Refining the Initial Categorization

We replace the categories provided by the data (I, B, O) with more distributional categories. The category I is divided into three distributional subcategories.

S $=$ AL* NU AR*

We suppose that structures are composed of a nucleus (NU) with optional left (AL) and right (AR) adjuncts. We here give informal definitions, the formal/distributional ones are given in the next sections. The nucleus is the main element (head) of a given structure: the noun for Noun Phrase, the verb for VP, VP for clause. The adjuncts correspond to auxiliary elements, for instance adjectives for a noun. They are in a dependence relation with the head of the structure. The adjuncts are characterized by their formal position (left/right) relative to the nucleus. Adjuncts are optional (a nucleus may occur alone). Apart from these two main categories, we need to introduce an additional element: the linker. The linker is a special element which builds an endocentric structure with two elements. It usually corresponds to coordination.

An element of the structure (nucleus or adjunct) might possess the break property. This notion is introduced in order to manage sequences where adjacent nuclei belong to different structures. When an element has the break property, its presence closes the preceding structure (if any), and opens a new one. The following sentence shows an example of the new categorization (compare with example Section 3):

InO earlyAB+ tradingNU inO HongNU KongNU MondayNUB+ ,O goldNU wasO quotedO atO $A 366.50NU anAB+ounceNU .O