Otwarty dostęp

Research on the Optimization of Intelligent English Vocabulary Teaching Paths Based on Reinforcement Learning Models

Xuefen Shi

Xuefen Shi

Yuxi Agriculture Vocation-Technical CollegeYuxi, China
Wyszukaj tego autora
Sciendo|Google Scholar
Shi, Xuefen
 oraz 
Jinli Pan

Jinli Pan

Yuxi Agriculture Vocation-Technical CollegeYuxi, China
Wyszukaj tego autora
Sciendo|Google Scholar
Pan, Jinli
  
29 wrz 2025
Research on the Optimization of Intelligent English Vocabulary Teaching Paths Based on Reinforcement Learning Models's Cover Image
O artykule
Poprzedni artykuł
Następny artykuł
Zacytuj
Pobierz okładkę
Data publikacji: 29 wrz 2025
Otrzymano: 06 sty 2025
Przyjęty: 27 kwi 2025
DOI: https://doi.org/10.2478/amns-2025-1135
Słowa kluczowe
© 2025 Xuefen Shi and Jinli Pan, published by Sciendo.
This work is licensed under the Creative Commons Attribution 4.0 International License.

Figure 1.

Markov decision process
Markov decision process

Figure 2.

The recommended model
The recommended model

Figure 3.

The distribution of the correct number of individual households
The distribution of the correct number of individual households

Figure 4.

All households are distributed
All households are distributed

Figure 5.

The value function of the initial state of a single door
The value function of the initial state of a single door

Figure 6.

The average of every 200 rounds ERR
The average of every 200 rounds ERR

Figure 7.

The variance of err per 200 rounds
The variance of err per 200 rounds

Figure 8.

The average of every 200 rounds ERR
The average of every 200 rounds ERR

Figure 9.

The variance of every 200 rounds of ERR
The variance of every 200 rounds of ERR

Figure 10.

System satisfaction survey
System satisfaction survey

Figure 11.

Attitude survey
Attitude survey

Figure 12.

Study results
Study results

System advantages and disadvantages survey

Issue number Part of the learner’s feedback
9 The system is more flexible than the English class four lexical learning software on the market
The system can analyze my vocabulary and recommend my vocabulary
The system can recommend me the vocabulary resources that I am interested in
This system will improve my motivation to learn English four words
This system will improve my efficiency of learning English four words
10 The department of lexical resources can increase the English interpretation of words
The system can increase the number of words

The comparison of the model and its model(k=9)

Data set Evaluation index BPR Q-Learning NeuCF FPMC DDPRG This model
Yelp Precision@9 0.0399 0.0397 0.0701 0.1021 0.1028 0.1162
Recall@9 0.1581 0.2106 0.5185 0.5002 0.5225 0.5453
NDCG@9 0.1362 0.2268 0.3234 0.3931 0.4006 0.4070
MovieLens Precision@9 0.0297 0.0406 0.0804 0.1148 0.1258 0.1357
Recall@9 0.1581 0.1400 0.4988 0.5057 0.5486 0.5852
NDCG@9 0.2782 0.1511 0.3438 0.4052 0.4255 0.4441
Last.fm Precision@9 0.0962 0.0706 0.0985 0.1355 0.1446 0.1574
Recall@9 0.3361 0.1506 0.5059 0.5258 0.5335 0.5454
NDCG@9 0.3447 0.1900 0.3672 0.3762 0.3868 0.3990

The comparison of the model and its model(k=7)

Data set Evaluation index BPR Q-Learning NeuCF FPMC DDPRG This model
Yelp Precision@7 0.0394 0.0202 0.0951 0.1284 0.1772 0.1954
Recall@7 0.1866 0.0998 0.5563 0.5235 0.5569 0.6025
NDCG@7 0.2652 0.2685 0.3856 0.4314 0.4460 0.4752
MovieLens Precision@7 0.0559 0.0523 0.1270 0.1256 0.1555 0.1753
Recall@7 0.2570 0.1686 0.5632 0.4684 0.5473 0.6249
NDCG@7 0.1588 0.1242 0.3999 0.3991 0.4525 0.5004
Last.fm Precision@7 0.1251 0.1025 0.1400 0.1634 0.1800 0.2063
Recall@7 0.4112 0.3000 0.5257 0.5570 0.5782 0.6034
NDCG@7 0.2594 0.2067 0.3764 0.3994 0.4090 0.4583

The comparison of the model and its model(k=5)

Data set Evaluation index BPR Q-Learning NeuCF FPMC DDPRG This model
Yelp Precision@5 0.0450 0.0398 0.0784 0.0995 0.1152 0.1258
Recall@5 0.1582 0.2244 0.5215 0.4954 0.5461 0.5753
NDCG@5 0.1358 0.2358 0.3353 0.3869 0.4026 0.4345
MovieLens Precision@5 0.0302 0.0501 0.0892 0.1261 0.1356 0.1450
Recall@5 0.1154 0.1496 0.5067 0.5598 0.5895 0.5984
NDCG@5 0.2786 0.1582 0.3598 0.4061 0.4452 0.4566
Last.fm Precision@5 0.0958 0.0775 0.1064 0.1257 0.1483 0.1660
Recall@5 0.3359 0.1594 0.5252 0.5240 0.5356 0.5584
NDCG@5 0.3452 0.1961 0.3796 0.3859 0.3891 0.4060

The comparison of the model and its model(k=3)

Data set Evaluation index BPR Q-Learning NeuCF FPMC DDPRG This model
Yelp Precision@3 0.0411 0.0424 0.0810 0.0897 0.0961 0.1159
Recall@3 0.1769 0.2458 0.5372 0.5076 0.5305 0.5522
NDCG@3 0.1595 0.2406 0.3598 0.3654 0.3819 0.4021
MovieLens Precision@3 0.0524 0.0525 0.0993 0.0922 0.1100 0.1257
Recall@3 0.1953 0.1581 0.4988 0.5357 0.5404 0.5543
NDCG@3 0.3161 0.3005 0.3685 0.4024 0.4299 0.4367
Last.fm Precision@3 0.0908 0.0851 0.1142 0.1202 0.1328 0.1453
Recall@3 0.3459 0.1953 0.5025 0.4953 0.5204 0.5402
NDCG@3 0.3530 0.2006 0.3680 0.3588 0.3975 0.4099
Język:
Angielski
Częstotliwość wydawania:
1 razy w roku
Dziedziny czasopisma:
Nauki biologiczne, Nauki biologiczne, inne, Matematyka, Matematyka stosowana, Matematyka ogólna, Fizyka, Fizyka, inne
Kanał RSS czasopisma