Cloud AI Service Engineering platform for development of universal intelligent services dedicated to various areas of applications - Project - Bridge of Knowledge

Search

Cloud AI Service Engineering platform for development of universal intelligent services dedicated to various areas of applications

The project will develop and implement the CAISE cloud platform destined for the domestic market, which will enable SMEs to effectively develop and deploy intelligent cloud services and applications. The platform will be based on TASKcloud, a cloud computing facility being developed at CI TASK for several years. TASKcloud will be expanded with further open source software, creating an execution environment for the planned CAISE platform. The DevOps methodology and CI/CD techniques will be used to develop the cloud and build the platform, enabling its continuous development during the project period. The CAISE platform will offer two categories of base services, for management and production. The first one concerns the management of clients, data, models and services, while the second will create an intelligent environment for the production of base and sectoral services. Services will be provided for content analysis of digital text documents edited in Polish, for image processing, as well as hybrid document content composed of texts and images. Thanks to the use of high performace computing (HPC) resources, new solutions related to deep learning methods will be used, including federated learning and Big Data, providing appropriate data for machine learning. The project will involve close collaboration with four EU companies (signed letters of intent in the field of creating multicloud solutions have already been signed) and three Polish SMEs acting as the project subcontractors verifying the functioning of the platform. Such verification will ensure that the platform achieves the highest technological readiness level and will be effectively used in at least three sectors of public administration: justice, health care and education. This will be an original implementation of a cloud-based AI solution constituting a significant product innovation. Integration of DevOps and CI/CD techniques and incorporation of quality management procedures in these processes will lead to a process innovation. In turn, the possibility of using the CAISE platform in other public sectors, such as security, energy or logistics, will prove the universality of the offered solution.

Details

Project's acronym:
CAISE
Financial Program Name:
IPCEI CIS (Cyberbezpieczeństwo – CyberPL)
Organization:
Narodowe Centrum Badań i Rozwoju (NCBR) (The National Centre for Research and Development)
Agreement:
KPOD.05.10-IW.10-0005/24 z dnia 2024-07-23
Realisation period:
2024-03-05 - 2026-08-31
Project manager:
prof. dr hab. inż. Henryk Krawczyk
Realised in:
Dział Usług Chmurowych
Request type:
National Reconstruction Plan
Domestic:
Domestic project
Verified by:
Gdańsk University of Technology

Filters

total: 2

  • Category

  • Year

  • Options

clear Chosen catalog filters disabled

Catalog Projects

Year 2024

  • Enhancing Word Embeddings for Improved Semantic Alignment

    This study introduces a method for the improvement of word vectors, addressing the limitations of traditional approaches like Word2Vec or GloVe through introducing into embeddings richer semantic properties. Our approach leverages supervised learning methods, with shifts in vectors in the representation space enhancing the quality of word embeddings. This ensures better alignment with semantic reference resources, such as WordNet....

    Full text to download in external service

  • Toward Sustainable Development: Exploring the Value and Benefits of Digital Twins
    Publication

    - Telecom - Year 2024

    The complexity and number of data streams generated by internal processes exceed the capabilities of most current simulation environments. Consequently, there is a need for the development of more advanced solutions that can handle any number of simultaneous simulations. One of the most promising ideas to address these and other challenges is the concept of a Digital Twin (DT), which refers to a digital representation or a virtual...

    Full text to download in external service

seen 148 times