Abstract:

In this paper, word segmentation techniques are used to analyze the form features of 3852 pairs of statistical terms, which are captured by parallel terms grasping technique. In addition, a model is established to compare the economic rate of statistical terms. Our results show that statistical terms are consisted of low-frequency modifier and high-frequency main word; that although hyphen will reduce the economic index, it will make the terms easier to understand; and that the economic index of Chinese terms is much higher than that of English. Based on these findings, we put forward two basic principles, and introduce the construction process of terminology database.

Key words: statistical term, terminology database, word segmentation, economic index