Find Sentiment of Discord Community

The Sentiment Analysis feature in Blaze leverages Machine Learning models to automatically detect the emotional tone of messages within your Discord community. This gives community managers and marketers a powerful tool to monitor community health, engagement trends, and emerging concerns over time.

How it Works

We deploy natural language processing (NLP) and ML algorithms to analyze the content of messages and categorize them into positive, negative, or neutral sentiment buckets. This analysis is performed over selected historical timeframes to identify sentiment shifts and conversation trends.

Word Cloud

The Word Cloud visualizes the most frequently discussed topics in your Discord community over the last 30 days. These topics are extracted from conversations and tagged based on the theme (e.g., pricing, product_update, scam, rewards).

Interpretation:

• Larger words = Higher frequency of mentions

• Topics relate to both user feedback and support issues

• Useful for spotting trending concerns or praises

Click into these positive or negative panels to see past messages sorted with the most positive or negative messages first.

Sentiment Breakdown Metrics

For any selected period, Blaze provides a breakdown of:

• Positive Messages Weighted sum of positive Discord messages. Includes percent change across two halves of the time period (e.g., increase/decrease in positivity).

• Negative Messages Weighted sum of negative messages, including percent change across two halves of the period. A spike may indicate dissatisfaction or issues.

• Neutral Messages Weighted sum of neutral statements (informational or passive messages) and their change over time.

Average Sentiment Over Time:

Tracks the community’s emotional tone over time, scored between -1 (negative) and +1 (positive). Averages near 0 suggest neutral sentiment. Campaign overlays can be added for contextual insights.

Example from dashboard: Current average sentiment is -0.01, indicating slightly negative to neutral community tone.

Average sentiment over time for any channel:

See average sentiment over time for any particular channel on the top chart, and see the total volume of messages being sent in that channel on the bottom chart. Showing the volume helps you get a sense for whether the average sentiment score is representative and based on enough data.

Volume and sentiment breakdown of all messages:

Volume of positive, negative and neutral messages over time. This gives you a sense of the breakdown of types of messages over time.

You can change the time period by clicking the 3 dots icon on the top right of tabs with charts.