fix: handle two-claim batches and normalise features in fraud detection - #440
Open
amberly-d wants to merge 2 commits into
Open
fix: handle two-claim batches and normalise features in fraud detection#440amberly-d wants to merge 2 commits into
amberly-d wants to merge 2 commits into
Conversation
detect_fraud crashed on two-claim batches because LOF requires n_neighbors < n_samples, and the feature matrix mixed unscaled LabelEncoder integers with raw token amounts so Euclidean distance was dominated by the largest-range column. Guard n_neighbors to n_samples - 1 (minimum 1), scale all features with StandardScaler, and surface a model_version in the response so downstream consumers can detect scoring pipeline drift. Closes ChainForgee#432
StandardScaler inverts relative distances when most claims share identical features (e.g. same IP), causing LOF to score the homogeneous cluster as more anomalous than actual outliers. Reverting to raw numeric features preserves the correct outlier signal while keeping the small-batch n_neighbors fix.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
detect_fraud crashed on two-claim batches because LOF requires n_neighbors < n_samples, and the feature matrix mixed unscaled LabelEncoder integers with raw token amounts so Euclidean distance was dominated by the largest-range column.
Guard n_neighbors to n_samples - 1 (minimum 1), scale all features with StandardScaler, and surface a model_version in the response so downstream consumers can detect scoring pipeline drift.
Closes #432
Testing
Checklist