Unsupervised authorship attribution
This addresses authorship attribution for texts with unknown authors, but it appears incremental as it builds on clustering methods without claiming major breakthroughs.
The paper tackles the problem of attributing parts of a written text to unknown authors without prior knowledge of writing styles, using multiple independent clusterings to identify similar and dissimilar text segments, and demonstrates results on texts with multiple writing styles.
We describe a technique for attributing parts of a written text to a set of unknown authors. Nothing is assumed to be known a priori about the writing styles of potential authors. We use multiple independent clusterings of an input text to identify parts that are similar and dissimilar to one another. We describe algorithms necessary to combine the multiple clusterings into a meaningful output. We show results of the application of the technique on texts having multiple writing styles.