Text classification software (link)
Pre-processing building a .sp file, namely a processing chain.
e.g.: [ORIGINAL] Bob is in the kitchen. I don't care at all!!!
[AFTER PROCESSING] Bob_NP Bob_NP_be_VBZ be_VBZ be_VBZ_in_IN ...
Dictionary computation building a dictionary from a .sp file and a corpus of documents
Data Processing building a numerical corpus from a .sp file, a dictionary and a corpus of documents. All documents are projected to build a bag of words.
Learning step learning (and saving) a model associated to a labeled set.
Testing a model Test a model in console mode
In all above programs, the existing files are not replaced. If a file exists, the associated computation is not done.
- Download and unpack the file in a directory
- Launch all scripts