Change Details

See https://en.wikipedia.org/wiki/Stochastic_context-free_grammar Implement a feature that, when given a set of sentences, produces a likelihood ratio that represents how likely it is that a sentence is to be generated from a subset of a corpus (e.g. "vandalism" or "spam" vs. non-vandalism and non-spam). Review for suitability: * https://github.com/aetilley/pcfg * https://www.cs.bgu.ac.il/~elhadad/nlp11/nltk-pcfg.html#pcfg

See https://en.wikipedia.org/wiki/Stochastic_context-free_grammar Implement a feature that, when given a set of sentences, produces a likelihood ratio that represents how likely it is that a sentence is to be generated from a subset of a corpus (e.g. "vandalism" or "spam" vs. non-vandalism and non-spam). Review for suitability: * https://github.com/aetilley/pcfg * https://www.cs.bgu.ac.il/~elhadad/nlp11/nltk-pcfg.html#pcfg Parsing: * https://spacy.io/ (Doesn't make CNF trees) * https://github.com/BLLIP/bllip-parser (Broken for Python 3.5) * http://nlp.stanford.edu/software/lex-parser.shtml

See https://en.wikipedia.org/wiki/Stochastic_context-free_grammar Implement a feature that, when given a set of sentences, produces a likelihood ratio that represents how likely it is that a sentence is to be generated from a subset of a corpus (e.g. "vandalism" or "spam" vs. non-vandalism and non-spam). Review for suitability: * https://github.com/aetilley/pcfg * https://www.cs.bgu.ac.il/~elhadad/nlp11/nltk-pcfg.html#pcfg Parsing: * https://spacy.io/ (Doesn't make CNF trees) * https://github.com/BLLIP/bllip-parser (Broken for Python 3.5) * http://nlp.stanford.edu/software/lex-parser.shtml