To become operational, data science applications require an efficient workflow for data access, processing, and storage that preserves data value and remains unaffected by the format or volume of the data. The research area in Data Engineering, Analysis, and Processing focuses on creating practical data science applications by ensuring their functionality with a dual focus: operationally, on the interfaces and mechanisms for data flow and access; scientifically, on data analysis, modeling, processing, and the presentation and interpretation of results.
Data Engineering, Analysis and Processing - ISIN
Research area in
Data Engineering, Analysis and Processing
This scientific sector aims to explore and analyze the human mental complex system to build models of human behavior. These models are used for a deeper understanding and interpretation of the human mind and mental states, for characterizing mental conditions such as typical, atypical, anomalous, or disordered, and for predicting and simulating human decision-making processes including recommendation systems, behavior inference, and service personalization. The analysis considers various factors that affect the mental system such as the environment and social interactions, and examines both observable outputs like body movements and posture, facial expressions, activities, mobility, verbal communication, and less evident ones like heart rate, electrodermal activity, and emotions.
Expertise: human behavioral sensing and modeling, recommendation systems, personalization services, sentiment analysis and emotion recognition, affective computing.
Expertise: human behavioral sensing and modeling, recommendation systems, personalization services, sentiment analysis and emotion recognition, affective computing.
Processing Big Data presents significant challenges, starting with designing an immutable, reliable, and continuously growing data storage system, and extending to architecting scalable applications and elastic data pipelines to process the data. The Big Data Processing scientific sector has a dual focus: developing elastic architectures for storing and retrieving large amounts of data, and creating data pipelines for analyzing and processing data using machine learning and data mining methodologies. These solutions leverage centralized, decentralized, and federated approaches. Additionally, this sector aims to build architectures that facilitate the visualization and management of data throughout the data processing pipeline.
Expertise: infrastructures for big data analysis and processing, data lakes, data meshes and data spaces, lambda architectures, unsupervised methodologies for topic extraction, semi-supervised modeling, collaborative learning.
Expertise: infrastructures for big data analysis and processing, data lakes, data meshes and data spaces, lambda architectures, unsupervised methodologies for topic extraction, semi-supervised modeling, collaborative learning.
The FINVA (Financial Intelligent Virtual Assistant) platform, developed in collaboration with and , has been recognized as an Innosuisse Success Story. The project's goal is to support financial institutions in reducing costs and improving compliance strategies.
Our research area boasts state-of-the-art infrastructures to support cutting-edge research in data engineering, analysis, and processing. These infrastructures include:
- Advanced computing facilities for large-scale data processing
- High-performance data storage systems
- Specialized laboratories for human behavioral analysis
- Integrated platforms for data modeling, analysis, and visualization