DeloitteJunior Consultant
Jan. 2019 - Mar. 2019South AfricaTasks were to build models that performs Natural Language Processing (NLP), specifically sentiment analysis. All models were coded in Python and libraries, such as FastText, Scikit-Learn, TensorFlow and Keras, was used to build, train and deploy the model. An example of a model that was built is given a query from a customer, based on the context of the text the customer was advised on which department/division would be more equipped to resolve or attend to the query. Additionally, a team project was issued where we had to analyse data given by the customer to come up with an explanation as to why the customers’ company was performing poorly. In this project, tools for data analysis and data analytics was used to perform the required tasks. Such tools included Pandas, Numpy,, MySQL Server.