Human-Centered Interactive Visual Data Analysis

"The purpose of computing is insight, not numbers."
– Richard Hamming

This PhD position focuses on the human-centered perspective on interactive visual data analysis. As such, interested candidates will not aim for the very last percents of accuracy of ML models but rather focus on human aspects in data science.

Possible directions include

• Human-Centered AI (HCAI)
• Personalized Similarity Search
• Personalized Training of ML models
• Personalized Recommender Systems
• Subjectivity in Data Labeling
• Human Factors in Data Science
• People-Oriented Visualization Analysis Design

Personalized Visual-Interactive Music Classification (link)

Overview of our system for the personalized visual classification of music collections. Upper left: querying and browsing interface for all songs. Lower left: class creation, labeling, and class prediction interface. Upper right: analysis of features and class distributions. Center right: filtering interface for classes and class collisions. Lower right: two interfaces for the meaningful selection of labeling candidates (based on dimensionality reduction and active learning). Throughout this work, we apply a test dataset for genre classification. Note in our examples, we give the true class label in [brackets] in the song names for verification

For general information about the position, please see the hub page for open PhD positions