Choose Up to 5 or 10 Factors, with Clearer Results

---

Choose How Many Factors to Include in Your Analysis

You can now choose whether your final model uses up to 5 or 10 factors in FactDecode's Design screen. The default setting is 10 factors.

Select 5 factors if you prefer a more focused model, or 10 factors if you want to consider a broader range of inputs. When you run an analysis, FactDecode uses your selected limit to guide factor selection and model training.

open_in_full

Improved Results Display Based on Model Quality

Depending on the available data and combination of factors, a model may not be able to identify meaningful factor contributions.

With this update, FactDecode no longer presents predictions or model-based charts as normal analysis results when the model cannot establish sufficient evidence for those outputs.

Instead, the analysis is marked as "Limited analysis." Available information, such as the analysis settings and data coverage, remains accessible so you can understand what was evaluated and why model-based results are unavailable.

Importantly, a model's predictions are not hidden simply because its predictive accuracy is low. When model training is valid, its results remain available regardless of how well the model performs.

Your Saved Analysis Results Remain Unchanged

This update does not modify or overwrite previously saved analysis results.

Changes to the factor limit apply to new analysis runs. Your existing results remain accessible through Run History.

Analyses marked as "Limited analysis" are also saved in your history, allowing you to revisit their settings and available data information later.