Optimising human inspection work in automated verbatim coding

Giacomo Berardi

Consiglio Nazionale delle Ricerche

Andrea Esuli

Consiglio Nazionale delle Ricerche

Fabrizio Sebastiani1

Qatar Computing Research Institute

Introduction

In the past ten years we have championed an approach to automatically coding open-ended answers (‘verbatims’) based on ‘machine learning’ (Giorgetti & Sebastiani 2003). Based on these principles we have built an automated verbatim coding system, which we have variously applied to coding surveys in the social sciences (Giorgetti et al. 2003), in customer relationship management (Macer et al. 2007) and in market research (Esuli & Sebastiani 2010).