Attribute-Constrained Event Pattern Indexing for Real-Time Flow Queries
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems face high computational complexity and scalability issues in filtering and querying large numbers of event sequences or journeys in contact centers, leading to inefficient and time-consuming query operations.
Innovation Solution
Implementing optimized indices and data structures, such as B-trees and B+-trees, to facilitate attribute-constrained queries for real-time event sequence visualization and analytics, allowing for efficient filtering and querying of event sequences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional querying methods are used on large event sequence datasets, then query operations can be performed, but computational complexity increases and query efficiency deteriorates
Solution Approach 1:
The patent pre-processes event sequences into structured patterns and builds inverted indexes before querying occurs. This preliminary organization of data into event pattern sets with positional ordering allows the system to avoid complex computations during actual query operations, resolving the contradiction between query efficiency and computational complexity
Solution Approach 2:
The patent segments large event sequences into smaller, manageable event patterns and organizes them into structured data formats. By dividing the complex event data into pattern sets with defined positional relationships, the system can efficiently query specific patterns without processing the entire dataset, thereby improving query efficiency while reducing computational complexity
2Productivity
If traditional querying methods are used on large event sequence datasets, then query operations can be performed, but scalability deteriorates
Solution Approach 1:
The system performs preliminary organization of event sequences into pattern sets and builds inverted indexes in advance. This pre-processing enables the system to scale efficiently as data grows, because the structured organization allows incremental additions without requiring complete re-processing of datasets, thus improving both query speed and scalability
Solution Approach 2:
The patent introduces a new dimensional organization by creating inverted indexes that map event patterns to their positions in sequences, rather than传统的 sequential scanning. This dimensional transformation enables efficient random access to event patterns regardless of dataset size, improving scalability while maintaining query speed
Data Source
AI summary
A method for performing attribute constrained queries for event sequence visualization may include receiving an event sequence dataset including event sequences. The method further includes determining singleton events in the dataset, where each singleton event is associated with an attribute. The method also includes generating event pattern sets corresponding to different distinct singleton events and that maintain a fixed positional order for each event sequence. The event pattern sets include pattern occurrence data for each event sequence positioned according to the fixed positional order. The method also includes identifying the fixed positional order maintained by each event pattern set, and generating an attribute constraint search structure, such as a B-tree, B+tree, or table, that includes keys and pointer values. The method may also include searching the search structure to identify key values satisfying a received attribute constraint, and outputting the sequence identifiers corresponding to the identified key values.