Textual sentiment can be computed in many ways, and pooled into a time series in many more. The package treats that multiplicity as the problem to be solved rather than a nuisance: it computes sentiment, generates many candidate aggregation schemes, and selects among them within the prediction step.
See the package site.