Predict Signals

What is Predict Signals

Predict Signals detects and flags potential anomalies in your data and enables you to take corrective actions through data-driven insights. These anomalies could arise from typos, mistakes, erroneous formulas, or conscious changes in planning assumptions. It gives you a detailed context around each signal by using an intuitive graph and lets you collaborate with your colleagues and peers to make more informed decisions.

How Predict Signals works

The Predict Signal's AI Engine, powered by machine learning and statistical/mathematical modeling, generates a predicted range representing the normal data ranges for a given dimension intersection. Any values falling outside of that range are tagged as Signals. These Signals are categorized as High risk, Medium risk, and Low risk depending on how far off they are from the predicted normal range.

AI Engine of Predict Signals

Predict Signals has a robust machine-learning pipeline that handles the entire data science cycle, starting from data pre-processing to model deployment and management.

The Model training phase in this ML pipeline is equipped with many time-series algorithms, statistical methods, and classical ML techniques with advanced ensemble and stacking methods to ensure consistent and accurate results. The Model training time is optimized to a maximum extent using multi-threading and parallel executions. You cannot change the algorithmic settings for the Model training as they are designed to adapt to varied customer data.