Inductive Analysis

Inducive analysis casts a wide net across contents in your data and forms a two-level category scheme based on the extracted insights.

Step 1: Define the high-level insight type you want to extract

Skimle goes through your documents to form some suggestions of potential high-level categories (called insight types in Skimle). Often, you want to define exactly what to extract from the corpus by adding your own insight type.

Steps 2 & 3: Data extraction and categorization

Skimle goes through each of your documents and extracts insights based on the user-defined categories and AI instructions. It then forms a two-level category hierarchy and assigns insights to the best matching category.

Note: Subcategories are only formed if the category has a meaningful number of insights assigned to it.

The above example shows the category hierarchy for thge inductive insight type "Cynicism". Note that only one of the categories has further subcategories; the other categories have too few insights to form meaningful subcategories.

Step 4: Developing category summaries

After the data has been categorised, Skimle automatically forms a summary for each category. The category summaries are visible in the categories view.