Abstract
In order to be useful, visualizations need to be interpretable. This paper uses a user-based approach to combine and assess quality measures in order to better model user preferences. Results show that cluster separability measures are outperformed by a neighborhood conservation measure, even though the former are usually considered as intuitively representative of user motives. Moreover, combining measures, as opposed to using a single measure, further improves prediction performances.
Original language | English |
---|---|
Title of host publication | NIPS Workshop on Interpretable Machine Learning in Complex Systems |
Place of Publication | Barcelona |
Publication status | Published - 2016 |
Event | Thirtieth Conference on Neural Information Processing Systems - Sagrada Familia, Barcelonne, Spain Duration: 5 Dec 2016 → 10 Dec 2016 |
Symposium
Symposium | Thirtieth Conference on Neural Information Processing Systems |
---|---|
Abbreviated title | NIPS 2016 (NeurIPS) |
Country/Territory | Spain |
City | Barcelonne |
Period | 5/12/16 → 10/12/16 |
Fingerprint
Dive into the research topics of 'Learning Interpretability for Visualizations using Adapted Cox Models through a User Experiment'. Together they form a unique fingerprint.Student theses
-
Interpretability and Explainability in Machine Learning and their Application to Nonlinear Dimensionality Reduction
Author: Bibal, A., 16 Nov 2020Supervisor: FRENAY, B. (Supervisor), VANHOOF, W. (President), Cleve, A. (Jury), Dumas, B. (Jury), Lee, J. A. (External person) (Jury) & Galarraga, L. A. (External person) (Jury)
Student thesis: Doc types › Doctor of Sciences
File