Hi there! I'm
Fabian Roth

Researcher & Data Scientist

creates advanced data models and analytical tools to extract valuable insights from large datasets. Specializes in machine learning and deep learning to develop innovative, accurate, and efficient solutions, with a focus on scalability and usability.

Years of Experience in Programming
+ 0
Years of Experience in Data Science
+ 0
Years of Experience in Lecturing
+ 0
Years of Experience in Deep Learning
+ 0

Biography

Currently, I’m a Researcher and PhD student at the Chair of Anthropogenic Material Cycles (ANTS) at RWTH Aachen University with a focus on sensor-based real-time material flow characterization. I began my journey as a student of Business Informatics in Data Science and Data Analytics at Trier University of Applied Sciences, where I gained strong knowledge in programming and machine learning. This interest has driven me to further research in deep learning, particularly in real-time object detection and instance segmentation. Before my PhD, I worked as a Researcher at German Research Center for Artificial Intelligence (DFKI) and Saarland University, where I developed an AI-based anomaly detection system using drones. With a strong background in applied AI, I am driven by the challenge of optimizing industrial processes. I aim to develop cutting-edge sensor-based AI solutions that enhance efficiency and sustainability in waste management and recycling by leveraging AI-driven image analysis for improved material characterization in sorting processes.

My passion

Data Science & Data Analysis

Cutting-edge object detection & instance segmentation

Deep Learning & Machine Learning

AI-driven material characterization & classification

Main Focus

My expertise covers the full pipeline from sensor-based data acquisition and preprocessing to feature extraction, real-time object detection, and material classification using AI.

01

undraw_video-files_cxl9

Data Acquisition

Collecting high-quality sensor data to enable precise material characterization.

02

undraw_data-processing_z2q6

Data Preprocessing & Feature Extraction

Cleaning, enhancing, and transforming raw data for AI-based analysis.

03

detection

Object Detection & Instance Segmentation

Using deep learning to identify different materials in dynamic environments.

04

undraw_algorithm-execution_rksm

AI-Driven Process Optimization

Extracting valuable insights from characterized materials to enhance decision-making.

Working Experience

Scientific Employee

  • Member of the research group Sensor Technology & Data Science
  • Research in the field of material science with focus on Sensor Technology & Data Science by using Deep Learning and Machine Learning for sensor-based material flow characterization
  • Lecture „Sensor Technology in Resource Management“
  • Supervision of bachelor & master thesis
  • Member of the CYA Steering Committee for Young Academics

Researcher

German Research Center for Artificial Intelligence (DFKI),
Mar 2023 – Aug 2023 | Saarbrücken

Junior Researcher

  • Chair of Business Administration, in particular Information Systems in the Service Sector (Information and Service Systems)

Student Assistant

German Research Center for Artificial Intelligence (DFKI),
Mar 2022 – Sep 2022 | Saarbrücken

  • Within the ’self-optimization‘ use case in the project SPAICER (funded by the Federal Ministry of Economic Affairs and Climate Action (BMWK) (funding reference 01MK20015A)) as a smart resilience service, it enables material analysis to gain insight into quality using a non-destructive testing method. This allows recommendations to be made for parameter optimization.
  • By leveraging an self-implemented application, object detection and an UAV with an infrared area camera the system is able to predict anomalies.
  • Grade 1.0

Student representative on the audit committee

Trier University of Applied Sciences,
Jan 2021 – Dec 2022 | Trier

Research Assistant

Trier University of Applied Sciences,
Oct 2019 – Feb 2022 | Trier

  • Advising and supporting students in their learning process
  • Lectures: ‘Fundamentals of Programming’,
                     ‘Programming’,
                     ‘Data Analysis with Python‘, and
                     ‚Networks‘
  • In addition to conducting the lectures themselves, this included the conception and creation of the lecture materials, the exercises as well as the creation and correction of the exams.
  • Conception and realization of the pre-courses ‘Fundamentals of Programming’ and ‘Programming’ as preparation for students for the upcoming lecture
  • Organizing tutor workshops and contributing to the development of tutor training in the Department of Economics.
  • Planning and realization of workshops as part of Girls‘ Day in the field of Data Analysis / Data Science for interested female pupils (career orientation project for girls)
  • Supporting the Department’s public relations work by attending trade fairs, City Campus and Open Day.
  • Participation in the introductory programme for first-year students („Study Buddy Programme“)

Student Assistant

Trier University of Applied Sciences,
Sep 2017 – Sep 2019 | Trier

Working Student, Project Management

IT-Haus GmbH,
Jul 2018 – Sep 2018 | Föhren

Education

PhD Student in natural sciences

Master of Science, Business Informatics

Trier University of Applied Sciences,
2019 – 2022 | Trier

Bachelor of Science, Business Informatics

Trier University of Applied Sciences,
2016 – 2019 | Trier

Apprenticeship, IT specialist for system integration

IT-Haus GmbH,
2013 – 2016 | Föhren

Vocational diploma

Higher vocational school for it systems,
2011 – 2013 | Wittlich

School-based apprenticeship, State certified assistant for it systems

Higher vocational school for it systems,
2011 – 2013 | Wittlich

My latest publications

Fabian Roth et al.

26 March 2025

Automated data acquisition method for sensor-based real-time material flow characterization of recyclable waste streams using sensor fusion: A case study

This work contributes to novel data acquisition methods for high resolution RGB area images of SBRTC applications to effectively address the challenges of noisy and biased real-world datasets making it easier to perform data splitting.

Get In Touch

I look forward to exciting discussions, feel free to reach me out!

Contact Info

Email

info@fabianroth.com

Contact me on LinkedIn