We present a visual analytics system for large-scale document retrieval tasks with high recall where any missing relevant documents can be critical. Our system utilizes a novel user-driven topic modeling called targeted topic modeling, a variant of nonnegative matrix factorization (NMF). Our system visualizes a topic summary in a treemap form and lets users keep relevant topics and incrementall...

Missing values imputation is often evaluated on some similarity measure between actual and imputed data. However, it may be more meaningful to evaluate downstream algorithm performance after than the itself. We describe a straightforward unsupervised algorithm, minimax approach based optimal recovery, derive probabilistic error bounds non-negative matrix factorization (NMF). Under certain geome...

Non-Negative Matrix Factorization (NMF) implies that a given nonnegative matrix is represented by a product of two non-negative matrices. In this paper, a factorization condition (consistent condition) on basis matrix is proposed firstly. For a given consistent basis matrix, although there exist infinite solutions (factorizations) generally, the sparse solution is unique with probability one, w...

Non-negative matrix factorization (NMF) has proven to be a useful decomposition technique for multivariate data, where the non-negativity constraint is necessary to have a meaningful physical interpretation. NMF reduces the dimensionality of non-negative data by decomposing it into two smaller non-negative factors with physical interpretation for class discovery. The NMF algorithm, however, ass...