This project is my second take on the “Real or Not ? NLP with Disaster Tweets” Kaggle competition, in which we are provided a list of tweets that can either be about a disaster or not. You can find my first attempt here Last time, I was rather new to data science. I focused heavily on feature engineering and bags of words to put into machine learning models and managed to obtain a decent score of 78% with ensemble models.

This project’s goal is to build a model that can classify consumer goods based on pictures and descriptions. I approached this project with basic and advanced methods for both NLP and image recognition. For basic methods, I used feature extractions and clustering, whereas for advanced methods I used supervised neural networks. Moreover, the dataset we were provided with was small, which made it necessary to either use dimension reduction for clustering algorithms or transfer learning for neural networks.

This project is my take on the “Real or Not ? NLP with Disaster Tweets” Kaggle competition, in which we are provided a list of tweets that can either be about a disaster or not. All of these tweets have been detected based on keywords such as “ablaze”, and it’s up to the competitor to build a model that determines wheter these words were used in the context of an actual event.

In this project, my goal was to familiarize myself with natural language processing and use it to build a “next word” predictive model such as the ones we have on our smartphones keyboards. For that, I used a database provided by SwiftKey. I divided this project into three parts : the App (hosted on ShinyApps), the project pitch (hosted on RPubs), and the GitHub repository that contains the App’s code.