Automatic Assessment of Quality of your Data for AI

Abstract

The saying Garbage In, Garbage Out resonates perfectly within the machine learning and artificial intelligence community. While there has been considerable ongoing effort for improving the quality of models, there is relatively less focus on systematically analysing the quality of data with respect to its efficacy for machine learning. Assessing the quality of the data across intelligently designed metrics and developing corresponding transformation operations to address the quality gaps helps to reduce the effort of a data scientist for iterative debugging of the ML pipeline to improve model performance. In this tutorial, we emphasize on the importance of data quality and its associated challenges in data, and highlights the importance of analysing data quality in terms of its value for machine learning applications. We will survey on important data-centric approaches to improve the data quality and the ML pipeline. We also will be focusing on the intuition behind them, highlighting their strengths and similarities, and illustrates their applicability to real-world problems. As part of hands on session, we first provide an overview on available data quality analysis tools like: Pandas Profilers, Amazon Deepqu, IBM's Data Quality for AI, etc. We will then showcase how an end users can assess the data quality for their structured (tabular) data using one of the available tool in detail.