| CSS Extraction by Query Model |
|
|
| Step 1: For
each query, we will do semantic tags |
|
| according to the
component dictionary |
|
| The example: |
|
|
|
| Word by word |
||
| translation of |
||
| the query: |
||
|
|
|
| Please help me
receive send |
||
| some about half
pull light of |
||
| news |
||
|
|
|
| Step 2: We analyze the CSSs found along with
their |
||
| left context
and right context in order to determine |
||
| the most probable CSS. |
||
|
|
|
|
|
| For this
example, the two CSS vectors are: [6,9,4] |
||||
| and [7,9,3].
From Query Model, p[6,9,4]=0 and |
||||
| p[7,9,3]=0.03.
Thus, the most probable core |
||||
| semantic
structure is [7, 9, 3] and the CSS “半
(half) |
||||
| 拉 (pull) 灯 (light)” is extracted. |
||||