| Peer-Reviewed

Design and Implementation of Journal Recommendation Model Based on L-BERT

Received: 5 June 2022     Published: 9 June 2022
Views:       Downloads:
Abstract

The output cycle and quantity of academic achievements of researchers can promote academic exchanges and integration to a certain extent. Therefore, it is particularly important to design a journal recommendation model for scientific researchers that is fast and relatively suitable for producing scientific research results. Based on this, focusing on the information knowledge of papers and the characteristics of journal topics, combined with deep learning technology, this paper proposes a journal-oriented recommendation model based on L-BERT (LDA-BERT), which realizes the purpose of recommending journals for scholars. Firstly, topic extraction is performed on papers in a specific journal, and the number of topic words is determined by perplexity. Secondly, the vectorized representation of the topic words is carried out by using the BERT model, and the fixed representation value is obtained through the mean value idea. Finally, Euclidean distance is used to determine the similarity between papers and journals. In order to verify the effect of L-BERT model, it is compared with Word2Vec model on the real data set. The results show that the recommended results of L-BERT model are improved by 17%, 10% and 13% respectively in Precision, Recall, and F1-Score, which fully shows that this research has a certain application value, and the research results can help researchers shorten the cycle of looking for journals.

Published in American Journal of Computer Science and Technology (Volume 5, Issue 2)
DOI 10.11648/j.ajcst.20220502.25
Page(s) 134-145
Creative Commons

This is an Open Access article, distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution and reproduction in any medium or format, provided the original work is properly cited.

Copyright

Copyright © The Author(s), 2022. Published by Science Publishing Group

Keywords

LDA Model, BERT Model, Euclidean Distance

References
[1] Wu Z, Wu J, Khabsa M, et al. Towards building a scholarly big data platform: Challenges, lessons and opportunities [C]// IEEE/ACM Joint Conference on Digital Libraries. IEEE, 2014: 117-126.
[2] Rossmann, Bernhard, Canzaniello, et al. The future and social impact of Big Data Analytics in Supply Chain Management: Results from a Delphi study [J]. Technological forecasting and social change, 2018, 130 (May): 135-149.
[3] Ahmed M M, Kader M A, Zamli K Z. Comparative Study to Measure the Quality of Big Scholarly Data and Its Hypothetical Mapping Towards Granular Computing [J]. Journal of Computational and Theoretical Nanoscience, 2018, 24 (10): 7810-7814.
[4] Wang D, Liang Y, Xu D, et al. A content-based recommender system for computer science publications [J]. Knowledge- Based Systems, 2018, 157: 1-9.
[5] Wang L C. Research on Paper Recommendation System Based on Deep Neural Network [D]. Jilin University, 2019.
[6] Yang Z H, and Brian D. Davison. “Venue Recommendation: Submitting Your Paper with Style.” 1 (2012): 681-686.
[7] Dai J. Research and Implementation of Cooperative Recommendation Model Based on LDA in Scientific Research Achievement Management System [D]. Liaoning University, 2017.
[8] Zhang Y, Dong L L, Li S H. Extraction of Elements of Protest Based on BERT Model and TextTeaser improved algorithm [J]. Journal of Physics: Conference Series, 2021, 1955 (1).
[9] DEVLIN J, CHANG M W, LEE K, et al. BERT: Bidirectional encoder representations from transformers for language understanding [J]. Computation and Language, 2018, 23 (2): 3-19.
[10] Zhang Y, Hao J B. Malicious code detection method based on attention mechanism and residual network: 1-10, 2022.
[11] Mikolov T, Chen K, Corrado G, et al. Efficient estimation of word representations in vector space [J]. CoRR, 2013, abs/1301.3781.
[12] Fan W, Wang H M, Xing Y. Auto-encoder based multi-view attributed network representation learning model [J]. Journal of Computer Applications, 2021, 41 (04): 1064-1070.
[13] Liu S Z, Zhu Y D, Chen R Z, Luo H Y, Zhao F, Sun Y, Wang B H. Traffic mode recognition algorithm based on residual temporal attention neural network [J]. Journal of Computer Applications, 2021, 41 (06): 1557-1565.
[14] Liu H L, Ma J, Zhang G X. Review of studies on deep learning-based content recommendation algorithms [J]. Computer Engineering, 2021, 47 (7): 1-12.
[15] Rajesh T M, Kulkarni Praveen. A Multi-Model Framework for Grading of Human Emotion Using CNN and Computer Vision [J]. International Journal of Computer Vision and Image Processing (IJCVIP), 2021, 12 (1).
[16] Zhou C, Li X L, Li Q L, Zhu D D, Chen S L, Jiang L R. Sparse Non-negative Matrix Factorization Algorithm Base on Cosine Similarly [J]. Computer Science, 2020, 47 (10): 108- 113.
[17] Qiu J P, Shen C. Analysis of Hot Topics in Domestic Big Data Research Based on LDA Model [J]. Journal of Modern Information, 2021, 41 (09): 22-31.
[18] Bao X, Chen H M, Xiao Q. Time-incorporated point-of- interest collaborative recommendation algorithm [J]. Journal of Computer Applications, 2021, 41 (08): 2406-2411.
[19] Yang X L. Survey for Performance Measure Index of Classification Learning Algorithm [J]. Computer Science, 2021, 48 (08): 209-219.
[20] Dong J W, Shi Y Q, Xie B R. Adaptive Recommendation System for Personalized Learning Resource Based on Artificial Intelligence [J]. China Computer & Communication, 2021, 33 (15): 166-168.
Cite This Article
  • APA Style

    Peng Liu, Yuzhi Xiao, Youpeng Qin, Lin Zhou. (2022). Design and Implementation of Journal Recommendation Model Based on L-BERT. American Journal of Computer Science and Technology, 5(2), 134-145. https://doi.org/10.11648/j.ajcst.20220502.25

    Copy | Download

    ACS Style

    Peng Liu; Yuzhi Xiao; Youpeng Qin; Lin Zhou. Design and Implementation of Journal Recommendation Model Based on L-BERT. Am. J. Comput. Sci. Technol. 2022, 5(2), 134-145. doi: 10.11648/j.ajcst.20220502.25

    Copy | Download

    AMA Style

    Peng Liu, Yuzhi Xiao, Youpeng Qin, Lin Zhou. Design and Implementation of Journal Recommendation Model Based on L-BERT. Am J Comput Sci Technol. 2022;5(2):134-145. doi: 10.11648/j.ajcst.20220502.25

    Copy | Download

  • @article{10.11648/j.ajcst.20220502.25,
      author = {Peng Liu and Yuzhi Xiao and Youpeng Qin and Lin Zhou},
      title = {Design and Implementation of Journal Recommendation Model Based on L-BERT},
      journal = {American Journal of Computer Science and Technology},
      volume = {5},
      number = {2},
      pages = {134-145},
      doi = {10.11648/j.ajcst.20220502.25},
      url = {https://doi.org/10.11648/j.ajcst.20220502.25},
      eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.ajcst.20220502.25},
      abstract = {The output cycle and quantity of academic achievements of researchers can promote academic exchanges and integration to a certain extent. Therefore, it is particularly important to design a journal recommendation model for scientific researchers that is fast and relatively suitable for producing scientific research results. Based on this, focusing on the information knowledge of papers and the characteristics of journal topics, combined with deep learning technology, this paper proposes a journal-oriented recommendation model based on L-BERT (LDA-BERT), which realizes the purpose of recommending journals for scholars. Firstly, topic extraction is performed on papers in a specific journal, and the number of topic words is determined by perplexity. Secondly, the vectorized representation of the topic words is carried out by using the BERT model, and the fixed representation value is obtained through the mean value idea. Finally, Euclidean distance is used to determine the similarity between papers and journals. In order to verify the effect of L-BERT model, it is compared with Word2Vec model on the real data set. The results show that the recommended results of L-BERT model are improved by 17%, 10% and 13% respectively in Precision, Recall, and F1-Score, which fully shows that this research has a certain application value, and the research results can help researchers shorten the cycle of looking for journals.},
     year = {2022}
    }
    

    Copy | Download

  • TY  - JOUR
    T1  - Design and Implementation of Journal Recommendation Model Based on L-BERT
    AU  - Peng Liu
    AU  - Yuzhi Xiao
    AU  - Youpeng Qin
    AU  - Lin Zhou
    Y1  - 2022/06/09
    PY  - 2022
    N1  - https://doi.org/10.11648/j.ajcst.20220502.25
    DO  - 10.11648/j.ajcst.20220502.25
    T2  - American Journal of Computer Science and Technology
    JF  - American Journal of Computer Science and Technology
    JO  - American Journal of Computer Science and Technology
    SP  - 134
    EP  - 145
    PB  - Science Publishing Group
    SN  - 2640-012X
    UR  - https://doi.org/10.11648/j.ajcst.20220502.25
    AB  - The output cycle and quantity of academic achievements of researchers can promote academic exchanges and integration to a certain extent. Therefore, it is particularly important to design a journal recommendation model for scientific researchers that is fast and relatively suitable for producing scientific research results. Based on this, focusing on the information knowledge of papers and the characteristics of journal topics, combined with deep learning technology, this paper proposes a journal-oriented recommendation model based on L-BERT (LDA-BERT), which realizes the purpose of recommending journals for scholars. Firstly, topic extraction is performed on papers in a specific journal, and the number of topic words is determined by perplexity. Secondly, the vectorized representation of the topic words is carried out by using the BERT model, and the fixed representation value is obtained through the mean value idea. Finally, Euclidean distance is used to determine the similarity between papers and journals. In order to verify the effect of L-BERT model, it is compared with Word2Vec model on the real data set. The results show that the recommended results of L-BERT model are improved by 17%, 10% and 13% respectively in Precision, Recall, and F1-Score, which fully shows that this research has a certain application value, and the research results can help researchers shorten the cycle of looking for journals.
    VL  - 5
    IS  - 2
    ER  - 

    Copy | Download

Author Information
  • School of Computer, Qinghai Normal University, Xining, China

  • School of Computer, Qinghai Normal University, Xining, China

  • School of Computer, Qinghai Normal University, Xining, China

  • School of Computer, Qinghai Normal University, Xining, China

  • Sections