On the Automatic Scoring of Handwritten Essays

Automating the task of scoring handwritten student essays is a challenging problem of AI. The goal is to assign scores which are comparable to those of human scorers even though both human and machine handwriting recognition do not achieve perfect transcription. The research described is a first attempt based on coupling two AI technologies: optical handwriting recognition and automated essay scoring. The test-bed is that of essays written by children in statewide reading comprehension tests in schools. The process involves several image-level operations: removal of pre-printed matter, segmentation of handwritten text lines and extraction of words. Constraints provided by the reading passage, the question and the answer rubric help recognize handwritten words. The method of essay scoring is based on using a vector space model and a machine learning approach to learn scoring parameters from a set of human-scored samples. System performance is compared to scoring done by human raters. ...

• Type: Conference
• Year: 2007
• Where: IJCAI
• Authors: Sargur N. Srihari, Rohini K. Srihari, Pavithra Babu, Harish Srinivasan
• Keywords: Artificial Intelligence, Essay Scoring, Handwriting Recognition, Handwritten Student Essays, IJCAI 2007