Partitioned Trie Processing for Scalable Customer Journey Visualization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing trie data structures face scalability issues and computational inefficiencies when used for customer journey visualizations, leading to significant processing overhead and latency in filtering and querying customer interactions with contact centers.
Innovation Solution
The method involves splitting a data frame of customer interactions into partitions based on organization identifiers and sequence identifiers, generating multiple trie data structures for each partition, merging these structures to create a combined organization trie, and performing analytics on the combined structure to generate visualizations, while utilizing separate execution units for concurrent processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If a single trie data structure is used to store all customer interaction events, then data completeness is maintained, but processing time and computational complexity increase significantly
Solution Approach 1:
The patent divides the customer interaction dataset into multiple partitions based on organization identifiers and sequence identifiers. Each partition is processed independently to create separate trie data structures, which are then merged. This segmentation reduces the computational complexity of processing each individual trie while maintaining the ability to retrieve complete customer journey information through the merged structure.
2Productivity
If multiple trie data structures are created for different partitions, then processing efficiency improves through parallel computation, but memory usage and merging overhead increase
Solution Approach 1:
The dataset is segmented into partitions that can be processed in parallel across multiple execution units. Each partition generates a separate trie data structure that fits within available memory constraints, enabling efficient parallel processing while avoiding the memory explosion that would occur with a single comprehensive trie.
Solution Approach 2:
Multiple partition-specific trie data structures are merged into a unified trie structure that preserves all customer interaction information. The merging process combines the benefits of parallel processing with the completeness of a single comprehensive data structure, allowing efficient querying of entire customer journeys.
3Speed
If concurrent processing is used to generate trie data structures, then processing speed increases, but coordination overhead and data consistency challenges arise
Solution Approach 1:
The processing task is divided into independent partition processing operations that can execute concurrently without interfering with each other. Each execution unit processes a specific partition independently, eliminating the need for complex coordination mechanisms while achieving parallel processing speedups.
Solution Approach 2:
Data is pre-partitioned by organization identifiers and sequence identifiers before concurrent processing begins. This preliminary organization ensures that each execution unit receives self-contained data partitions that can be processed independently, avoiding the need for complex inter-process communication and coordination during the trie generation phase.
4Measurement precision
If filtering is performed on the combined trie data structure, then accurate analytics are achieved, but query performance degrades with large datasets
Solution Approach 1:
The filtered view functionality partitions the large combined trie data structure into smaller, manageable segments based on organization identifiers. Each partition can be queried independently and efficiently, and the results are combined to provide accurate analytics. This approach maintains measurement precision while dramatically improving query performance by avoiding the need to traverse the entire large dataset.
Data Source
AI summary
A method for providing efficient trie data structure processing according to an embodiment includes splitting, based on organization identifiers and sequence identifiers, a data frame indicative of a set of events associated with customer interactions with automated agents of a contact center to produce a set of multiple partitions, and producing a set of multiple trie data structures, including generating a trie data structure for each partition. Each trie data structure represents aggregate counts of a corresponding subset of event sequences associated with a corresponding organization. The method also includes merging multiple trie data structures of the set of trie data structures to produce a combined organization trie data structure and storing the combined organization trie data structure to enable generation of a visualization of the combined organization trie data structure.


