Many enterprises instituted a work-from-anywhere environment due to necessity this year, and in-person events were cancelled. With the absence of our in-person events, we all attended virtual ones. What became very clear was attendees often had the same questions.
Industry experts tend to use a lot of terms in a lot of different contexts without a lot of definition. This is compounded by a fresh wave of technological innovation and new demand for automation to make remote workers’ lives easier and maintain or improve enterprise efficiency.
So, we decided to start some conversations and answer those most-asked questions about intelligent document processing. Here are the results. If you have a topic you want covered or you want to pose a question, let us know. We’re planning a new set of these mini-webinars based on the success of ones below.
Handwriting Recognition
Handwriting recognition is the ability for software to read and convert handwriting – handprinted text or cursive – into machine-print. Interesting developments have been made over the last several years in handwriting recognition. It is still one of the most difficult parts of information or business information to wrangle. Key developments in handwriting recognition are discussed here. This includes handwriting recognition innovations and its importance today.
What Is Natural Language Processing or NLP?
Natural Language Processing or NLP essentially deconstructs text, whether it’s a text from a phone or social media or text within a document. NLP is most generally defined as the automated processing or manipulation of natural language – speech or text – by software so that information can be inferred from it. What we’re focused here on text within a document, whether that document is born-digital such as a word document, converted into a PDF or if it’s a document that’s scanned and then subsequently run through OCR software. Discover the true importance of NLP to Intelligent Capture.
The Wrong and Right Way to Measure Intelligent Capture
Measuring an Intelligent Capture system can be either done in a proof of concept, in a head-to-head evaluation of solutions or to optimize an existing intelligent capture implementation. Measuring a system means measuring at the task level, which is any type of action that a user takes to process a document. The wrong and the right way to measure intelligent capture and assess its accuracy. Find out how to measure the quality of your results.
How Many Sample Are Necessary for Your Intelligent Capture to be Successful?
One of the most common questions that organizations needing to implement intelligent document processing have is, “how many samples do we need?” There are two primary drivers for how many samples should be used and two areas where samples are necessary.
Secret Life of a Confidence Score
A confidence score essentially is a number assigned to a task, but it’s not just a number. We derive the number from the intelligent document processing system. The confidence score typically ranges between 0-100, they might go beyond a hundred. There are some systems that have maximum confidence scores of 1,500. Ultimately, competence scores come from the software that outputs them and are associated with a specific task. How do we derive confidence scores and what do they mean when applied in intelligent automation? Both answered here.
Essential Playbook for Conducting an Intelligent Document Processing – IDP Evaluation
Intelligent Document Processing is essentially software that can classify, locate and extract data from documents and these documents can be born-digital (such as email or PDF) or they can be hard copy such as (scanned or faxed documents) for use in business systems, internal applications, RPA and other automation platforms across the organization. IDP software can leverage simple technologies such as OCR to very advanced machine learning or deep learning. The essential playbook for conducting an Intelligent Document Processing (IDP) Evaluation is presented here based on several highly successful use cases.
Ultimately, these mini-webinars offer a basis for understanding the terminology used in the industry, how to focus on having really good sample sets and then what to evaluate in the output of intelligent capture systems.
An intelligent capture system should be able to be configured quickly and also easily maintained in production. It is nice to have a straightforward, easy-to-use UI, but the engine behind that UI must be capable of accurate document classification, data location, extraction and verification so that the output is consistent and highly accurate for real straight through processing.