Analysis of Summarization Evaluation Experiments

The goals of my dissertation are: 1) to propose a French terminology for the presentation of evaluation results of automatic summaries, 2) to identify and describe experimental variables in evaluations of automatic summaries, 3) to highlight the most common tendencies, inconsistencies and methodological problems in summarization evaluation experiments, and 4) to make recommendations for the presentation of evaluation results of automatic summaries. In this paper, I focus on the second objective, i.e. identifying and describing variables in summarization evaluation experiments.