Open Moussed opened 9 years ago
cool! but i think it's hard to make...
i'm in italy, and i have some problem to get NFC-reader, when i got it, i try to help for this feature, because it's really nice idea.
thank you ) it's not really hard because in machine learning there are already functions to do this kind of classification ;) What is difficult is to organize nfc data by column in a csv file, add in the same line the app measure and reader mesure. With hundreds of lines it will be easier for the model to make good predictions Look a this python library that can be used for this problem. http://scikit-learn.org/stable/tutorial/statistical_inference/supervised_learning.html
Can we get something like a line of data corresponding to each Nfc measure ? If we can put in a csv file each Nfc mesure with hits correspondonding nfc data and the real value(mesured by the reader) in a line, we can then create a machine learning (data mining) model that reconize pattern in this dataset.Then this model will be able to predict what real value corresponds to a new line of Nfc data .(the prediction of the model is trained on the historical data )
I hope you will understand me .
We just need a csv file with the dataset descripted before. I will then use it to try to create this kind of model. I don't know how to get the Nfc meseaure and the corresponding nfc data from the app .if some one can explain me please :)i would like to save them in a dataset(csv file) for each measure .