Temporal Point Process Bucketing for Entity Resolution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing bucketing algorithms in Master Data Management (MDM) systems are limited in effectively grouping records with fictional identity data, as they rely on identity attributes and are not well-suited for handling intentionally obscured or altered identities, leading to inefficiencies in entity resolution and scalability issues when dealing with large datasets.
Innovation Solution
The implementation of temporal point processes for bucketing records, which groups records based on temporal similarities and patterns, allowing for the identification of records with fictional identity data by analyzing the temporal relationships and patterns in data records, thereby improving entity resolution and scalability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional bucketing algorithms are used that rely on identity attributes, then records with clear identities can be grouped effectively, but records with fictional or obscured identity data cannot be properly identified or grouped
Solution Approach 1:
The patent changes the parameter used for bucketing from identity-based attributes to temporal attributes. Instead of grouping records by name, ID, or other identity characteristics, the system groups records by temporal patterns such as time intervals between events, frequency of occurrences, and temporal sequences. This parameter change enables effective grouping of records with fictional or obscured identities while maintaining high accuracy for records with clear identities.
2Productivity
If entity resolution processes handle all records uniformly using identity attributes, then processing is straightforward, but scalability is reduced when dealing with large datasets containing fictional identity data
Solution Approach 1:
The patent segments the entity resolution process into two distinct phases: bucketing based on temporal attributes, and subsequent detailed matching within buckets. This segmentation allows the system to efficiently process large datasets by first grouping records using computationally efficient temporal pattern matching, then applying more resource-intensive matching algorithms only to smaller subsets of records within each bucket. This dramatically improves scalability while maintaining resolution accuracy.
3Productivity
If records are grouped based on semantic criteria alone, then grouping is simple and fast, but temporal patterns and relationships between events are not captured
Solution Approach 1:
The patent creates a multi-functional bucketing mechanism that simultaneously captures both semantic similarity and temporal patterns. The bucketing system evaluates multiple attributes including semantic criteria (name similarity, attribute matching) and temporal criteria (time intervals, event sequences, frequency patterns) in an integrated manner. This universal approach allows the system to maintain high grouping speed while preserving comprehensive information about both the content and temporal relationships of records.
Data Source
AI summary
An embodiment establishes a designated attribute value as a semantic criterion for grouping records in a bucket, identifies a first set of records having attribute values that satisfy the semantic criterion, and adds the first set of records to the bucket. The embodiment detects that the first set of records represent a first series of events that occurred in succession at respective times. The embodiment derives a temporal attribute value representative of a time pattern formed by the times of the first series of events and designates the temporal attribute value as a temporal criterion for grouping records in the bucket. The embodiment identifies a second set of records that represent a second series of events and satisfy the temporal criterion and adds the second set of records to the bucket based at least in part on the second set of records satisfying the temporal criterion.