Heat Maps

 Create 20 prompts designed to create a heat map of an image, text file, or a pdf file. These go beyond simple text extraction and ask for interpretation, comparison, and summarization of the data shown in charts, graphs,  photos, tables, infographics, etc.

 

Okay, here are 20 prompts designed to generate heatmaps based on interpretation, comparison, and summarization of data within images, text files, or PDFs. These prompts assume an AI capable of understanding context, performing analysis, and visualizing the results as a heatmap overlay.

Heatmap Prompts for Interpretation:

1. Input: Image of a complex scientific graph (e.g., multi-line plot with error bars).
   Prompt: "Analyze the provided graph image. Generate a heatmap where the color intensity over specific regions (lines, data points, shaded error areas) indicates the rate of change or volatility. Hotter colors (reds/oranges) should represent steeper slopes or wider error bands, signifying rapid change or higher uncertainty."

2. Input: PDF of a research paper.
   Prompt: "Process the provided research paper PDF. Create a heatmap overlaid on each page, highlighting sections based on the novelty or significance of the claims being made, as inferred from the surrounding text (e.g., phrases like 'novel finding', 'significant result', 'we demonstrate for the first time'). Hotter colors for stronger claims of novelty/significance."

3. Input: Image containing a table of financial results (e.g., quarterly earnings).
   Prompt: "Analyze the financial table in the image. Generate a heatmap directly on the table cells containing numerical values. The color should represent the magnitude of deviation from the column's average value (or a provided baseline, e.g., previous year's value if identifiable). Red for significantly above average, blue for significantly below."

4. Input: Text file containing customer reviews for a product.
   Prompt: "Analyze the sentiment expressed in the provided customer reviews text file. Generate a heatmap visualizing the sentiment intensity paragraph by paragraph (or sentence by sentence). Use a diverging color scale: strong positive sentiment in green, strong negative sentiment in red, neutral/mixed in yellow/white."

5. Input: Image of a photograph (e.g., a landscape, cityscape, or event).
   Prompt: "Interpret the provided photograph to identify areas of high visual complexity or detail density. Generate a heatmap overlay where hotter colors indicate regions with intricate textures, numerous distinct objects, or complex patterns."

6. Input: PDF of a legal contract.
   Prompt: "Scan the provided legal contract PDF. Generate a heatmap highlighting clauses or sections associated with potential risks, obligations, or liabilities for 'Party A' (assuming roles can be identified). Intensity should correlate with the inferred severity or importance of the risk/obligation."

7. Input: Image of an infographic explaining a process.
   Prompt: "Analyze the provided infographic image. Generate a heatmap focusing on the sections that represent potential bottlenecks or critical decision points within the depicted process. Hotter colors should indicate steps inferred to be more complex, time-consuming, or crucial based on annotations, icons, or flow."

Heatmap Prompts for Comparison:

1. Input: Image containing two similar bar charts side-by-side (e.g., comparing results before and after an intervention).
   Prompt: "Compare the two bar charts in the image. Generate a heatmap overlaid on the second chart. The color of each bar should represent the percentage change relative to the corresponding bar in the first chart. Green for positive change, red for negative change, intensity scaled by magnitude."

2. Input: PDF report containing data tables across multiple pages comparing different regions/products.
   Prompt: "Analyze the tables within the PDF report. Generate a heatmap specifically on the numerical cells representing 'Sales Performance'. Compare each cell's value against the overall average sales performance calculated across all comparable tables in the document. Highlight cells significantly outperforming (hot colors) or underperforming (cool colors) the global average."

3. Input: Text file containing two different versions of an article or draft.
   Prompt: "Compare the two versions of the text provided (assume they are concatenated or identifiable). Generate a heatmap on the second version indicating the degree of textual change from the first version on a paragraph level. Hotter colors for paragraphs with substantial edits (additions, deletions, significant rewrites), cooler colors for minor or no changes."

4. Input: Image containing a scatter plot with data points colored by category.
   Prompt: "Analyze the scatter plot image. For each category identified by color/legend, calculate its centroid. Generate a heatmap overlay on the plot area representing the density difference between two specific categories (e.g., 'Category A' vs. 'Category B'). Hotter colors where Category A is denser, cooler colors where Category B is denser."

5. Input: PDF document containing meeting minutes.
   Prompt: "Analyze the meeting minutes PDF. Identify action items assigned to different individuals or teams. Generate a heatmap highlighting sections discussing action items. Compare the number or apparent urgency of action items assigned to 'Team X' versus 'Team Y'. Use a diverging colormap where red indicates areas dominated by Team X's actions and blue indicates areas dominated by Team Y's actions."

Heatmap Prompts for Summarization:

1. Input: Long text file (e.g., a novel chapter or a lengthy article).
   Prompt: "Perform extractive summarization on the provided text file to identify the most important sentences. Generate a heatmap overlay on the text where the background color intensity of each sentence corresponds to its calculated importance score in