API Anomaly Detection via ML Prediction and Routing

Solution Overview

Problem

Conventional approaches fail to proactively detect and address application programming interface (API) issues, often leading to reactive handling of outages and performance problems, which can result in API timeouts, errors, and reduced customer satisfaction.

Innovation Solution

An API anomaly detection and resolution platform that uses machine learning algorithms to predict anomalous API behavior by analyzing transaction parameters, designating an API as anomalous, and routing requests to secondary APIs before issues occur, thereby preventing outages and maintaining service continuity.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If conventional monitoring approaches are used, then system simplicity is maintained, but API issues and outages cannot be detected in advance

Engineering Contradiction:

ImproveAPI availabilityVSAvoidmonitoring system complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The system performs preliminary actions by collecting baseline parameters during normal API operation and training machine learning models in advance. This allows the system to predict anomalies before they occur, enabling proactive routing decisions that prevent outages rather than reacting to failures after they happen.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent introduces an intermediary anomaly detection platform that sits between the API and the monitoring system. This platform collects parameters, trains machine learning models, and makes routing decisions, thereby decoupling the complexity of anomaly detection from the core API system while improving reliability.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Measurement precision

If machine learning algorithms are implemented for anomaly prediction, then API outage detection accuracy is improved, but processing time and computational resources increase

Engineering Contradiction:

Improveanomaly detection accuracyVSAvoidprocessing time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The system performs preliminary actions by collecting baseline parameters during normal API operation and training machine learning models in advance. This allows the system to predict anomalies before they occur, enabling proactive routing decisions that prevent outages rather than reacting to failures after they happen.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system applies machine learning algorithms selectively - using them for parameter analysis and anomaly prediction rather than processing every single API transaction in real-time. This partial application of complex processing maintains accuracy while reducing overall processing time and computational overhead.

Inventive Principle:

Principle #16Partial or excessive action

3Productivity

If API requests are routed to secondary APIs proactively, then service continuity is maintained, but system complexity and routing overhead increase

Engineering Contradiction:

Improveservice continuityVSAvoidrouting system complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent introduces an intermediary anomaly detection platform that sits between the API and the monitoring system. This platform collects parameters, trains machine learning models, and makes routing decisions, thereby decoupling the complexity of anomaly detection from the core API system while improving reliability.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The system implements feedback mechanisms where routing decisions are based on continuous monitoring of API parameters and anomaly predictions. The system learns from past performance data and adjusts routing strategies accordingly, creating a closed-loop system that improves service continuity while managing complexity through adaptive control.

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS11989287B2Application programming interface anomaly detection

Publication Date: 2024.05.21 DELL PROD LP

AI summary

A method comprises collecting parameters corresponding to processing by a first application programming interface of at least one application programming interface transaction, analyzing the parameters using one or more machine learning algorithms, and predicting, based at least in part on the analyzing, whether the at least one application programming interface transaction is anomalous. In the method, the first application programming interface is designated as being in an anomalous state responsive to predicting that the at least one application programming interface transaction is anomalous. One or more application programming interface requests for the first application programming interface are routed to a second application programming interface responsive to the anomalous state designation.