Blog - Cognitive Data Capture, Machine Learning, AI

The Digitalization of Accounting

We are at the peak of a technological revolution.

In 2017, almost 2 billion invoices were exchanged all over the world and 90% of those invoices were still processed manually. For accounting firms, this means time, money and resources wasted on manual data entry and employees left feeling unfulfilled in a repetitive job. Rossum is dedicated to creating a world without manual data entry, putting an end to these mundane and tedious tasks still done by humans.

Over the past few months, we have hosted a series seminars on the “Digitalization of Accounting” to help Czech accounting firms realize the utilization of artificial intelligence for invoice data capture. Karin Fuentesova, Chief of AI Data Set, presented at each seminar on the importance of implementing AI into accounting firms’ operations. In this blog, we will divulge details that were shared at each seminar with the goal of encouraging accounting firms to put an end to the mundane data entry that is still plaguing them.

Continue reading →

Founder Interview with Chief AI Architect Petr Baudiš

In our second Founder Interview blog, I sit down with Petr Baudiš, Chief AI Architect of Rossum, on our terrace on a surprisingly warm late autumn afternoon. We discuss growing up programming, starting Rossum and what new technologies Rossum has been developing. Named one of Financial Times’ New Europe Challenger in 2017, Petr has published many scientific papers which have even been referenced in AlphaGo’s Nature paper. Petr is a disruptor in the world of AI and his work at Rossum is doing just that. Read on and you’ll see why.

Continue reading →

TAGS: founder interview Startup tech

Peeking into the Neural Network Black Box

If you have ever come in touch with neural networks, you are probably familiar with the black box problem [1, 2]. Compared to many other algorithms from the glass box category [2], neural networks are inherently difficult to dissect. This should come as no surprise. We use neural networks to find solutions to problems which are difficult for humans to put into the language of algorithms. Whenever it is difficult for an expert to find features which would help any other Machine Learning algorithm, that’s typically where neural networks come into the picture and blow the competition out of the water.

So what can be done? As it turns out, the inherent difficulty of explaining how neural networks work does not deter everyone and there are actually people who, I would say, are even drawn to the difficulty [1]. I still think the proposed methods are far from explaining how a neural network works on the entire dataset, on all classes etc. Right now, in my opinion, the explanation methods are most powerful in a sample-by-sample examination. They are, for example, able to tell you which parts of the image play the most important role in an image classification task. I would still argue that there are important conclusions to be drawn about the task as a whole.

Even though the methods are very general, in the end, I will of course focus on models we use at Rossum for invoices. If you last until the end of this blog post, I promise you will see some nice results regarding invoices and tips on how to use those in your own projects. Continue reading →

TAGS: AI Machine Learning Neural Networks

Tensorflow for cropping and rendering

In tensorflow, there are many usable features for tasks besides just for learning. One worth mentioning is the function for rendering framed rects for given boundingboxes that can be used for fast evaluation of predictions of an image. But what if we want to get fancy? We have recently faced a task, to crop given areas from an image, process them into features and render them back onto the image. This could be done in keras or tensorflow so that it can be fast and embedded in a model. Or even possibly – overlay an image, not with framed boundingboxes, but alphablend with filled rectangles. To note, we want a feature vector rendering, not a full graphics renderer with tensorflow, which already exists (at least in the form of this example).

Here we will be introducing the concepts with the code, but we have also published the code that should be able to run in ipython notebook. Continue reading →

TAGS: Keras Machine Learning TensorFlow

Blue Prism RPA Meets Rossum AI

Blue Prism has recently unveiled its Digital Exchange, an intelligent automation marketplace that presents downloadable assets with pre-built artificial intelligence. This revolutionary one-stop shop helps companies add skills to digital workers through cognitive and disruptive technologies. We are thrilled to be a part of the Blue Prism ecosystem, as well as being a new Technology Affiliance Partner. Find out how our partnership works, and how you can incorporate automated invoice data capture into your business process with minimum human involvement.

Continue reading →

TAGS: AI Blue Prism RPA

Inside line items: Our progress and evaluation techniques

Our researchers have been hard at work solving line item extraction from invoices over the past few months. We have made several leaps forward we will talk about in the post, and you can test to see how it works in our live demo, right now! The rest of the post is aimed at our fellow data science geeks and all about our performance indicators concerning line items.

Quick update about line items in Elis

We have a lot of results we are eager to share, as well as just the story of our journey towards breathtaking table extraction capabilities. Whether it’s intense hackathons, exciting competition among multiple teams, or just endless thorny issues on how to annotate something as complex and varied as invoice table data – but that’s for another blog post.

The gist of the story, short & sweet, is that users of the Elis verification interface can now enjoy dramatic quality improvement in the semi-automatic line item capture we offer, and fully automatic table extraction is now available in our Data Extraction API as an experimental feature.

This is why you can test to see how it works right now in our live demo. To follow up on our previous line items blog post, let us show you a few examples. It simply works for many layouts already.

Continue reading →

TAGS: AI line items

Blue Prism Continues to Drive Intelligent Automation Capabilities Through New Affiliate Technology Alliance Partnerships

Blue Prism’s recent announcement on the newest additions to their Technology Alliance Program.

K2, Loop AI Labs, re:Infer and Rossum join Blue Prism’s Technology Alliance Program As Affiliates Making it One of the Broadest and Most Innovative in the Industry

Austin, Texas, and London — November 13, 2018 — Blue Prism (AIM: PRSM), today announced the addition of K2, Loop AI Labs, re:Infer and Rossum to its Technology Alliance Program as affiliates, strengthening an already broad array of technology and independent software vendors (ISVs), that provide new and breakthrough intelligent automation capabilities. These partners enable customers to leverage Blue Prism’s Robotic Process Automation (RPA) platform to build out best-in-class solutions that incorporate cutting edge, artificial intelligence (AI), machine learning, OCR and data extraction capabilities. Continue reading →

TAGS: AI Blue Prism Partner Technology

Will the Virtual Office replace the Traditional Office?

As technology is advancing, the workplace as we know it is changing forever. The Mad Men-esque office environment is long gone, and in its place we are left with a question – will the traditional office survive and how will it continue to evolve in the future?

Below we have outlined the pros and cons for a traditional office and a virtual office. We dive deep and try to determine which is the better choice in the end for the employee and employer. Read on and decide for yourself.

Traditional Office: A “traditional office” has changed its meaning over the last century – from rooms separated by department, to cubicles, to open office plans – a singular definition is not possible. For the sake of this blog, let’s consider that a “traditional office” is one where the employees of a company commute to the same physical location where they conduct their work – no matter the layout of the office.

Now think of where you work – does it meet this definition? How do you feel about this type of office? What are some ways that could improve your current office situation? Read on to discover some advantages and disadvantages of a traditional office environment.

Continue reading →

TAGS: AI office virtual