Transform Operational Efficiency
with Machine Learning Tools
Information Integration Through Transparent, Trainable Machine Learning
Just one look was all it took. IT officials at Oklahoma State University knew just minutes into a demo that this machine learning solution will provide much faster data delivery to systems campus-wide.
They eliminated separator sheets required for manual classification in their previous system – a system that was supposedly “the world’s best.”
And classification is just the tip of the iceberg.
The I.T. team at OSU saved hundreds of thousands of dollars and gained ROI in just 6 months through:
In another government use case, Grooper reduced forms processing times over 95 percent.
Using near-perfect information intelligence through machine learning, the agency’s intelligent document processing is free from templates and slow, manual data entry workflows.
How Machine Learning Tools Improve Efficiency
Improved operational efficiency using machine learning is only possible through visual training.
Grooper machine learning is trained using small sample sets. Built-in visual training data provides quick training and testing. Approved users easily duplicate and adjust pre-trained models to train new document types.
Take a look at at this article in our learning center to see Grooper machine learning and weightings in action.
Grooper uses the TF-IDF machine learning algorithm for document classification and data extraction
A visual-based design studio provides the training interface. Because of this, users don’t use programming languages like Python, NumPy, Apache Spark, TensorFlow, etc. Although these tools have their place, we’ve made it easier to get results without them.
Training applied to specific document types is stored along with a document copy. Rapid review and testing is easier for new document types because users see the training effects.
Machine learning also powers data extraction. “Feature Extractors” in Grooper are paired with easy-to-build regular expression “Value Extractors.”
They’re trained with the same visual interface. As a result, users can choose the correct choice from many similar matches found in the document.
Now it’s easy to find data, and Feature Extractors are the intelligence responsible for accurate data extraction.
Feeding a neural network? Well as you know, accurate data is a requirement for success. So be confident that you have the best data from all available sources.
Grooper’s AI tools easily find any data on structured and unstructured data. Held back by poorly scanned documents? Not a problem with built-in computer vision, over 100 image processing tools, and layered OCR.
Transform Your Operations with ML Tools!
Companies and government agencies transform operations with Grooper machine learning. This case study shows how machine learning provides:
Download our case study to learn more:
“Grooper has saved OSU hundreds of thousands of dollars and the ROI was seen in less than six months after going live. This product has taken data processing, document scanning, and import automation to a whole new level. It’s now in virtually every department including our president’s office.”Erin Girton, Database Administrator/Content Management & Capture Administrator, Oklahoma State University
“Grooper allows my staff to process ten times more volume than we could with our previous image capture solution; our office thinks Grooper is worth its weight in gold.”Marie Ramsey-Hirst, Court Clerk, Canadian County
“Grooper will give us the access to more contract data than ever before by quickly extracting the data across thousands of lengthy contracts, allowing employees to spend time on value-adding data analysis rather than extraction.”Glena Brauer, Supervisor – Marketing Contracts and Compliance, Chesapeake Energy
“Practically every unit here within our agency uses Grooper on some level. Grooper has cut down on our indexing time by 70 percent.”Ryan Freeman-Smith, Manager, Oklahoma Health Care Authority
“All of the information in some file some place is very hard to retrieve. We ought to be able to get it with the click of a button, and now we can. Using the system that Grooper provides to scan the documents and mine the data eliminates many manual processes. My belief is that in the long-term it doesn’t really cost us anything because it’s going to pay us dividends over and over again. I would be surprised if anybody has a system that’s any better than this.”Gary Ridley, former Oklahoma Secretary of Transportation, former Director of Oklahoma Dept of Transportation and former Director of Oklahoma Transportation Authority
Featured Case Studies
Thousands of companies choose BIS to enrich products and services with unique data-centric solutions. Here are some of their stories.