Formal Approach to Data Accuracy Evaluation

Usually, data quality is defined by multiple attributes that allow classifying the output (such as completeness, freshness, and accuracy) or methods exploiting these dependability, performance, protection). Among suggested attributes, we will discuss one of principal categories: accuracy. Scientific experiments, decisionmaking, retrieval are examples situations require a formal evaluation approach to The should be adaptable distinct understandings accuracy enduser expectations. This study investigates defines dimensions metrics affect its evaluation. investigation generates problems in user expectation specification database models. work describes our proposed for defining an algorithm considers distribution inaccuracies relations. decomposes query accordance with accuracy, labels every part value, addresses possibility enforcing using values. mainly contributes proposing explicit suggesting some improvement actions reinforce