Sequential Event Classification Using Milestone Graph Constraints
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Raw sequential event data is often not meaningful to end users, contains errors, duplications, or is received out of order, posing challenges in applications where the order of events is critical.
Innovation Solution
A method and system that apply machine learning and graph techniques to classify and transform sequential event data, automatically converting generic status codes into meaningful milestones, using a milestone graph to enforce real-world constraints and optimize classification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If sequential raw event data is used directly, then the data contains important information about each event, but the information is not readily apparent or discernable to end users
Solution Approach 1:
The patent introduces an intermediary processing system that includes a sequence detector and classification module. This intermediary component receives raw sequential event data, processes it through machine learning models, and transforms it into meaningful classifications that are discernable to end users, thereby resolving the contradiction between information discernability and processing complexity
Solution Approach 2:
The patent replaces manual information extraction and classification processes with automated machine learning-based classification systems. The sequence detector and classification module automatically analyze raw event data and assign meaningful classifications without requiring manual intervention, thus improving information discernability while managing processing complexity through automation
2Productivity
If sequential event data is processed without validation, then processing speed is maintained, but errors, duplications, and out-of-order data create issues in event sequencing
Solution Approach 1:
The patent implements preliminary validation and sorting actions before final classification. The sequence detector identifies and corrects out-of-order events, and the classification module validates data integrity before processing. This preliminary action ensures event sequence accuracy is maintained while allowing continuous processing to preserve productivity
Solution Approach 2:
The patent incorporates feedback mechanisms where the classification module continuously validates incoming event data against established patterns and constraints. When errors or duplications are detected, the system provides feedback to adjust processing and maintain sequence accuracy, thereby resolving the contradiction between processing speed and sequence reliability
3Loss of information
If machine learning and graph techniques are applied to classify events, then meaningful milestones are generated with improved visibility, but computational complexity increases
Solution Approach 1:
The patent segments the complex classification task into distinct functional modules: a sequence detector for identifying event patterns, a classification module for assigning meanings, and a milestone generator for creating structured outputs. This segmentation allows each component to be optimized independently, improving event meaning visibility while managing overall computational complexity through modular architecture
Solution Approach 2:
The patent transforms raw event parameters into classified milestone parameters through machine learning models. By changing the parameter representation from raw data formats to structured milestone formats with meaningful classifications, the system improves event visibility while the models are trained to perform these transformations efficiently, balancing computational complexity
Data Source
AI summary
A method and system are provided that apply a combination of machine learning and graph techniques to classify and transform sequential event data. In some embodiments, the method and system are applied to generate raw data in the shipping industry to automatically classify a sequence of status codes extracted from EDI data files corresponding to a series of physical events experienced by a shipping container into a sequence of meaningful milestones to provide improved visibility regarding the actual status of the shipping container. The method and system can be applied to classify and transform sequential event data for use in the shipping industry and in other applications.