Sentiment analysis can used to gauge the effectiveness of a campaign. However, combining all the tasks necessary to perform this can be challenging. Read how a variety of MathWorks toolboxes for data import, processing and visualising were used to create sentiment analysis model.
Timeseries Forecasting uses historical data to predict future responses and is a powerful technique for predicting success in business. Time-series Regression in the Econometrics Toolbox is a technique that can be used to forecast and predict future values. This regression technique is often more reliable than classical linear models. A Time-series regression is a model made up of the relevant predictors, a constant and a random disturbance. These models can be used to forecast performance metrics and enable confident decision making.
Alluvial flow diagrams can be very useful in analysing how large categorised timeseries data sets change over time. This can be used as a quick analysis technique to visualise the change in customer behaviour and product adoption over time. After this visualisation has been performed, additional techniques can be employed for further detailed analysis.
NLE, a multi-utility provider used MATLAB to create algorithms that could predict customer behavior. These algorithms were then deployed to the cloud where massive amounts of data could be processed. Using MATLAB in the cloud enabled NLE to get answers quickly, enabling faster decision making.