Asian Science Citation Index is committed to provide an authoritative, trusted and significant information by the
coverage of the most important and influential journals
to meet the needs of the global scientific community.
In order to improve the comprehensive performance and expand the scope of application of traditional term co-occurrence models, this study proposes Unified Term Co-occurrence Model. It unites two types of traditional term co-occurrence models (which are called mother models of Unified Term Co-occurrence Model) and could make a series of compound models of them for various research conditions. Precision and stability are two key performance indicators of term co-occurrence models. The first type of traditional term co-occurrence models are good at stability and the second type of traditional term co-occurrence models are good at precision. The experimental results in this study confirm that precision and stability of Unified Term Co-occurrence Model (UTCM) are not lower than both of its mother models. Then, a new measure for comprehensive performance is proposed and Unified Term Co-occurrence Model (UTCM) achieves better comprehensive performance compared with both of its mother models. Researchers can use unified term co-occurrence model instead of traditional models as an important tool to get more rational experimental results in relative research fields such as information retrieval, natural linguistic processing and computational linguistics, etc.
