This assignment presents the design, investigation, and findings of machine learning pipelines for text and time series datasets. Each sample in the text dataset corresponds to a Twitter 'tweet' with a unique identification number and sentiment label. The time series dataset provides one-minute interval readings on six weather features from a Raspberry Pi computer in Newcastle between November 2021 and December 2022.
Each dataset has a separate project objective. The text dataset(s) objective is to apply a combination of different approaches to predict the sentiment of the test set and improve the prediction results. The time series dataset objective is to try to predict future values 5, 10, 15 and 30 minutes into the future as well as 1, 2, 6 and 12 hours.