Interpretable and reconfigurable clustering of document datasets by deriving word-based rules

摘要: Clusters of text documents output by clustering algorithms are often hard to interpret. We describe motivating real-world scenarios that necessitate reconfigurability and high interpretability clusters outline the problem generating clusterings with interpretable reconfigurable cluster models. develop a algorithm toward outlined goal building models; it works rules disjunctions conditions on frequencies words, decide membership document cluster. Each is comprised precisely set satisfy corresponding rule. show our approach outperforms unsupervised decision tree huge margins. purity f-measure losses achieve as little 5% 3% respectively using approach.