A method for detecting artificial and non-scientific texts in the collection of documents.
Oleg Yu. BakhteevMargarita KuznecovaAleksey RomanovYuriy ChehovichPublished in: Russ. Digit. Libr. J. (2017)
Keyphrases
- detection method
- significant improvement
- preprocessing
- similarity measure
- document collections
- high accuracy
- objective function
- clustering method
- dynamic programming
- information retrieval systems
- cost function
- relevant documents
- retrieval systems
- support vector machine
- database
- k means
- pairwise
- natural language
- neural network