Multidimensional Data Reversion: Free Text Isn't Free

In this episode of Multidimensional Data Reversion, David Yanofsky and Shannon Capone Kirk discuss methods for analyzing data, particularly focusing on free text analysis. They explore the importance of human involvement in building rules and identifying scope, contrasting it with the assumption that data analysis is purely automated. The conversation covers strategies for evaluating free text, including sentiment analysis and rubrics for third-party scoring, while setting aside generative AI for later discussion. They emphasize the value of directly asking individuals about their feelings and experiences to reduce bias and improve the accuracy of sentiment analysis, especially in employee feedback surveys. The hosts also address common client inquiries about data analysis tools and the challenges of using AI in e-discovery, highlighting the need for clear questions and calibrated approaches to data review.