Anomaly Detection for Survey Scores Using ML Prediction Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems lack the capability to efficiently detect anomalies in customer survey data using machine learning, relying on manual analysis to differentiate between real changes and regular day-to-day variations.
Innovation Solution
A system configured to receive historic survey score data, generate training data, and train prediction models to forecast expected survey scores. It compares actual scores to predicted scores, using user-defined confidence levels and standard deviation augmentations to determine if scores are anomalous.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual analysis is used to differentiate between real changes and regular day-to-day variations, then companies can identify anomalous survey scores, but the process is time-consuming and inefficient
Solution Approach 1:
The patent replaces manual mechanical analysis with an automated machine learning system that uses prediction models, confidence intervals, and standard deviation bands to detect anomalies. The system automatically compares actual survey scores against predicted scores and statistical thresholds, eliminating the need for manual review while maintaining detection accuracy.
Solution Approach 2:
The patent introduces statistical intermediaries (confidence intervals and standard deviation bands) that serve as objective criteria for anomaly detection. These statistical measures act as mediators between raw survey data and anomaly identification, providing a systematic framework that replaces subjective manual judgment with quantifiable thresholds.
2Loss of information
If traditional systems collect and categorize survey data, then companies have comprehensive data metrics, but they lack automated anomaly detection capability
Solution Approach 1:
The patent implements a feedback-based anomaly detection system where prediction models generate expected scores based on historical data, which are then compared against actual scores. The system provides feedback through confidence intervals and standard deviation bands, automatically flagging anomalies when actual scores fall outside expected ranges, thus enabling automated detection without losing data metric information.
Solution Approach 2:
The patent performs preliminary actions by training prediction models on historical survey data before actual anomaly detection is needed. The system pre-establishes confidence intervals and statistical baselines, so when new survey data arrives, anomaly detection can occur automatically and immediately without requiring real-time manual analysis.
3Reliability
If companies analyze survey metrics over time to identify trajectory changes, then they can detect real changes in customer experience, but they cannot efficiently distinguish anomalies from regular variation
Solution Approach 1:
The patent changes the analytical parameters by introducing statistical measures (confidence intervals, standard deviation bands) alongside traditional metric tracking. Instead of simply monitoring metric trajectories, the system evaluates whether changes exceed statistically significant thresholds, providing a more reliable distinction between genuine anomalies and regular variation while maintaining manageable system complexity.
Data Source
AI summary
An anomaly detection system using machine learning to generate predicted survey scores for a given duration and a given metric based on historic survey score data. The system compares a predicted survey score to the actual survey score and identifies anomalous actual survey scores. The anomaly detection system trains a plurality of survey score prediction models using historic survey score data. Each survey score prediction model is based on a specific survey score metric and a specific duration. The survey score prediction models generate expected survey score results for the given duration and the given metric. Based on the user-determined filtering and tolerances, the system determines if the actual survey score result is anomalous. The system generates reports for the detected anomalies and continually updates the survey score prediction models with newly obtained actual survey results, thereby improving the anomaly detection accuracy over time.