C-SALT: Mining Class-Specific ALTerations in Boolean Matrix Factorization
This work addresses a more complex class-dependency structure in datasets for data mining, though it appears incremental as it builds on existing rank-one factorizations.
The paper tackles the problem of identifying class-specific alterations in Boolean matrix factorization for labeled binary data, showing that the method can filter structure matching the model assumption and is justified in real-world applications.
Given labeled data represented by a binary matrix, we consider the task to derive a Boolean matrix factorization which identifies commonalities and specifications among the classes. While existing works focus on rank-one factorizations which are either specific or common to the classes, we derive class-specific alterations from common factorizations as well. Therewith, we broaden the applicability of our new method to datasets whose class-dependencies have a more complex structure. On the basis of synthetic and real-world datasets, we show on the one hand that our method is able to filter structure which corresponds to our model assumption, and on the other hand that our model assumption is justified in real-world application. Our method is parameter-free.