Who wrote ‘The World of Saddam Hussein?’ A supervised machine learning approach
Abstrak
Abstract The 2003 novel ‘The World of Saddam Hussein’ (TWOSH), published under the pseudonym Mahdy Haidar, has long concealed the true identity of its author. This study compiles the works of eight authors previously speculated by literary critics as potential authors. Using a supervised machine learning approach with a large set of classification algorithms, I employed two distinct feature sets: (1) top frequent words and (2) top free-occurring function word compounds lacking lexical attachments. My findings suggest that (1) the author of TWOSH can be discovered with high certainty and (2) the innovative method of utilizing top free-standing function word compounds proves nearly as effective as the traditional top N words approach, offering two key advantages: the exclusion of lexical items, providing a more accurate reflection of authorial style, and the elimination of the expensive and inaccurate morphological segmentation requirement.
Penulis (1)
Emad Mohamed
Akses Cepat
- Tahun Terbit
- 2024
- Bahasa
- en
- Sumber Database
- CrossRef
- DOI
- 10.1093/llc/fqae080
- Akses
- Open Access ✓