An Automatic Error Detection Method for Engineering English Translation Based on the Deep Learning Model

Table 2

LSTM-CRF algorithm for automatic translation error detection in engineering English.

---

LSTM-CRF training procedure

---

Neighboring = ((−1,1), (1,−1), (1,1), (−1,1),(−1,0),(0,1),(1,0),(1,−1))

aliment = enter (f2s, s2f);

DIAL; END(t2s); END(f2t);

Start-DIAG(): for target f = rots.

for source f

added.to = 0.…sn

if (t aliment of f)

 for next point (t-ne, f-ne):

if((t-ne not aliment and s-ne not aliment) and

 (t-ne, s-ne) in union (f2s, s2f))

add next point (t-ne, s-ne)

END(a):

 for target w t-ne = 0.…to

  for source word s-new = 0.…sn

   if((t-ne not aligned or s-ne not next) and

   (t-ne, s-ne) in a)

   add point (t-ne, s-ne);

---