Data Profiling for Change Rules

2026-06-05

AI Summary

Existing database systems struggle to effectively model temporal dynamics, contextual dependencies, and causal relationships among attributes. To address this limitation, this work proposes Change Rules (CRs)—a novel rule-based paradigm that explicitly captures antecedent-consequent attribute changes within ordered tuple sequences, thereby overcoming the constraints of traditional data quality rules in modeling temporal and contextual patterns. The authors introduce CR-Miner, an efficient algorithm that employs a level-wise candidate generation strategy to identify change intervals, integrating declarative dependency specifications with sequence analysis techniques. Experimental results demonstrate that CR-Miner achieves a 40–50% average speedup over state-of-the-art methods while significantly enhancing the granularity and efficiency of trend analysis and causal inference.

Abstract

Understanding data change is critical towards understanding trends, normal vs. abnormal behaviours, recognizing patterns, and the causes of change. Existing database systems have limited support for change management, relying on statistics, triggers, and constraints. Data quality rules model sequential changes along a restricted set of attributes, quantify change among unordered tuples, and have limited ability to model the context under which attribute changes occur. In this paper, we introduce Change Rules (CRs) that quantify the sequential changes among ordered tuples in both the antecedent and consequent attributes. CRs aim to address the limitations of existing declarative dependencies to support trend analysis and causal relationships that trigger change among attributes. We propose CR-Miner, an automated algorithm for CR discovery that generates candidate change intervals in a level-wise manner. Experimental results show that CR-Miner achieves an average runtime improvement of 40-50% over existing baselines.

Problem

• Data Profiling
• Change Rules
• Sequential Changes
• Trend Analysis
• Causal Relationships

Innovation

• Change Rules
• CR-Miner
• data profiling
• sequential change
• trend analysis