Compare obtained results

[bcdasilv]

Compare the results obtained from running the sentiment analysis tools/models.
This data analysis work should be reported in text format supported by tables and charts (e.g. bar plots).

You may notice the need to re-run some sentiment classification using some or all models/tools. That's common to come across a mistake after measuring all the data in the dataset and discover that you need to re-run part or all analysis. For example, you may find out you passed a wrong parameter, or you skipped an important step, or you filtered out valuable data.