Dear Madam or Sir,
I have great experience in Python, R, Java, C++, Matlaband Scala for natural language processing and machine learning. Moreover, as a mathematician and software developer I am familiar both with the theoretical background as well as the practical aspects of the implementation. I have implemented many machine learning algorithms from scratch in Python and Matlab to get a deep understanding.
I have great experience in Pythons module nltik for natural language processing. Should this module be used or should it be done from scratch using grammar rules/stemming rules from external files? Do you have any preferences for the classifier (e.g. simple classic classifiers like naive bayes, decision trees with boosting/bagging, preprocessing with PCA, LDA, and ISOMAP)? Do you have any specifications for the features (like classic BoW, n-gramms,...) or should the be learned from the data?
I am looking forward to discuss the detail.
Best regards