Yuling Liu, Xingming Sun, Yongping Liu

Partition the dictionary table D into 4 small tables D₁, D₂, D₃, D₄ according to the set of parts of speech P = {n, v, a, d} and construct a Huffman tree H_x for D_x using step 2) described in Algorithm 1.

(3)	Extract all the note pages from the stego-document T`.
(4)	Parse sentences of the note pages by using the morphological analyzer to obtain a sequence of words with part-of-speech w1/t1 w2/t2….
(5)	If tt ∈ P, decode the corresponding bits of word ww in the following way:
(a)	Starting from the root of the Huffman tree Hx, traverse it to the leaf node ww and record the path traversed.
(b)	Analyze the path traversed and set the current bit of C` to 0 if the path goes down a left child; or to 1. Go to the next bit of C` for each child traversed.
(6)	Repeat step 5) until C`L+m has been retrieved.