Learning Users’ Intention of Legal Consultation through Pattern-Oriented Tensor Decomposition with Bi-LSTM

Algorithm 1

Pattern-oriented tensor decomposition method.

Input: Tensor which represents users’ legal consulting statement, , and tensor which is users’ legal

   consultation pattern, .

Output: The core tensor which is close to users’ legal consultation pattern in the layer of the tensor structure,

   , and its corresponding factor matrices and , , .

1 Initialize factor matrices and ;

2 for  to thresholdvalue  do

3   for to do

4     ;

5     , , ;

6     ;

7    end

8    for to do

9     ;

10     , , ;

11      ;

12  end

13 end

14  ;

15 return , , ;