We use cookies

This website uses cookies to provide necessary website functionality, improve your experience and analyze our traffic. By using our website, you agree to our Privacy Policy and our cookies usage.

Contact us

Our friendly team would love to hear from you.




    or contact us directly at office@cognitum.eu

    This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
    ionicons-v5-e

    Thank you for your interest!

    We will contact you as soon as possible.

    Back to homepage
    CogniPy Graph Database and Knowledge Graph Development
    Back to stories

    In-memory Graph Database and Knowledge Graph

    Open source product development compatible with Semantic Web standarts.

    Highlights

    Our client from the tax-fraud detection field was interested in our tool FluentEditor, an ontology editor, but wanted to use particular components in the functionality of the product they developed. Then the idea arose to create CogniPy – a python module, which grew from FluentEditor and could be used as a building block that s/he needs and builds her own application. We have discovered that we need to open source our technology, as we believe it will be beneficial for the community of researchers working in different fields of science. 

    CogniPy is a Graph Database and Knowledge Graph with Natural Language Interface which simplifies and optimizes the work with data in general. Thanks to it, you can implement your time-consuming data driven projects easier and make the resulting solutions more user-friendly for non-technical specialists e.g. medical experts. Besides, as it is compatible with OWL, a well known semantic web standard, and therefore it contributes to OWL community growth.

    Industry

    Technology

    Science

    Team

    12

    Duration

    months

    Country

    Poland

    Challenge

    The target audience is data scientists, data analysts, data and machine learning engineers. This open source is a Python package as we know that 90% of data scientists work with is done in this programming language. Moreover, most of them are using jupyter notebooks, spyder or similar Python based environments and CogniPy is compatible with all of them. Also it is compatible with Pandas that is the standard for tabular data analytics.

    The main challenge we were faced with, is the application of the grammar of natural language (English) on top of semantics rules, although this language is still considered relatively simple.

    Solution

    Other analogues allow modeling data in the form of graphical diagrams, which we found difficult to use for domain experts. Therefore, we decided to change the principle of data modeling and developed a tool that displays it in the form of natural language.

    As a result, we launched a very intuitive tool, connecting data and non-technical people. From the point of view of possible application in various industries, great opportunities open up.

    Healthcare and Pharma. If knowledge is transmitted in the form of graphical diagrams, specialists have to undergo separate training or courses in order to be able to perform knowledge modeling. Using natural language data, no such training is required as it has the form of sentences. With CogniPy you will be able to skip this stage of staff training. Possible use: systems and applications for taxonomy, data on drugs and their interaction with each other, clinical researches, integration with complex structures.  

    Automotive and Manufacturing. Possible use: systems and applications for the description, connection and interaction of multiple different auto and assembly parts, integration with complex structures. Learn more about Smart Solution for Continuous Line Operation.

    Aerospace and Defense. Possible use: systems and applications for aeronautics, connection and interaction of multiple different construction parts. Learn more about Smart Business Intelligence Tool.

    Security and Regtech. Possible use: systems and applications for collection, storage and presentation of sensitive data, integration with complex structures. Learn more about Live Comply Regtech Solution.

    Telecom. Possible use: systems and applications for collection, storage and presentation of sensitive data, integration with complex structures. Learn more about Cybersecurity Assessment PoC.

    From the technical side, there are existing graph databases that require high computer power to process it. But for your prototypes, you usually don’t need an external server to store your graph while prototyping, so your working device is enough. 

    CogniPy can work with Excel spreadsheets via the Pandas package. It also has an autocomplete function and a predictive editor. Moreover, CogniPy allows you to import data using Pandas, or if you scrap it from the Internet you can integrate these data with other data sources. Moreover developers can use results from CogniPy that can have a form of tabular data being the result of standard SPARQL queries, with other technologies, including Tensorflow or PyTorch to build training sets designed for Deep Learning.

    In the future, we have plans for the next iterations of this open source library. We plan to include more complex languages in terms of semantics (Polish, other European languages) and Latin, which is a universal language in medicine.

    To learn more about the tool please visit our webpage.

    Technology Specification

    Cognitum Software House Python
    Cognitum Software House C#
    Cognitum Software House Java
    Cognitum Software House Jena

    library:

    Python, C#, Java, Jena

    Turn your ideas into projects!

    Let’s talk about your project, just click the button and we will reach you as soon as possible.

    Contact Us