can better understand the data production process
How We Can Help

We've helped clients create new and innovative data sets, replicate published measures, and respond to revise-and-resubmit requests from top-tier journals, by helping to:

  • Provide many classic NLP measures, such as: length, sentiment/tone, readability, year-over-year change analysis, keyword search, text classification, and more
  • Analyze EDGAR filings, including risk factors, MD&A, and financial statement footnotes
  • Use a variety of LLMs to augment and replicate human analysis, including GPT (OpenAI), Claude (Anthropic), Gemini (Google), and Llama (Meta)
  • Streamline and maximize human effort when hand-coding is required
  • Extract structured data from PDFs
  • Use APIs from Google, SerpAPI, and others to build large-scale datasets
  • Create topic models to determine relative mix of corporate disclosures
  • Implement new and existing measures to quanitify earnings conference calls, analyst reports, and other public disclosures
Who We Are
Our Mission
Quality
Our carefully-designed systems help us provide the largest quantity of accurate and reliable data possible from a variety of sources
Understanding
We strive to communicate as clearly and thoroughly as possible to ensure our clients understand how their data was created and are able to explain it to others
Reproducibility
We believe reproducibility is critical to useful data analysis and support this goal with thorough documentation of our processes and results
Our Expertise
Our founder—Stephen V. Brown, Ph.D. —has extensive experience in information technology, spending over a decade as a full-time systems and software engineer before getting his Ph.D. in Business Administration from the University of Florida. He has published novel research using cutting-edge NLP techniques in top-tier academic journals, with extensive hands-on experience analyzing corporate disclosures, developing machine learning and AI models, and producing high-quality datasets for statistical analysis. You can learn more on his personal web site .
Portrait of Stephen Brown in New York City
More Information
Frequently Asked Questions
Get in Touch
Email
[email protected]