Qualitative Evaluation of Text for Cleaner Knowledge Base
Created 2 preprocessing models, that can be used by any UNIX and DOS system.
Also developed an 'understandability' metric.
Features need to be loaded into a cloud based chatbot model, of any choosing.
The goal of this project was to familiarize myself with NLP practices, using spacy and CoreNLP. I was able to learn application of various functions within these NLP toolkits. Also researched Squad 2.0 dataset to help develop features and took a great deal of time reading about BERT.