1. bookAHEAD OF PRINT
Informacje o czasopiśmie
License
Format
Czasopismo
eISSN
2444-8656
Pierwsze wydanie
01 Jan 2016
Częstotliwość wydawania
2 razy w roku
Języki
Angielski
access type Otwarty dostęp

Intelligent Recommendation System for English Vocabulary Learning – Based on Crowdsensing

Data publikacji: 15 Jun 2022
Tom & Zeszyt: AHEAD OF PRINT
Zakres stron: -
Otrzymano: 20 Feb 2022
Przyjęty: 27 Mar 2022
Informacje o czasopiśmie
License
Format
Czasopismo
eISSN
2444-8656
Pierwsze wydanie
01 Jan 2016
Częstotliwość wydawania
2 razy w roku
Języki
Angielski
Abstract

With the advent of the artificial intelligence era, the traditional methods of online learning can no longer meet people's personalised needs. Take the English vocabulary memory APP, which is widely used in the market of learning software, as an example. At present, there are various factors that make it difficult for learners to improve their learning efficiency, reduce their learning enthusiasm and enhance their learning effects. Therefore, in this paper, the application of machine learning algorithm in intelligent recommendation of English vocabulary learning is studied, and an intelligent recommendation system for English vocabulary learning based on crowdsensing is established. In its functional module, machine learning algorithms such as clustering algorithm, word vector training and collaborative filtering recommendation are used to realise the intelligent recommendation in the system. Finally, accurate resources of recommendation for user groups with different needs, interest characteristics and abilities improves the efficiency of English vocabulary learning.

Keywords

Introduction

In today's society, with the development of data transmission technology, the reduction of network cost and the improvement in the coverage that mobile devices offer, the development rate of online learning is greatly accelerated, and online learning platforms and massive online courses are constantly updated. The main battlefield for people to learn online has also spread from the website to the massive open online course platform under mobile phone and WeChat applet, even in educational UGC platforms such as forums and blogs [1]. Courses that are free of charge can be classified into audio or video, customised or universal. Among them, language learning, as a major platform of online learning, has blossomed everywhere, whether on the PC or on the mobile device, and nearly 100 English learning software can be downloaded in the mobile application market. At the same time, learners also gradually change their traditional ideas and begin to adapt to the fragmented learning mode with short content and loose structure, and this trend is especially obvious in the case of English vocabulary learning [2]. Online education makes up for the imbalance of educational resources in traditional education, and realises the sharing of educational resources. By making use of wireless mobile communication technology and equipment, learners can acquire educational resources indiscriminately, ignore the limitations of time and space, and obtain the benefit of digital learning anytime and anywhere.

However, the explosive growth of learning resources has a negative impact on both “teaching” and “learning” [3]. It is difficult for users to select the resources suitable for them from the massive data, and the phenomena such as high participation rate, low completion rate, uneven resource quality and homogeneous competition in platform become more and more prominent. Learners took a great effort to register in the platform and try to find high-quality resources, which indirectly led to the situation of low completion rate and low active period of users. For English vocabulary learning, the rigid resource push form leads to the difficulty that users learn words every day, which is largely determined by the selected dictionary, instead of depicting accurate user portraits according to different learners’ vocabulary size and personal abilities [4,5,6]. Therefore, in this paper, artificial intelligence is combined with English vocabulary learning to build an intelligent recommendation model for English vocabulary learning. According to the calculation of similar users and related words, a list of intelligent recommended words is formed, and a resource feature model is constructed. By calculating the correlation between words, an intelligent recommendation system for English vocabulary learning based on crowdsensing is designed and realised.

Related theories
Adaptive learning

Adaptive Learning originated in the 1970s, studies and simulates people's learning from the perspective of information processing and is essentially a computer program [7]. This program can generate new rules in the process of problem solving, and add them to the program, so as to solve similar problems more effectively. It mainly includes the following four aspects:

Change what: Adaptive learning can change the learning contents by distinguishing the difficulty level of a task, or change the form of representation and path selection of the content [8].

Why change: It focuses on three levels, namely learning parameters, interaction between learners and systems, and teaching variables.

When to change: The first adjustment is made based on learner characteristics, followed by continuous modelling and further adjustment based on interactive parameters.

How to change: The common adaptation is controlled by the system, including learners’ complete control over learning environment and content, and sharing adaptation.

As shown in Figure 1, the four-dimensional view of adaptive learning very intuitively shows a core problem in the design of adaptive learning system. It can be divided into learner-oriented and educator-oriented adaptive learning. The latter of these is characterised by a big conflict between curriculum management and the existing teaching system, and there are few mature products at present. As for the former, the system is familiar with the functions of acquiring interactive data of learners, creating a learning model, selecting information and displaying information. Learner-oriented adaptive learning can be realised through the cycle of four basic stages: acquisition, analysis, selection and presentation.

Fig. 1

Four-dimensional views of adaptive learning

Mobile crowdsensing

Mobile Crowd Sensing (MCS) [9] refers to collecting data by using smart mobile device sensors in physical space and mobile social networks in cyberspace. Owing to its inherent high mobility and scalability, it has been widely used in many sensing applications [10], as shown in Figure 2.

Fig. 2

Application range of mobile crowd sensing

Group intelligence has the following two characteristics [11, 12]: First, there is no central control, but it is distributed in such a manner as to be suitable for the working state under the network environment, and will not affect the solution of the whole problem due to the mistakes of decision-making and data anomalies. Second, groups are self-organised. Each individual in the group can communicate indirectly for information transmission and cooperation, and thus each individual can change the environment. However, the complex behaviour of the group is reflected by the intelligence in the process of individual interaction. This kind of wisdom emerging from the group is beyond the wisdom of its constituent individuals which is called group intelligence [13].

Group intelligence is applied in the educational environment, that is, individual learners have certain rules about their abilities and behaviours to learn specific knowledge, and they can communicate indirectly. Therefore, for new learners, it has great reference and application value. In the teaching process, the conclusion of group intelligence can be used to improve teaching methods, update teaching contents and share learning strategies, so as to help learners optimise the learning process and improve teaching efficiency.

Design of intelligent recommendation system for English vocabulary learning
Demand analysis

The role of the system is divided into learner and manager. The main functions of learners include: personal information maintenance, uploading corpus, completing tests, and vocabulary learning. The main functions of the administrator are to review corpus, update thesaurus, publish tests, recommend new words and so on.

Corpus update: It is used to solve the problems of singleness and lack of timeliness of professional corpus [14]. Through user-uploaded corpus, system audit, word segmentation, keyword extraction and other technical means, the role of swarm intelligence can be brought into full play to realise real-time updating of corpus.

Accurate recommendation of new words: It is used to solve the problem of low accuracy of intelligent vocabulary recommendation. Similar words of learned words and co-occurrence words are used as important sources of recommended words for mixed recommendation, which improves the efficiency of new word recommendation of the system.

Systematic recommendation strategy: It is used to solve the problem of time-consuming memorisation by users. This system is different from the way in which other memorising software show all usages to users, since it divides the words into resource types, which reduces the presentation of interference information, helps users to improve the efficiency of memorising words and makes personalised adjustments to the source and quantity of words.

Overall architecture design

As shown in Figure 3, the architecture of intelligent vocabulary recommendation system is divided into three layers: user layer, business layer and data layer. The layer provides data storage services and undertakes the responsibility of ensuring the reliability and security of data. According to the actual demand of the system, data layer stores user information database, user log behaviour database, vocabulary database, corpus, comment data published by users, test data and so on. The business layer implements the core business logic of the recommendation system, including similar words’ mining and adopting a user-based collaborative filtering algorithm to recommend words to learners, where the clustering algorithm is used to locate the learning style of users and adjust the push mode. The user layer is responsible for the interaction between learners and the system, the server response to the user's request and the display of the results, such as word learning, registration, corpus uploading, testing and other functions. Moreover, all data of user behaviour generated by this layer will be recorded in the log database.

Fig. 3

Systematic architecture

Functional design
Vocabulary learning

The core of English vocabulary learning is word learning [15]. When users recite words, the system will judge whether they are new users who need to complete the test first. Users who have completed a round of tests can choose to take the test again if they have not conducted a new round of tests for a long time or have formed an impression that the results of the previous round of tests do not constitute an accurate reflection of their actual standard. The system will recommend adaptive test questions for users with different needs according to the information registered by users in the early stage of registration, for example, recommending real questions or high-quality prediction questions over the years for users who choose CET-4 and CET-6 as their learning objectives; or selecting foreign original news such as BBCNews or NewYorkTimes for journalism students who need to improve their professional skills [16,17,18]. According to these learning data, the system can find highly relevant words, and form a list of recommended words for users to learn. As shown in Figure 4, when learning a new word, users can choose whether to collect it as a new word, comment on it or directly skip to learning the next word. In this way, the number of users’ new vocabulary books keeps increasing, and every time the software is restarted and the user starts learning, the system will recommend a new batch of words according to the updated vocabulary. If there is a need to review, the user can click on the new word book directly. This function helps users quickly find new words every time they learn, improves the efficiency of memorising words and rapidly expands their vocabulary.

When the user collects words, the system will automatically ask the user whether the words need to be marked as core words. If not marked, the system will judge whether the words belong to proper nouns according to the characteristics of the words themselves. If not, they will be recorded as ordinary words. In the stage of review, users will only be shown basic usages such as definitions and phonetic symbols. If the recorded word is regarded as a core word, not only will the basic usage be displayed but also advanced usage, such as example sentences, will be further developed.

Fig. 4

Subsystem of word learning

Corpus update

In the process of language learning, students with different professional backgrounds have personalised needs for English vocabulary, for example, students majoring in computer science need to master the latest technological achievements; students majoring in history need to learn a wide range of place-names and toponyms; and students majoring in news need to master the current events to help them develop their ability at news writing [19,20,21]. However, a professional dictionary usually can’t fundamentally solve their needs. The editing of the dictionary needs relevant corpora from different fields to be collected, and combined with opinions of experts, so as to sort out and integrate them. As shown in Figure 5, for students with different professional backgrounds, the system adopts the way in which users can customise the corpus, which allows users to upload different forms of corpus materials, including speech videos, interview recordings, news reports, professional papers, etc., so as to supplement the existing professional corpus. Managers can review the authority and correctness of data. By reviewing a high-quality corpus prepared in this manner, the professional thesaurus can subsequently be updated by such operations as speech-to-text conversion, word segmentation, core word extraction and calculation of word correlation. With the help of speech recognition tools, crawlers and abundant network resources, the source of the corpus shows an unprecedented affluence. With NLTK, Gensim and other natural language processing packages, it is easy to remove stop words, extract core words and train word vector function by calculating TF-IDF [22]. Through the co-occurrence word matrix and the trained word vector, co-occurrence words and similar words can be mined to prepare for the subsequent recommendation words. At the same time, administrators can review the corpus materials uploaded by users, continuously extract the latest materials and update the existing corpus.

Fig. 5

Subsystem of corpus maintenance

Progress test

The test module in commonly-used vocabulary learning systems is particularly poor [23]. Many software retain the function of vocabulary testing, but they are additional functions in systematic design. Except for the data of vocabulary range and ranking, which does not achieve the effect of supporting and assisting users to improve the efficiency of memorising words. In addition, besides pre-test and summative test, the role of formative assessment in the learning process cannot be ignored. An online vocabulary learning system can’t represent all the input sources of English resources for users in a certain period of time, and users’ vocabulary size should be a dynamic value. Especially with the coming of examination and grading, a user's vocabulary can be improved dramatically in a relatively short time. Therefore, as shown in Figure 6, users can choose to retake the test at any time when they feel that their vocabulary has greatly improved, or that the last test did not accurately reflect their true level. Afterwards, the system constantly updates the recommended vocabulary list according to the test results of users, so as to ensure the efficiency of memorising words.

Fig. 6

Subsystem of progress test

Implementation of intelligent recommendation system for English vocabulary learning
Data acquisition

At present, the open corpus data set is very rich, for example, Wikipedia has opened the download channel of its multilingual webpage data. Take news corpus as an example; there is a plethora of available data sets, including: English News Data of Reuters Corpora [24] (Reuters English news text and corresponding data of news category), data set of New York Times (registered developer account application api, which can be traced back to 1851, any searchable and indexable article), data set of 20Newsgroups [25] (about 20,000 newsgroup documents are collected, which are evenly divided into newsgroup collections of 20 different topics), etc. The experimental data set used in this paper is 3,000 news reports from the figure-eight website [26, 27].

Corpus pre-processing

In the process of processing the original corpus, there are many meaningless symbols and function words, and thus it is necessary to remove stop words, extract stem, restore word form and carry out other operations. Firstly, the Word_tokenize () function in NLTK package is used to segment the text data, and the information is processed into the smallest unit that the computer can handle. Similarly, NLTK package is adopted to remove function words (conjunctions, prepositions, articles, etc.) and punctuation marks from the corpus. First, by defining an empty list and traversing the segmented text list, the empty list is appended if the word does not exist in the deactivated word list. Secondly, Porter's stem extraction tool is used to extract the same stem from different inflections of English words. Finally, the WordNetLemmatizer function is used to restore the morphology. Thus, it is enough simply to understand the data structure of the incoming function. When these steps are implemented in Python language, importing packages and calling functions can be easily implemented. The example code is as follows:

fromnltk.corpusimportstopwords

fromnltk.stem.porterimportPorterStemmer

fromnltk.stemimportWordNetLemmatizer

Text=paragraph.lower()# Enter the paragraph text paragraph and convert it into lowercase characters.

Text _ list = nltk.word _ tokenize (text) # participle

english_punctuations=[’,’,’.’,’:’,’;’,’?’,’(’,’)’,’[’,’]’,’&’,’!’,’*’,’@’,’#’,’$’,’%’]

text_list=[wordforwordintext_listifwordnotinenglish_punctuations]

# Remove punctuation marks

Stop _ list = set (stop words.words (“English”)) # Take out all English stop words.

stopped_token=[wordforwordintext_listifwordnotinstopword_list]

Stemmed _ token = [porter stemmer (). stem (I)foriinstall_token] # stem extraction

lemmatized_token=[WordNetLemmatizer().lemmatize(i)foriinstemmed_token]

Gensim training word vector

In order to solve the “cold-start” of the system, the method of mining co-occurrence words and similar words of words is also adopted to realise mixed recommendation of words. The experimental corpus is trained by word2vec model of Gensim toolkit in python [28, 29], where the trained model is used to output similar words. It is very convenient to train word vectors in Gensim. We save the trained model and load it, and then call the most_similar () method to get similar words.

Mining co-occurrence words

Taking the professional category as the boundary, the system pre-processes the corpus materials uploaded by administrators and users, and the number of core words obtained is limited (assuming that there are N words). By creating a co-occurrence word matrix, the common frequency of all words can be obtained. The higher the frequency, the greater the correlation between words, and it is classified as a list of preferred words, thus realising mixed recommendation. The specific implementation logic is shown in Table 1. The N*N matrix is constructed with all N core words as the number of rows and columns. The data on the diagonal line are uniformly defined as 0, and the data on both sides of the diagonal line are symmetrically distributed. The number in ath row and bth column is 10, and the number of common occurrences of words a and b across all the corpora is 10.

Principle of co-occurrence word matrix

Order 1 2 3 4 ……… N

Order Trump Corruption Scandal Governor Senate
1 Trump 0 5 4 0 0 5
2 Corruption 5 0 1 2 0 3
3 Scandal 4 1 0 5 0 2
4 Governor 0 2 5 0 0 8
…… 0 0 0 0 0 0
N Senate 5 3 2 8 0 0

Owing to the large number of words, it is an important step to filter out unimportant words in the text and then calculate their total number. The text feature extraction algorithm TF-IDF is used to extract feature word vectors. The purpose of this algorithm can be understood as calculation of the importance of a word in a document [30, 31]. For the word i of a document j, the calculation of its TDIDF value is shown in Eq. (1): w(i,j)=tf×idf=Count(i,j)Size(j)×log(NDocs(i,D)) w(i,j) = tf \times idf = {{Count({\rm{i}},{\rm{j}})} \over {Size({\rm{j}})}} \times \log \left( {{N \over {Docs(i,D)}}} \right) Term frequency (TF) calculates the frequency of word i appearing in document j. Inverse document frequency (IDF) is obtained by dividing the total number of documents N by the number of documents containing the word i, and then taking the logarithm of the quotient. We multiply TF and IDF, and the greater the TF-IDF value, the more important the word is to the article. Finally, the words with value of TF-IDF greater than 0.01 are extracted to form the feature word set of the corpus, which is realised using the following algorithm:

Vector = tfidfvectolizer (stop _ words = stopwords _ 1ist) # Call the tfidfvectolizer method to convert the input corpus into a word vector.

tf idf=vectorfittransform(corpus)

Word _ list = vector.get _ feature _ names0 # Get all the words of the word bag model.

weight_list=tfidf.toarray()

for i in range(len(weight list))

Print(len(weight_list)) # Number of all words

Print (“-the first”, i+1, “tf-idf weight of paragraph text”)

for j in range(len(word_list)):

If weight list[i][j]>0.01: # Only words with values greater than 001 are reserved.

print(word_list[j],weight_list[i][j])

else:

j+−1

To build a word network, first, a common empty matrix needs to be built; we fill the first row and the first column of the matrix with feature words, and then calculate the common comments of feature words in each corpus. The main calculation of this segment code focuses on calculating the contribution times of feature words, which requires constant recursive operation. In order to improve the operation efficiency, the location of each feature word is stored in a dictionary {’feature words ’:[1,3,5,29,45,89]}. Then, we take out the corresponding position list of two feature words that need to be compared from the dictionary, turn the list into a set and use the method of intersection between two sets to ascertain the co-occurrence frequency of the two feature words. The final implementation algorithm is as follows:

defcount_matrix(matrix,formated_data):

Calculate the number of times each keyword appears.

# Construct a dictionary where keywords appear. keyword_location

keywordlistmatrix=[o][1:]

forwordinkeywordlist:

keyword_location[word]=[]

i=1

foreach_lineinformated_data:

eachline=str(each_line)

#pring(eachline)

ifwordineachline:

print(word)

keyword_location[word].append(i)

i=i+1

print(’Construct a dictionary of keyword occurrence locations’)

print(keywordlocation)

#Traverse the line, skip elements with subscript 0

forcolinrange(1,len(matrix)):

counter=0

ifcol>=raw:

ifraw=col:

matrix[raw][col]=counter

else:

counter=len(set(keyword_location[matrix[raw][0]])&set(keyword_location[matrix[0][col]]))matrix[raw] [col]counter

else:

matrix[raw][col]=matrix[col][raw]

print(’Finish!)

return matrix

Functional test

As the selected corpus is sourced from news articles, the experimental design utilised the services of 20 undergraduate students majoring in journalism in S University who had studied with traditional memorisation software for more than 3 months to test the function of similar vocabulary recommendation. They were asked to learn the randomly recommended vocabulary, which contains 30 words. Then, according to their condition of marking new words, intelligent recommendation was used to mine the co-occurrence words and similar words of new words, and a word list of 30 words is formed again. After learning two lists, the users evaluated the word qualities of the different lists, and verified whether the words intelligently recommended by the system increased the proportion of recommended new words in practical application.

According to their feedback, there is no significant difference in the probability of recommended new words between the intelligent recommended word list and the randomly composed word list. However, the users are generally satisfied with the intelligent recommended word list, and can clearly distinguish the randomly composed word list from the intelligent recommended word list. After further interviews, it was ascertained that users were of the opinion that the correlation between words was strong in the intelligent recommended word list, which can remind them of the specific usage scenarios of similar words, help them associate memories and finally enhance their initiative and participation in English vocabulary learning.

Conclusion

From the perspective of artificial intelligence, in this paper, an intelligent recommendation model for memorising and learning English vocabulary based on crowd sensing is implemented. By taking adaptive learning as the main idea, an intelligent recommendation system that meets users’ needs is built, where the personalised intelligent recommendation function of machine learning algorithms, such as clustering algorithm, word vector training and collaborative filtering recommendation, is realised. The data set of public news is selected as experimental corpus that the functions of corpus processing, core word extraction, co-occurrence words and similar words mining are realised. The results of the deployment of a small-scale experimental test indicated that users were generally satisfied with the intelligent recommendation system for English vocabulary learning, and their initiative and participation were significantly improved.

Fig. 1

Four-dimensional views of adaptive learning
Four-dimensional views of adaptive learning

Fig. 2

Application range of mobile crowd sensing
Application range of mobile crowd sensing

Fig. 3

Systematic architecture
Systematic architecture

Fig. 4

Subsystem of word learning
Subsystem of word learning

Fig. 5

Subsystem of corpus maintenance
Subsystem of corpus maintenance

Fig. 6

Subsystem of progress test
Subsystem of progress test

Principle of co-occurrence word matrix

Order 1 2 3 4 ……… N

Order Trump Corruption Scandal Governor Senate
1 Trump 0 5 4 0 0 5
2 Corruption 5 0 1 2 0 3
3 Scandal 4 1 0 5 0 2
4 Governor 0 2 5 0 0 8
…… 0 0 0 0 0 0
N Senate 5 3 2 8 0 0

Bloom's taxonomy of educational objectives[M]. Longman, 2009. Bloom's taxonomy of educational objectives [M]. Longman 2009 Search in Google Scholar

Boyle E A, Duffy T, Dunleavy K. Learning styles and academic outcome: The validity and utility of Vermunt's Inventory of Learning Styles in a British higher education setting[J]. British journal of educational psychology, 2003, 73(2): 267–290. BoyleE A DuffyT DunleavyK Learning styles and academic outcome: The validity and utility of Vermunt's Inventory of Learning Styles in a British higher education setting [J] British journal of educational psychology 2003 73 2 267 290 10.1348/0007099036062697612828816 Search in Google Scholar

Dunn R, Griggs S A, Olson J, et al. A meta-analytic validation of the Dunn and Dunn model of learning-style preferences[J]. The Journal of Educational Research, 1995, 88(6): 353–362. DunnR GriggsS A OlsonJ A meta-analytic validation of the Dunn and Dunn model of learning-style preferences [J] The Journal of Educational Research 1995 88 6 353 362 10.1080/00220671.1995.9941181 Search in Google Scholar

Elaish M M, Shuib L, Ghani N A, et al. Mobile English Language Learning (MELL): a literature review[J]. Educational Review, 2019, 71(2): 257–276. ElaishM M ShuibL GhaniN A Mobile English Language Learning (MELL): a literature review [J] Educational Review 2019 71 2 257 276 10.1080/00131911.2017.1382445 Search in Google Scholar

Eberhart R C, Shi Y, Kennedy J. Swarm intelligence[M]. Elsevier, 2001. EberhartR C ShiY KennedyJ Swarm intelligence [M] Elsevier 2001 Search in Google Scholar

Figure Eight. News article/Wikipedia page pairings. [DB/OL], 2015-03-27/2020-01-10 Figure Eight News article/Wikipedia page pairings [DB/OL], 2015-03-27/2020-01-10 Search in Google Scholar

Wehrwein E A, Lujan H L, DiCarlo S E. Gender differences in learning style preferences among undergraduate physiology students[J]. Advances in physiology education, 2007, 31(2): 153–157. WehrweinE A LujanH L DiCarloS E Gender differences in learning style preferences among undergraduate physiology students [J] Advances in physiology education 2007 31 2 153 157 10.1152/advan.00060.200617562903 Search in Google Scholar

Ying R, He R, Chen K, et al. Graph convolutional neural networks for web-scale recommender systems[C]//Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. 2018: 974–983. YingR HeR ChenK Graph convolutional neural networks for web-scale recommender systems [C] Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining 2018 974 983 10.1145/3219819.3219890 Search in Google Scholar

IResearch. Q3 2019 China Online Education Industry Data Release Report. [J/OL] 2019-12-27/2020-01-10. IResearch Q3 2019 China Online Education Industry Data Release Report [J/OL] 2019-12-27/2020-01-10 Search in Google Scholar

Guo B, Wang Z, Yu Z W, et al. Mobile crowd sensing and computing: The review of an emerging human-powered sensing paradigm[J]. ACM Computing Surveys, 2015, 48 (1): 1–31. GuoB WangZ YuZ W Mobile crowd sensing and computing: The review of an emerging human-powered sensing paradigm [J] ACM Computing Surveys 2015 48 1 1 31 10.1145/2794400 Search in Google Scholar

Shu L, Chen Y F, Huo Z Q, et al. When mobile crowd sensing meets traditional industry[J]. IEEE Access, 2017, 5: 15300–15307. ShuL ChenY F HuoZ Q When mobile crowd sensing meets traditional industry [J] IEEE Access 2017 5 15300 15307 10.1109/ACCESS.2017.2657820 Search in Google Scholar

Yu D J, Chen C, Wu J H, et al. Personalized game recommendation based on implicit feedback[J]. Acta Electronica Sinica, 2018, 46(11): 2626–2632. (in Chinese) YuD J ChenC WuJ H Personalized game recommendation based on implicit feedback [J] Acta Electronica Sinica 2018 46 11 2626 2632 (in Chinese) Search in Google Scholar

Xiong H Y, Zhang D Q, Chen G L, et al. iCrowd: Near-optimal task allocation for piggyback crowdsensing[J]. IEEE Transactions on Mobile Computing, 2016, 15(8): 2010–2022 XiongH Y ZhangD Q ChenG L iCrowd: Near-optimal task allocation for piggyback crowdsensing [J] IEEE Transactions on Mobile Computing 2016 15 8 2010 2022 10.1109/TMC.2015.2483505 Search in Google Scholar

Hao Gang, Li Yanyan, Zheng Yafeng. Socialized recommendation model and method for large-scale online learning [J]. Modern Distance Education Research, 2018, (03): 94–103. HaoGang LiYanyan ZhengYafeng Socialized recommendation model and method for large-scale online learning [J] Modern Distance Education Research 2018 03 94 103 Search in Google Scholar

Chen Jiayan. Research on personalized recommendation of learning resources based on learning behavior characteristics [D]. Nanjing University of Posts and Telecommunications, 2018. ChenJiayan Research on personalized recommendation of learning resources based on learning behavior characteristics [D] Nanjing University of Posts and Telecommunications 2018 Search in Google Scholar

Gagne R M. Learning outcomes and their effects: Useful categories of human performance[J]. American psychologist, 1984, 39(4): 377. GagneR M Learning outcomes and their effects: Useful categories of human performance [J] American psychologist 1984 39 4 377 10.1037/0003-066X.39.4.377 Search in Google Scholar

Hwang G J, Fu Q K. Trends in the research design and application of mobile language learning: a review of 2007–2016 publications in selected SSCI journals[J]. Interactive Learning Environments, 2019, 27(4): 567–581. HwangG J FuQ K Trends in the research design and application of mobile language learning: a review of 2007–2016 publications in selected SSCI journals [J] Interactive Learning Environments 2019 27 4 567 581 10.1080/10494820.2018.1486861 Search in Google Scholar

Honey P, Mumford A. Learning styles questionnaire[M]. Organization Design and Development, Incorporated, 1989. HoneyP MumfordA Learning styles questionnaire [M] Organization Design and Development, Incorporated 1989 Search in Google Scholar

He Liyan, Cui Yuhang, Lin Lin, et al. A study of college English vocabulary learning strategies based on mobile learning mode [J]. Economist, 2018, (12): 206–208. HeLiyan CuiYuhang LinLin A study of college English vocabulary learning strategies based on mobile learning mode [J] Economist 2018 12 206 208 Search in Google Scholar

Li Hongmei, Xu Zhenhua. A Case Study of Intelligent Teaching System Supporting English Learning [J]. distance education journal, 2012, 30(03): 66–72. LiHongmei XuZhenhua A Case Study of Intelligent Teaching System Supporting English Learning [J] distance education journal 2012 30 03 66 72 Search in Google Scholar

Li Peijun. The Construction of College English Vocabulary Teaching Mode Based on Mobile Applied Learning Tools [J]. Journal of Heilongjiang Institute of Education, 2016, 35(05): 139–141. LiPeijun The Construction of College English Vocabulary Teaching Mode Based on Mobile Applied Learning Tools [J] Journal of Heilongjiang Institute of Education 2016 35 05 139 141 Search in Google Scholar

Li Cao. Research on the application of context in English vocabulary teaching [J]. Teaching and Management, 2013, (09): LiCao Research on the application of context in English vocabulary teaching [J] Teaching and Management 2013 09 Search in Google Scholar

Zhang Shufang. The application of association strategy in English vocabulary teaching [J]. Teaching and Management, 2013, (24): 139–141. ZhangShufang The application of association strategy in English vocabulary teaching [J] Teaching and Management 2013 24 139 141 Search in Google Scholar

Zheng Qinghua, Dong Bo, Qian Buyue, Tian Feng, Wei Bifan, Zhang Weizhan, Liu Jun. Current situation and development of wisdom education research Trend [J]. Computer Research and Development, 2019,56(01):209–224. ZhengQinghua DongBo QianBuyue TianFeng WeiBifan ZhangWeizhan LiuJun Current situation and development of wisdom education research Trend [J]. Computer Research and Development 2019 56 01 209 224 Search in Google Scholar

Zhang Nan. Research and implementation of online education-oriented learning resource recommendation method [D]. xidian university, 2018. ZhangNan Research and implementation of online education-oriented learning resource recommendation method [D] xidian university 2018 Search in Google Scholar

Zhao Jian, zhangxin Chuan, Li Jiaming, He Chen. Summary of research on swarm intelligence 2.0 [J]. Computer Engineering, 2019,45(12):1–7. ZhaoJian zhangxinChuan LiJiaming HeChen Summary of research on swarm intelligence 2.0 [J] Computer Engineering 2019 45 12 1 7 Search in Google Scholar

Zhang Ruonan. An exploratory study on the effect of word memory APP on junior high school English learning [D]. Shanghai Normal University, 2018. ZhangRuonan An exploratory study on the effect of word memory APP on junior high school English learning [D] Shanghai Normal University 2018 Search in Google Scholar

20 Newsgroups. [DB/OL] http://qwone.com/~jason/20Newsgroups, 2008-01-14/2020-01-10 20 Newsgroups [DB/OL] http://qwone.com/~jason/20Newsgroups, 2008-01-14/2020-01-10 Search in Google Scholar

Waterman D A. Adaptive production systems[R]. CARNEGIE-MELLON UNIV PITTSBURGH PA DEPT OF COMPUTER SCIENCE, 1974. WatermanD A Adaptive production systems [R] CARNEGIE-MELLON UNIV PITTSBURGH PA DEPT OF COMPUTER SCIENCE 1974 10.21236/ADA013570 Search in Google Scholar

Wehrwein E A, Lujan H L, DiCarlo S E. Gender differences in learning style preferences among undergraduate physiology students[J]. Advances in physiology education, 2007, 31(2): 153–157. WehrweinE A LujanH L DiCarloS E Gender differences in learning style preferences among undergraduate physiology students [J] Advances in physiology education 2007 31 2 153 157 10.1152/advan.00060.200617562903 Search in Google Scholar

Polecane artykuły z Trend MD

Zaplanuj zdalną konferencję ze Sciendo