Enabling Deeper Linguistic-Based Text Analytics—Construct Development for the Criticality of Negative Service Experience

Significant progress has been made in linguistic-based text analytics particularly with the increasing availability of data and deep learning computational models for more accurate opinion analysis and domain-specific entity recognition. In understanding customer service experience from texts, analysis of sentiments associated with different stages of the service lifecycle is a useful starting point. However, when richer insights into issues associated with negative sentiments and experiences are desired to inform intervention, deeper linguistic analyses such as identifying specific touchpoints and the context of the service users become important. While research in this direction is beginning to emerge in some domains, we are yet to see similar efforts in the domain of healthcare. We present in this paper the results from our construct development effort for quantifying how critical a negative patient experience is using different elements of the available textual feedback as a key basis for prioritizing interventions by service providers. This involves the identification of the different dimensions of the construct, associated linguistic markers and metrics to compute the criticality index. We also present the results of the application of our developed conceptualization to linguistic-based text analysis of a small dataset of patient experience feedback.