Improved Target Data Identification for More Consistent Analysis

Published on September 10, 2026

ChatGPT Image 2026年8月7日 17_10_42
FactDecodeでは、Projectで選択した分析対象データ(Target)の扱いを見直し、分析時の判定をより一貫したものにしました。これにより、表示名や設定情報の状態に左右されにくくなり、選択した分析対象をより確実に識別したうえで分析を実行できるようになっています。

Consistent handling of the selected Target

FactDecode now identifies the target Dataset selected in a Project consistently throughout the analysis workflow.

The Dataset actually selected in the Project is used as the basis for identifying the Target. This keeps the Project configuration and the data used for analysis aligned throughout the workflow.

Target identification that is less dependent on display names

Dataset display names are useful for making data easier to recognize. However, the visible name of a Dataset should not determine which Dataset is treated as the analysis Target.

With this update, FactDecode identifies the Target based on the Dataset selected in the Project rather than its display name.

This makes Target identification less sensitive to name changes or older display information and helps keep the Target clearly distinguished from other Datasets and Factors used in the analysis.

Your existing Project workflow remains unchanged

This update does not change how you create Projects or select a Target.

You can continue to select the data you want to analyze, configure the Datasets, Factors, prediction horizon, and other settings, and then run your analysis as before.

The improvement works behind the scenes to make the handling of the selected Target more consistent while preserving the existing Project workflow.