Articles posted on a forum often contain new Internet terms related to opinion elements. Consequently, existing Chinese opinion-mining systems may exhibit low recall and precision because they cannot recognize these new Internet terms. Therefore, we designed an algorithm to elaborate on the opinion elements of such articles by extracting the new terms. By ignoring any uncommon opinion element that appears only once, we determine whether the new term identified through manual judgment is a useful opinion element for a specific domain and add it to the thesaurus. In comparison with semi-automatic annotation methods, our approach can save considerable labor. The same Chinese word may have different meanings depending on the context, and this fact is prone to cause difficulties by changing the polarity or meaning of certain opinion elements, leading to errors in the analysis results of many Chinese systems. We designed appropriate algorithms to address this problem. Meanwhile, this system extracts the opinion elements from an article based on its established thesaurus and simultaneously considers various sentence patterns, the default topic, and clause priority to determine the opinion tendency of the author.
