Introduction
QInsights currently offers sentiment analysis for semi-structured data in Excel format, making it ideal for open-ended survey responses, customer feedback, or social media comments. If you want to conduct a sentiment analysis for text or PDF documents, you can still do this by using a prompt in Conversational Analysis. See below for an example prompt you can use.
Traditional sentiment analysis often categorizes feedback as positive, neutral, or negative, but with the power of generative AI, it becomes much more dynamic. QInsights allows you to define custom dimensions that align with your data, such as "satisfied" vs. "dissatisfied" or "rational" vs. "emotional". Additionally, you can refine your analysis by focusing on specific topics and engaging interactively through follow-up questions, also making sentiment analysis a collaborative process between human researcher and AI-assistant.
How it Works
1. Select Your Data
- Select an Excel file for Analysis.
- Next, choose the column you want to analyse. Sentiment analysis evaluates data case by case (= row by row).
2. Define Dimensions
Use standard sentiment categories, such as positive, neutral, and negative, or customize dimensions based on your needs.
Examples include:
- Satisfied, neutral, dissatisfied
- Rational, emotional
- Planned, impulsive
- Supportive, critical, neutral
3. Focus the Analysis
- Narrow the scope to a specific topic, especially useful for datasets like social media comments or product feedback. Examples include:
• Social media: "Classify responses that directly refer to the post’s content as like, neutral or dislike.”
• Customer feedback: "Disappointment or satisfaction with regard to usability"
4. Review Results
- QInsights produces a pie chart showing the distribution of your defined dimensions.
- Hover over the chart to see the percentage distribution, or prompt Q for a detailed list of dimensions and their corresponding distributions.
Example Prompt: "List all dimensions with their percentage distribution."
A summary of the analysis is displayed below the pie chart, written by Q, your AI assistant. This may provide sufficient insights for your needs, but you can always ask follow-up questions to delve deeper.
You can export the detailed evaluation of each case as an Excel file for further review or reporting.
Best Practices for Effective Sentiment Analysis
- Focus on One Column at a Time
- If your dataset includes separate questions for "benefits" and "challenges," analyze these columns individually to avoid conflicting results.
- Keep Prompts Simple
Instead of overly complex or multi-layered queries, focus on straightforward questions:
- rational vs. emotional regarding product design
- Positive, neutral, or negative with regard to pricing
Avoid stacking too many conditions in one prompt, like:
• ❌ Group responses first by similarity, then analyse for rational vs. emotional reasons.
Results like the following might be interesting for smaller data sets, but become quickly difficult to read:
--> If a pie chart like this was generated, simplify your prompt.
Tailor Dimensions to Your Needs
Examples
- Happy, frustrated (e.g. for customer feedback)
- Supportive, critical, neutral (e.g. for evaluating user responses to a policy.)
- Disappointed, neutral, excited
Sentiment Analysis for Word / PDF documents
You can use the following prompt in the Conversational Analysis if you want to conduct a sentiment analysis for Word or PDF files, e.g. for an interview, a focus group or a document.
Here is an example prompt. Refine to fit your specific needs:
Perform a sentiment analysis on the following [document Name]:
1. Identify statements or segments that convey a clear sentiment (positive, negative, or neutral).
2. Tally the number of statements for each sentiment category.
3. Summarize these sentiments, providing relevant quotes or short excerpts that illustrate each category.
4. Highlight any recurring themes or patterns.
5. Conclude with an overall summary and key insights.
Output Format (follow these exact instructions):
1. Sentiment Breakdown
- Present the counts of positive, negative, and neutral statements.
- After this section, insert one blank line.
2. Key Themes
- List the main topics discussed and the prevailing sentiment for each.
- After this section, insert one blank line.
3. Representative Quotes
- Provide example quotes or excerpts illustrating each sentiment category.
- After this section, insert one blank line.
4. Overall Summary & Insights
- Provide a concise narrative capturing the sentiment trends and their potential implications.
Be sure to use bold characters for section headers and leave one blank line between each section.