Recognition of Chinese Legal Elements Based on Transfer Learning and Semantic Relevance

Algorithm 2

Element identification method for Chinese legal documents.

Input: Combination T of sentences and elements of legal instruments

Output: List of elements for each sentence

1.

2. For t in T do

3. Using the BERT model for sentence t.sen and element t.lab representations, it obtains the sentence vector bs, bl

4. Calculating similarity using cosine similarity

5. Calculation using schematic functions:

6. If I(x) =1: Result[t.sen].Attend(t.lab)

7. Else: Continu

8. End for

9. Return result