{"id":23835,"date":"2021-04-13T05:11:17","date_gmt":"2021-04-13T05:11:17","guid":{"rendered":"https:\/\/rbsitsoftwaresolution.com\/blog\/docugamis-new-model-for-understanding-documents-cuts-its-teeth-on-nasa-archives.html"},"modified":"2021-04-13T05:11:17","modified_gmt":"2021-04-13T05:11:17","slug":"docugamis-new-model-for-understanding-documents-cuts-its-teeth-on-nasa-archives","status":"publish","type":"post","link":"https:\/\/www.rbsitsoftwaresolution.in\/blog\/docugamis-new-model-for-understanding-documents-cuts-its-teeth-on-nasa-archives.html","title":{"rendered":"Docugami\u2019s new model for understanding documents cuts its teeth on NASA archives"},"content":{"rendered":"

You hear so much<\/span> about data<\/em> these days that you might forget that a huge amount of the world runs on documents<\/em>: a veritable menagerie of heterogeneous files and formats holding enormous value yet incompatible with the new era of clean, structured databases. Docugami<\/a> plans to change that with a system that intuitively understands any set of documents and intelligently indexes their contents \u2014 and NASA is already on board.<\/p>\n<\/p>\n

If Docugami\u2019s product works as planned, anyone will be able to take piles of documents accumulated over the years and near-instantly convert them to the kind of data that\u2019s actually useful to people.<\/p>\n

Because it turns out that running just about any business ends up producing a ton of documents. Contracts and briefs in legal work, leases and agreements in real estate, proposals and releases in marketing, medical charts, etc, etc. Not to mention the various formats: Word docs, PDFs, scans of paper printouts of PDFs exported from Word docs, and so on.<\/p>\n

Over the last decade there\u2019s been an effort to corral this problem, but movement has largely been on the organizational side: put all your documents in one place, share and edit them collaboratively. Understanding the document itself has pretty much been left to the people who handle them, and for good reason \u2014 understanding documents is hard!<\/p>\n

\n

Hugging Face raises $40 million for its natural language processing library<\/a><\/p>\n<\/blockquote>\n

\n

Think of a rental contract. We humans understand when the renter is named as Jill Jackson, that later on, \u201cthe renter\u201d also refers to that person. Furthermore, in any of a hundred other contracts, we understand that the renters in those documents are the same type<\/em> of person or concept in the context of the document, but not the same actual<\/em> person. These are surprisingly difficult concepts for machine learning and natural language understanding systems to grasp and apply. Yet if they could be mastered, an enormous amount of useful information could be extracted from the millions of documents squirreled away around the world.<\/p>\n

What\u2019s up, .docx?<\/h2>\n

Docugami founder Jean Paoli says they\u2019ve cracked the problem wide open, and while it\u2019s a major claim, he\u2019s one of few people who could credibly make it. Paoli was a major figure at Microsoft for decades, and among other things helped create the XML format \u2014 you know all those files that end in x, like .docx and .xlsx? Paoli is at least partly to thank for them.<\/p>\n

\u201cData and documents aren\u2019t the same thing,\u201d he told me. \u201cThere\u2019s a thing you<\/em> understand, called documents, and there\u2019s something that computers<\/em> understand, called data. Why are they not the same thing? So my first job [at Microsoft] was to create a format that can represent documents as data. I created XML with friends in the industry, and Bill accepted it.\u201d (Yes, that Bill.)<\/p>\n

The formats became ubiquitous, yet 20 years later the same problem persists, having grown in scale with the digitization of industry after industry. But for Paoli the solution is the same. At the core of XML was the idea that a document should be structured almost like a webpage: boxes within boxes, each clearly defined by metadata \u2014 a hierarchical<\/em> model more easily understood by computers.<\/p>\n

\"Illustration<\/a><\/p>\n

Image Credits:<\/strong> Docugami<\/p>\n

\u201cA few years ago I drank the AI kool-aid, got the idea to transform documents into data. I needed an algorithm that navigates the hierarchical model, and they told me that the algorithm you want does not exist,\u201d he explained. \u201cThe XML model, where every piece is inside another, and each has a different name to represent the data it contains \u2014 that has not been married to the AI model we have today. That\u2019s just a fact. I hoped the AI people would go and jump on it, but it didn\u2019t happen.\u201d (\u201cI was busy doing something else,\u201d he added, to excuse himself.)<\/p>\n

The lack of compatibility with this new model of computing shouldn\u2019t come as a surprise \u2014 every emerging technology carries with it certain assumptions and limitations, and AI has focused on a few other, equally crucial areas like speech understanding and computer vision. The approach taken there doesn\u2019t match the needs of systematically understanding a document.<\/p>\n

\u201cMany people think that documents are like cats. You train the AI to look for their eyes, for their tails \u2026 documents are not like cats,\u201d he said.<\/p>\n

It sounds obvious, but it\u2019s a real limitation. Advanced AI methods like segmentation, scene understanding, multimodal context, and such are all a sort of hyperadvanced cat detection that has moved beyond cats to detect dogs, car types, facial expressions, locations, etc. Documents are too different from one another, or in other ways too similar, for these approaches to do much more than roughly categorize them.<\/p>\n

As for language understanding, it\u2019s good in some ways but not in the ways Paoli needed. \u201cThey\u2019re working sort of at the English language level,\u201d he said. \u201cThey look at the text but they disconnect it from the document where they found it. I love NLP people, half my team is NLP people \u2014 but NLP people don\u2019t think about business processes. You need to mix them with XML people, people who understand computer vision, then you start looking at the document at a different level.\u201d<\/p>\n

Docugami in action<\/h2>\n

\"Illustration<\/a><\/p>\n

Image Credits:<\/strong> Docugami<\/p>\n

Paoli\u2019s goal couldn\u2019t be reached by adapting existing tools (beyond mature primitives like optical character recognition), so he assembled his own private AI lab, where a multidisciplinary team has been tinkering away for about two years.<\/p>\n

\u201cWe did core science, self-funded, in stealth mode, and we sent a bunch of patents to the patent office,\u201d he said. \u201cThen we went to see the VCs, and SignalFire basically volunteered to lead the seed round at $10 million.\u201d<\/p>\n

Coverage of the round didn\u2019t really get into the actual experience of using Docugami, but Paoli walked me through the platform with some live documents. I wasn\u2019t given access myself and the company wouldn\u2019t provide screenshots or video, saying it is still working on the integrations and UI, so you\u2019ll have to use your imagination \u2026 but if you picture pretty much any enterprise SaaS service, you\u2019re 90% of the way there.<\/p>\n

As the user, you upload any number of documents to Docugami, from a couple dozen to hundreds or thousands. These enter a machine understanding workflow that parses the documents, whether they\u2019re scanned PDFs, Word files, or something else, into an XML-esque hierarchical organization unique to the contents.<\/p>\n

\u201cSay you\u2019ve got 500 documents, we try to categorize it in document sets, these 30 look the same, those 20 look the same, those five together. We group them with a mix of hints coming from how the document looked, what it\u2019s talking about, what we think people are using it for, etc.,\u201d said Paoli. Other services might be able to tell the difference between a lease and an NDA, but documents are too diverse to slot into pre-trained ideas of categories and expect it to work out. Every set of documents is potentially unique, and so Docugami trains itself anew every time, even for a set of one. \u201cOnce we group them, we understand the overall structure and hierarchy of that particular set of documents, because that\u2019s how documents become useful: together.\u201d<\/p>\n

\"Illustration<\/a><\/p>\n

Image Credits:<\/strong> Docugami<\/p>\n

That doesn\u2019t just mean it picks up on header text and creates an index, or lets you search for words. The data that is in the document, for example who is paying whom, how much and when, and under what conditions, all that becomes structured and editable within the context of similar documents. (It asks for a little input to double check what it has deduced.)<\/p>\n

It can be a little hard to picture, but now just imagine that you want to put together a report on your company\u2019s active loans. All you need to do is highlight the information that\u2019s important to you in an example document \u2014 literally, you just click \u201cJane Roe\u201d and \u201c$20,000\u201d and \u201cfive years\u201d anywhere they occur \u2014 and then select the other documents you want to pull corresponding information from. A few seconds later you have an ordered spreadsheet with names, amounts, dates, anything you wanted out of that set of documents.<\/p>\n

All this data is meant to be portable too, of course \u2014 there are integrations planned with various other common pipes and services in business, allowing for automatic reports, alerts if certain conditions are reached, automated creation of templates and standard documents (no more keeping an old one around with underscores where the principals go).<\/p>\n

\n

No code, workflow and RPA line up for their automation moment<\/a><\/p>\n<\/blockquote>\n

\n

Remember, this is all half an hour after you uploaded them in the first place, no labeling or pre-processing or cleaning required. And the AI isn\u2019t working from some preconceived notion or format of what a lease document looks like. It\u2019s learned all it needs to know from the actual docs you uploaded \u2014 how they\u2019re structured, where things like names and dates figure relative to one another, and so on. And it works across verticals and uses an interface anyone can figure out in a few minutes. Whether you\u2019re in healthcare data entry or construction contract management, the tool should make sense.<\/p>\n

The web interface where you ingest and create new documents is one of the main tools, while the other lives inside Word. There Docugami acts as a sort of assistant that\u2019s fully aware of every other document of whatever type you\u2019re in, so you can create new ones, fill in standard information, comply with regulations and so on.<\/p>\n

Okay, so processing legal documents isn\u2019t exactly the most exciting application of machine learning in the world. But I wouldn\u2019t be writing this (at all, let alone at this length) if I didn\u2019t think this was a big deal. This sort of deep understanding of document types can be found here and there among established industries with standard document types (such as police or medical reports), but have fun waiting until someone trains a bespoke model for your kayak rental service. But small businesses have just as much value locked up in documents as large enterprises \u2014 and they can\u2019t afford to hire a team of data scientists. And even the big organizations can\u2019t do it all manually.<\/p>\n

NASA\u2019s treasure trove<\/h2>\n

\"\"<\/a><\/p>\n

Image Credits:<\/strong> NASA<\/p>\n

The problem is extremely difficult, yet to humans seems almost trivial. You or I could glance through 20 similar documents and a list of names and amounts easily, perhaps even in less time than it takes for Docugami to crawl them and train itself.<\/p>\n

But AI, after all, is meant to imitate and transcend human capacity, and it\u2019s one thing for an account manager to do monthly reports on 20 contracts \u2014 quite another to do a daily report on a thousand. Yet Docugami accomplishes the latter and former equally easily \u2014 which is where it fits into both the enterprise system, where scaling this kind of operation is crucial, and to NASA, which is buried under a backlog of documentation from which it hopes to glean clean data and insights.<\/p>\n

If there\u2019s one thing NASA\u2019s got a lot of, it\u2019s documents. Its reasonably well-maintained archives go back to its founding, and many important ones are available by various means \u2014 I\u2019ve spent many a pleasant hour perusing its cache of historical documents<\/a>.<\/p>\n

But NASA isn\u2019t looking for new insights into Apollo 11<\/a>. Through its many past and present programs, solicitations, grant programs, budgets, and of course engineering projects, it generates a huge amount of documents \u2014 being, after all, very much a part of the federal bureaucracy. And as with any large organization with its paperwork spread over decades, NASA\u2019s document stash represents untapped potential.<\/p>\n

Expert opinions, research precursors, engineering solutions, and a dozen more categories of important information are sitting in files searchable perhaps by basic word matching but otherwise unstructured. Wouldn\u2019t it be nice for someone at JPL to get it in their head to look at the evolution of nozzle design, and within a few minutes have a complete and current list of documents on that topic, organized by type, date, author and status? What about the patent advisor who needs to provide a NIAC grant recipient information on prior art \u2014 shouldn\u2019t they be able to pull those old patents and applications up with more specificity than any with a given keyword?<\/p>\n

\n

How to access \u2018America\u2019s Seed Fund,\u2019 the $3 billion SBIR program<\/a><\/p>\n<\/blockquote>\n

\n

The NASA SBIR grant, awarded last summer, isn\u2019t for any specific work, like collecting all the documents of such and such a type from Johnson Space Center or something. It\u2019s an exploratory or investigative agreement, as many of these grants are, and Docugami is working with NASA scientists on the best ways to apply the technology to their archives. (One of the best applications may be to the SBIR and other small business funding programs themselves.)<\/p>\n

Another SBIR grant with the NSF differs in that, while at NASA the team is looking into better organizing tons of disparate types of documents with some overlapping information, at NSF they\u2019re aiming to better identify \u201csmall data.\u201d \u201cWe are looking at the tiny things, the tiny details,\u201d said Paoli. \u201cFor instance, if you have a name, is it the lender or the borrower? The doctor or the patient name? When you read a patient record, penicillin is mentioned, is it prescribed or prohibited? If there\u2019s a section called allergies and another called prescriptions, we can make that connection.\u201d<\/p>\n

\u201cMaybe it\u2019s because I\u2019m French\u201d<\/h2>\n

When I pointed out the rather small budgets involved with SBIR grants and how his company couldn\u2019t possibly survive on these, he laughed.<\/p>\n

\u201cOh, we\u2019re not running on grants! This isn\u2019t our business. For me, this is a way to work with scientists, with the best labs in the world,\u201d he said, while noting many more grant projects were in the offing. \u201cScience for me is a fuel. The business model is very simple \u2014 a service that you subscribe to, like Docusign or Dropbox.\u201d<\/p>\n

The company is only just now beginning its real business operations, having made a few connections with integration partners and testers. But over the next year it will expand its private beta and eventually open it up \u2014 though there\u2019s no timeline on that just yet.<\/p>\n

\u201cWe\u2019re very young. A year ago we were like five, six people, now we went and got this $10 million seed round and boom,\u201d said Paoli. But he\u2019s certain that this is a business that will be not just lucrative but will represent an important change in how companies work.<\/p>\n

\u201cPeople love documents. Maybe it\u2019s because I\u2019m French,\u201d he said, \u201cbut I think text and books and writing are critical \u2014 that\u2019s just how humans work. We really think people can help machines think better, and machines can help people think better.\u201d<\/p>\n

\n

How artificial intelligence will be used in 2021<\/a><\/p>\n<\/blockquote>\n

\n

\n

Noogata raises $12M seed round for its no-code enterprise AI platform<\/a><\/p>\n<\/blockquote>\n

\n

Article source: http:\/\/feedproxy.google.com\/~r\/techcrunchIt\/~3\/jJ2jlImvz4E\/<\/p>","protected":false},"excerpt":{"rendered":"

You hear so much about data these days that you might forget that a huge amount of the world runs on documents: a veritable menagerie of heterogeneous files and formats <\/p>\n","protected":false},"author":1,"featured_media":23836,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"jetpack_post_was_ever_published":false,"_jetpack_newsletter_access":"","_jetpack_dont_email_post_to_subs":false,"_jetpack_newsletter_tier_id":0,"_jetpack_memberships_contains_paywalled_content":false,"_jetpack_memberships_contains_paid_content":false,"footnotes":"","jetpack_publicize_message":"","jetpack_publicize_feature_enabled":true,"jetpack_social_post_already_shared":true,"jetpack_social_options":{"image_generator_settings":{"template":"highway","default_image_id":0,"font":"","enabled":false},"version":2}},"categories":[19],"tags":[25],"class_list":["post-23835","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-enterprise-software","tag-enterprise"],"aioseo_notices":[],"aioseo_head":"\n\t\t\n\t\n\t\n\t\n\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t