* Output should be saved as a .pickle file
* I should be able to input the percentage of documents used for testing and training
* The classifier can be NaiveBayes
* I want to be able to use Ngrams
* I want some way to control noise (such as minimum information gain)
* The classifier should output the most informative features as well as the accuracy rate
* The classifier should be capable of dealing with large datasets (over 350 MB and 100,000 text files)
I can only pay for results. You will show me screenshots of your progress and I will then release a milestone.
Please get back to me with budget and deadline.
13 freelance font une offre moyenne de £726 pour ce travail
I will guarantee delivery within 10 days with proof of work prior to payment for £1000. And you'll be able to know that it's finally done, and you can put an end to the ever growing costs and weeks.
Hello! We are software developers and have strong experience with similar projects, we would love to work on it , lets discus it further via chat Thanks Kind regards Robin
Hi, I have experience with statistics and data mining , using NLTK, Weka. I have published some papers in the area ([url removed, login to view]). Could you send me more details about this project?