In the Topics Report, Loop uses automatic sentiment analysis and Natural Language Processing (NLP) techniques to display a graph of topics based on Tags that you have configured with keywords and auto-tagging. You can add additional Tags and configure auto-tagging terms at any time to add a bar to this report. Each bar in the graph represents a specific Tag and shows the total number of messages broken down by sentiment.
The display options allows you to change the way the data is presented.
Sentiment:
Stacked – Shows the total number of messages while separating the bad into segments for each sentiment. These can be one of 5 sentiments: Very Angry, Angry, Neutral, Happy, and Very Happy. The sentiment is determined by Loop’s natural language processing algorithm.
Average – Show the average sentiment for the Tag.
Bar – Shows a separate bar for each sentiment.
Sort by:
Tag Names – Sorts the bars alphabetically by Tag name.
Counts – Sorts the bars by quantity of associated messages, from high to low.
Tag Temperature – Sorts the bars by average sentiment, from Very Happy to Very Angry.
Show:
Count – Show the Y-axis as the number of messages.
Percentages – Show the Y-axis as the percentage of total messages.
Clicking on a bar or segment displays the messages associated with the Tag and Sentiment. This works the same way as the conversation view in the Trending Topics report.