Our previous work
Custom AI products
Over the years we have worked with quite a variety of organisations. What they had in common? The wish to do their work better and smarter. With data, machine learning and AI.
In this time we have also developed AI products or better yet AI building blocks that we can personalise according to your wishes.
Will you profit from this as well?
Read on!keyboard_arrow_rightBack to all projectskeyboard_arrow_right"We have AI building blocks that can help you save time."
Saving time in research and sharing data
The upside of well-written code (and good IP agreements) is that you can reuse it. And the nice thing about machine learning (and as such language models) is that they can be trained further to specialise in specific subject areas and in different languages.
With our customisable AI products
We can do this with smaller and bigger pieces of code and machine learning models. The biggest of which are our 'AI products' Selectical and Pseuduck which we can personalise to support you in your work.
Scroll to products directlykeyboard_arrow_rightBased on (multilingual) NLP
NLP, or Natural Language Processing, concerns the use of linguistics, data science, artificial intelligence and machine learning to let a computer 'understand' natural language. (Think of ChatGPT.) With NLP you can e.g. make it easier to search through texts, summarize them, anonymise them, judge them, ask questions, etc.
Most language models are based on English, while in other countries – like in the Netherlands – organisations often operate completely in their own language. That is why we also work with Dutch NLP and other languages.
Pseuduck: our anonymisation tool
Well, actually Pseuduck is an pseudonymisation tool, but if you can pseudonymise you can also anonymise. It all comes down to recognizing different entities like names, cities, phone numbers, addresses, gender, professions, etc. This we do with Named Entity Recognition (NER).
With it you can recognise whether there is personal data in your texts and documents and mask, rename or delete them according to your wishes. Ideal for when you want to share data with customers, colleagues, partners, for research, etc.
More on Pseuduckkeyboard_arrow_right
Selectical: our literature selection tool
Finding and/or selecting relevant documents, texts or new publications for your research (whether this concerns due dilligence or systematic literature research), it can be quite a task.
If you have to do this often, you might have thought at some point "wouldn't it be nice to have a personal assistent".
Well, is exists! It's called Selectical. :)
More on Selecticalkeyboard_arrow_rightPseuduck | Pseudonymising | Anonymising | Privacy
Ensuring privacy when sharing data
'Just sharing data or documents' is not that simple. Often it contains personal and private information. So if you want to comply to the GDPR, you will often need to put some time and effort into anonymising or pseudonymising the data, before you can work with it.
Today, most anonymisation work is done by hand. This a lot of work, costs quite some time and is – like any other crappy job – prone to errors. Especially when it concerns free text!
Pseuduck to the rescue for GDPR proof texts
Pseuduck is a pseudonymisation tool developed by us that makes it easier to make your free texts – ranging from small text fields to large (collections of) documents – GDPR proof.
The benefits of Pseuduck are that we can teach it to specialise further in different fields and sectors, to pseudonymise or anonymise your texts, using your jargon and according to your needs and requirements.
What sensitive texts do you have to work with?
Contact us for a trial!keyboard_arrow_rightSelectical | Literature Research | NLP | Active Learning
Literature research in 1/3 of the time
Researchers doing (systematic) literature reviews need to screen many thousands research papers and select the ones that are relevant to a specific subject. Usually out of these thousands of papers, only a few hundred are actually useful. Making this selection manually takes a lot of useful time and is simply boring.
Without pre-labeled data
Usually we train an AI model using labeled data. Using examples. But that is not possible in this case! The challenge is that since every literature study is different, there are no examples yet. Labeling these examples just so happens to be your job. That doesn't mean that you can't use AI though!
And still use AI
For this we use a technique called Active Learning: this type of AI learns from human input, and can improve itself constantly while working. This way the researcher can start working the same way as usual, but with an AI learning in the background. Once the AI is certain enough that it has identified the right patterns, it will take over the labeling. This way the researcher only has to validate or correct the AI's work. This saves a lot of time!
Read more about Selecticalkeyboard_arrow_rightAsk for a demo!keyboard_arrow_rightWant to own your own AI product?
Before we get to work on solutions together, we will make clear agreements on the goals, code and knowledge (re-) usage and eventual intellectual property ownership. Many arrangements are possible. We like to work and think along with you!