Brian Tvenstrup, Lindon Ventures
This session will walk you through how to use RapidMiner and Text Mining on customer service call transcripts. First, we will start with making sure we understand the business need for the model. Then we will use RapidMiner web mining operators to access call transcripts via API and develop ETL scripts to transform complex JSON objects into working documents and datasets. We will be using the text mining extension to complete all necessary text preprocessing and processing and perform weight-based feature selection on the text after word vectorization. We will build a predictive model, comparing multiple machine learning algorithms and selecting the best candidate to take through optimization.