Suche

M.Sc. Dominik Schiller


profile_dominik_schiller

Wiss. Mitarbeiter

E-Mail: schiller{ @ }hcm-lab.de
Telefon: +49 821 598 - 2328
Raum: 2022
Hausanschrift: Universitätsstr. 6a
86159 Augsburg


Forschungsinteressen

Stichwörter: Machine Learning, Sentiment Analysis, Affective Computing

Publikationen

  • DrinkWatch: A Mobile Wellbeing Application Based on Interactive and Cooperative Machine Learning (Digital Health 18 DOI )
  • Design of a Knowledge-Based Agent as a Social Companion (Procedia Computer Science DOI)
  • Infected Phonemes: How a Cold Impairs Speech on a Phonetic Level (Interspeech 2017 - DOI)
  • Kristina: A knowledge-based virtual conversation agent (PAAMS 2017 DOI)
  • A multimodal annotation schema for non-verbal affective analysis in the health-care domain (MARMI' 16 DOI)
  • Modeling User’s Social Attitude in a Conversational System (Chapter in Human–Computer Interaction Series DOI)

Lehre

Abschlussarbeiten

Offen:

In Arbeit:

  • Understanding the inner workings of Deep Convolutional Neural Networks: A correlation analysis of machine learned and engineered features on speech related audio tasks
  • Empirische Betrachtung von Repräsentationsmöglichkeiten domänenspezifischer Textsegmente für maschinelle Lernverfahren im Bezug auf Emotionserkennung in sozialen Netzwerken

Abgeschlossen:

  • Comparative analysis of online machine learning methods for sentiment classification of social media data
  • Using Denoising Autoencoders to Improve Audio Signal Quality in an Online Environment
  • Automatic Detection of Novel Object Categories Using Convolutional Neural Networks