ComOM

Task descriptionThe rapid growth of online shopping and e-commerce platforms has led to an explosion of product reviews. These reviews often contain valuable information about users’ opinions on various aspects of the products, including comparisons between different devices. Understanding comparative opinions from product reviews is crucial for manufacturers and consumers alike. Manufacturers can gain insights into the strengths and weaknesses of their products compared to competitors, while consumers can make more informed purchasing decisions based on these comparative insights. To facilitate this process, we propose the “ComOM - Comparative Opinion Mining from Vietnamese Product Reviews” shared task.The goal of this shared task is to develop natural language processing models that can extract comparative opinions from product reviews. Each review contains comparative sentences expressing opinions on different aspects, comparing them in various ways. Participants are required to develop models that can extract the following information, referred to as a “quintuple,” from comparative sentences:Subject: The entity that is the subject of the comparison (e.g., a particular product model).Object: The entity being compared to the subject (e.g., another model or a general reference).Aspect: The word or phrase about the feature or attribute of the subject and object that is being compared (e.g., battery life, camera quality, performance).Predicate: The comparative word or phrase expressing the comparison (e.g., “better than,” “worse than,” “equal to”).Comparison Type Label: This label indicates the type of comparison made and can be one of the following categories: ranked comparison (e.g., “better”, “worse”), superlative comparison (e.g., “best”, “worst”), equal comparison (e.g., “same as,” “as good as”), and non-gradable comparison (e.g., “different from,” “unlike”).