Survey on the Current Situation of College English Teaching in China’s Universities and the Direction of College English Teaching Reform and Development

Table 7

Algorithm 1.

---

Algorithm 1. Learning preference and aversion features

---

Input: Graph A(Q,E,M)

Dimensionality d = 200

Number of negative samples Z = 5

Output: Students’ preference features φ1∈R|Q1|d

Students’ dislike features in φ2∈R|Q1|d

1:Initialization in φ1∈R|Q1|d, Q1∈student node

φ2∈R|Q1|d in initialization, Q1∈student node

Initialization in φ'∈R|Q1|d, Q2∈English writing literature node

where:Q1∪Q2 = Q

2:Extract the information of network edges E1∪E2 = E

Where: the weight of edge E1 is greater than 1/2 of the maximum weight

The weight of edge E2 is less than 1/2 of the maximum weight

3:According to the degree of the network English writing literature nodes, the negative sampling table of English writing literature is generated

Preferred feature table N1, disliked feature table N2

4 for Exy in E1 do

5: Sampling Z English writing documents from table N1, sampling English writing documents in φ′ to form a list

6: Input the list, φ1(x) and φ′(y) to the objective function

7:end for

8:Get the vector representation φ′ of the English writing literature nodes

9 for Exy in E2 do

10:sample Z English writing documents from table N2, sample English writing documents φ′ to form a list

11: Input the list, φ2(x) and φ′(y) to the objective function

12. end for

13: return,φ1φ2

---