Home

While working at GFaI, my main R&D duties were in the project AMENAMIN. In this project we developed a computer system to bring together ambient energy sensors and ambient intelligence systems for senior citizens in need of assistance. The main results were published in a paper titled "Towards ambient intelligent care and assistance systems for patients with dementia". More in detail, the project presents solutions for the following three research areas:

• • Personalized health care using energy harvesting technologies: Ambient intelligent care and assistance system for detection of dangerous events using self-sufficient fall sensors and fast aid for acute incontinence (in German)

  • Telemonitoring: Monitor system for eliminating bedsores using self-sufficient sensors placed under bed-posts (in German)

  • Telediagnostics: Automated analysis of the current mental state of patients with dementia using self-sufficient interactive devices (in German).

During my doctorate studies at the Humboldt University of Berlin I did some research in the topics distribution of wealth, investment strategies and portfolio optimization. In the first part of my PhD. thesis titled "Adaptive investment strategies for different scenarios" you can find some interesting findings, which led to the following papers:

• Investments in Random Environments (pre-print)

• Risk-Seeking versus Risk-Avoiding Investments in Noisy Periodic Environments (pre-print) 

The last part of my PhD. thesis shows a simple model to study coalition formation and statistical mechanics of complex networks, these investigations led to the following papers:

• The Investors Game: A model for Coalition Formation

• Formation of Common Investment Networks by Project Establishment between Agents  (pre-print).

My first interaction with data mining algorithms was during my master thesis where I applied different data visualization approaches and methods like association rules and probabilistic graphical models to analyze data. The following publications show some interesting findings on these topics:

• Interaction Graphs for Multivariate Binary Data

• Análisis de Accesos de Internet Mediante Modelos Gráficos Generalizados

At ASU, I had the opportunity to continue working on data mining, this time together with researchers from Intel Corporation. The project involved intensive programming tasks involving similarity measures, association rules, knowledge discovery and pattern recognition, leading to the following paper:

• Characterization and Analysis of Sales Data for the Semiconductor Market: an Expert System Approach.

It was during my master studies that I started to realize of the immense power of machine learning. My very first research work in the area of artificial intelligence was a student project called SIMAHO, where software agents had to find some food in a simulated environment. The success of this project motivated me to try other stochastic optimization methods for finding optimal paths and solutions to problems, leading to the following paper showing some interesting results in the field of portfolio optimization using genetic algorithms:

•  Adaptive investment strategies for periodic environments (pre-print)

Although I have not publish papers in the field of NLP, I have plenty professional practical experience in the field. For example, at acrolinx I worked on a project for sentence clustering and information retrieval of technical documentation. Moreover, at text & form I worked in a project for text retrieval from translation memories using N-grams indexing. Later, at moovel,  I improved search engines for digital maps focusing on solutions for mobile applications. Just recently I started to work with LLMs and Generative AI, a very promising and very fast-moving research and development area.

I started very early in my career to work in the field of modelling and simulation. My very first contribution to the field was my bachelor thesis 

• A programming language to specify stochastic Petri nets 

Moreover, the following contributions in the area of dynamical systems for weight change behavioral interventions have had much success:

• A dynamical model for describing behavioural interventions for weight loss and body composition change

• Dynamic energy-balance model predicting gestational weight gain.