Ontology-based AI

MakoLab provides unparalleled expertise, innovation and experience in ontology-based, symbolic AI. Our dedicated, expert, twenty-person R&D team, led by industry veterans like our founder, Mirek Sopek, PhD, has been driving AI advancements since 2008. We focus on developing future products and services MakoLab investments and commercial projects.

How symbolic AI with ontologies boost businesses

Data interoperability

The seamless integration of diverse data sources.

Scalability

Growing your knowledge base without losing coherence.

Increased transparency

Clear understanding of AI decisions.

Enhanced accuracy

The implementation of logical rules to reduce errors and hallucinations.

Smoother compliance

Trouble-free adherence to regulatory requirements.

Advanced reasoning

AI systems’ capabilities of understanding and handling complex queries and insights are facilitated.

The power of symbolic AI with ontologies Symbolic AI leverages human-readable symbols and logical rules. This makes it a game-changer in terms of understanding and reasoning. Ontologies enable the structured knowledge representation that is crucial for efficient information retrieval and processing, whereas symbolic AI is transforming data management across industries. It is the route to effectively standardising your data, mapping relationships, managing knowledge and transferring it.

What we offer

MakoLab provides end-to-end solutions tailored to your industry and your company’s requirements.

Custom ontology development

Sector-specific ontologies.

Semantic search solutions

Enhanced search functionalities.

Financial data standardisation

Streamlined regulatory compliance information management.

Automotive data management

Improved vehicle data analysis and fraud detection.

Ontologies and other symbolic AI solutions have been pivotal to rekindling our belief in principles that have been deemed the canon of knowledge for 2,500 years. Now, we’re smoothly integrating those principles into modern machine learning solutions.

Mirek Sopek, founder of MakoLab

MakoLabʼs proven achievements. A selection of our ontology-based project successes

Search insights

This proprietary semantic search engine significantly improves search result relevance, making it easier for users to find exactly what they need. Renault is one of the notable MakoLab clients using this technology to enhance their search functionalities. The challenge here was improving the precision and usefulness of search results in large datasets.

Schema.org

Schema.org is a collaborative project that provides a collection of shared vocabularies for structuring data on the internet. It helps search engines understand the web page content better, improving search results and making the web more accessible. Under the leadership of Dr Mirosław Sopek, MakoLab has contributed significantly to the development of schema.org extensions, particularly for the automotive and financial industries.

DG FISMA Project

MakoLab developed a tool for the European Commission’s Directorate General for Financial Stability, Financial Services, and Capital Markets Union (DG FISMA). The tool uses machine learning to find definitions in EU documents and report obligations. This project highlights MakoLab’s ability to handle complex regulatory data and provide solutions that aid legal professionals with document retrieval and analysis. The challenge was to streamline the identification of the relevant regulatory information in extensive legislative documents.

Used Cars and Car Audio Ontology

MakoLab developed these ontologies in collaboration with Professor Martin Hepp, in order to support the semantic descriptions of automotive products and offers. These ontologies facilitate better data management and integration within the automotive industry. The challenge we addressed was organising and standardising diverse automotive data to enhance interoperability and data reuse.

Financial Industry Business Ontology (FIBO)

The Financial Industry Business Ontology (FIBO) is a comprehensive framework that standardiszes the terminology and concepts used in the financial industry. It covers a wide range of financial instruments, business entities and regulatory requirements, ensuring that financial data is consistent, accurate and interoperable across different systems and organisations. MakoLab’s involvement in FIBO focused on formalising these financial concepts in a machine-readable format.

Car damage detection

MakoLab has created image analysis models for detecting vehicle damage. Created for Polish automotive insurers, these models recognise the various parts of a car and assess damage on the basis of photographic data. This showcases the practical application of ontologies in the automotive insurance sector. Ontology was used to understand how the various parts of a car are related to each other, as in how to perform a specific operation on a particular part, like removing a door, for instance, and where a specific part is located, such as the ‘rear left door’ being situated towards the back of the vehicle on its left-hand side.

IDMP

The IDMP Ontology (IDMP-O) extends the ISO standards for the Identification of Medicinal Products (IDMP). The aim is to create a digital, machine-processable standard for ensuring semantic interoperability across organizations and regulatory jurisdictions. This helps not only in aligning pharmaceutical data standards globally, but also in enhancing drug safety, innovation and operational efficiency. MakoLab contributed to the development of the IDMP-O by enhancing the ISO standards for the IDMP. Our work focused on creating a digital, machine-processable standard to ensure semantic interoperability across organisations and regulatory jurisdictions.

J3016 (SAE)

The J3016 (SAE) ontology provides a taxonomy and definitions for driving automation systems, detailing six levels from no automation (Level 0) to full automation (Level 5) for on-road motor vehicles. Led by Dr Mirek Sopek, MakoLab played a significant role in transforming these definitions into a machine-readable format. This formalisation process involved creating a structured ontology that captures the relationships and attributes of each automation level precisely.

UrbanOnto

UrbanOnto is a project focused on the Historical Ontology of Urban Spaces. It aims to understand the evolution of urban areas through digital tools and frameworks and it involves creating detailed ontologies that map out the historical changes in city layouts and structures. MakoLab contributed to UrbanOnto by developing the digital tools and frameworks, helping to standardise and process historical survey maps and comparative studies on urban spaces.

Search insights