Experimental Evaluation of Automated and Manual Data Cleaning Systems: A Case Study Using Organizational Data

Abstract

This paper presents a comprehensive comparison between computerized and manual data cleaning methods across different types of errors, including missing values, incorrect values, and inconsistent formats. The comparison covers efficiency along the three most important dimensions of data quality: completeness, accuracy, and consistency. Using a common Kaggle dataset of 50,000 employee records, we build a systematic evaluation framework with diagnostic measures and resource-utilization profiles. The experiments evaluate cleaning quantitatively on a given distribution of anomaly types: 30% missing values (15,000 records), 25% incorrect values (12,500 records), and 45% inconsistent formats (22,500 records). Automated cleaning significantly helps normalize mixed formats (92% success) and handle invalid values (88%). However, manual cleaning outperforms automatic methods on complex, context-dependent cases, with 95% accuracy in cases that require domain knowledge. The paper shows clear benefits for both approaches: automated cleansing excels in speed and cost on large datasets, while manual cleaning is useful in difficult cases that require domain knowledge. These findings add value to the literature by presenting empirical evidence of the effectiveness of both approaches and by providing firms with a structured, knowledge-based filter for choosing suitable cleaning solutions in line with their data, needs, and organizational structure.