Abstract
To reliably and accurately classify complicated "big" datasets, machine learning models must be continually improved. This research proposes straightforward yet competitive neural networks for text classification, even though graph neural networks (GNN) have reignited interest in graph-based text classification models. Convolutional neural networks (CNN), artificial neural networks (ANN), and their refined “fine-tuned” models (denoted as FT-CNN and FT-ANN) are the names given to our proposed models. The models presented in this paper demonstrate that our simple models like (CNN, ANN, FT-CNN, and FT-ANN) can perform better than more complex GNN ones such as (SGC, SSGC, and TextGCN) and are comparable to others (i.e., HyperGAT and Bert). The process of fine-tuning is also highly recommended because it improves the performance and reliability of models. The performance of our suggested models on five benchmark datasets (namely, Reuters (R8), R52, 20NewsGroup, Ohsumed, and Mr) is vividly illustrated. According to the experimental findings, on the majority of the target datasets, these models—especially those that have been fine-tuned—perform surprisingly better than SOTA approaches, including GNN-based models.
Similar content being viewed by others
Data availability statement
The datasets used in this work are publicly available. When the paper is accepted, the source code would exist in our Github repository (https://github.com/aliamer).
Abbreviations
- SOTA:
-
State-of-the-art
- BoW:
-
Bag of Words
- TFIDF:
-
Term frequency—Inverse document frequency
- VSM:
-
Vector space model
- GNN:
-
Graph neural networks
- CNN:
-
Convolutional neural networks
- ANN:
-
Artificial neural networks
- CNN-rand:
-
CNN randomly initializes word embeddings
- CNN-pre:
-
CNN uses pretrained word embeddings.
- FT-CNN:
-
Fine-tuned Convolutional neural networks
- FT-ANN:
-
Fine-tuned artificial neural networks
- SGC:
-
Simplifying Graph Convolutional Networks
- SSGC:
-
Simple Spectral Graph Convolutional Networks
- TextGCN:
-
Text Graph Convolutional Network
- HyperGAT:
-
Hypergraph Attention Networks
- BERT:
-
Bidirectional Encoder Representations from Transformers
- DAN:
-
Deep Averaging Networks
- MLP:
-
Multilayer Perceptron
- HeteGCN:
-
Heterogeneous graph convolutional network
- TensorGCN:
-
Tensor Graph Convolutional Networks
- GAT:
-
Graph Attention Network
- LSTM:
-
Long short-term memory
- Bi-LSTM:
-
Bidirectional LSTM
- GloVe:
-
Global vectors for word representation
- ReLU:
-
Rectified linear unit
- SVM:
-
Support vector machine
- SWEM:
-
Simple word embedding-based model
- NGNN:
-
Network in graph neural network model
- T-VGAE:
-
Topic variational graph auto-encoder
- CFE:
-
Category-based feature engineering model
- Han-LT:
-
Heterogeneous attention network for semi-supervised long text classification.
- InducT-GCN:
-
An inductive text classification model based on GCN
- Syntax-AT-Capsule:
-
An enhanced capsule network text classification model
References
Frank M, Drikakis D, Charissis V (2020) Machine-learning methods for computational science and engineering. Computation 8(1):15
Abdalla HI, Amer AA (2022) On the integration of similarity measures with machine learning models to enhance text classification performance. Inf Sci 614:263–288
Diera A, Lin BX, Khera B, Meuser T, Singhal T, Galke L, Scherp A (2022) Bag-of-words vs. sequence vs. graph vs. hierarchy for single-and multi-label text classification. arXiv preprint arXiv:2204.03954
Wu F, Souza A, Zhang T, Fifty C, Yu T, Weinberger, K (2019) Simplifying graph convolutional networks. In International conference on machine learning (pp. 6861–6871). PMLR
Zhu, H., & Koniusz, P. (2021, May). Simple spectral graph convolution. In International conference on learning representations
Ruan S, Chen B, Song K, Li H (2022) Weighted naïve Bayes text classification algorithm based on improved distance correlation coefficient. Neural Comput Appl 34(4):2729–2738
Zhang L, Jiang L, Li C (2019) A discriminative model selection approach and its application to text classification. Neural Comput Appl 31(4):1173–1187
Jiang M, Liang Y, Feng X, Fan X, Pei Z, Xue Y, Guan R (2018) Text classification based on deep belief network and softmax regression. Neural Comput Appl 29(1):61–70
Mirończuk MM, Protasiewicz J (2018) A recent overview of the state-of-the-art elements of text classification. Expert Syst Appl 106:36–54
Liu S, Nimah I, Menkovski V, Mocanu DC, Pechenizkiy M (2021) Efficient and effective training of sparse recurrent neural networks. Neural Comput Appl 33(15):9625–9636
Guo S, Yao N (2020) Generating word and document matrix representations for document classification. Neural Comput Appl 32(14):10087–10108
Jacob D, Ming-Wei C, Kenton L, Kristina T (2019) BERT: pre-training of deep bidirectional transformers for language understanding. n NAACL-HLT (1), pages 4171–4186.ACL
Victor S, Lysandre D, Julien C, Thomas W (2019) Distilbert, a distilled version of BERT: smaller, faster, cheaper and lighter. CoRR, abs/1910.01108
Yi T, Mostafa D, Dara B, Donald M (2020) Efficient transformers: a survey. CoRR, abs/2009.06732
Quentin Fournier, Gaétan Marceau Caron, and Daniel Aloise. 2021. A practical survey on faster and lighter transformers. CoRR, abs/2103.14636
Tolstikhin IO, Houlsby N, Kolesnikov A, Beyer L, Zhai X, Unterthiner T, Yung J, Steiner A, Keysers D, Uszkoreit J et al (2021) MLP-mixer: an all- MLP architecture for vision. Adv Neural Inform Process Syst 34:24261
Hanxiao L, Zihang D, David RS, Quoc VL (2021a) Pay attention to MLPs. CoRR, abs/2105.08050
Rahul R, Sundararajan S, Arun I, Ramakrishna B, Vijay L (2021) HeteGCN: heterogeneous graph convolutional networks for text classification. In WSDM, pages 860–868. ACM
Hui L, Danqing Z, Bing Y, Xiaodan Z (2021b) Improving pretrained models for zero-shot multi-label text classification through reinforced label hierarchy reasoning. arXiv preprint arXiv:2104.01666
Huang L, Ma D, Li S, Zhang X, Wang H (2019) Text level graph neural network for text classification. arXiv preprint. arXiv:1910.02356
Xie Q, Huang J, Du P, Peng M, Nie JY (2021). Inductive topic variational graph auto-encoder for text classification. In: proceedings of the 2021 conference of the North American chapter of the association for computational linguistics: human language technologies (pp. 4218–4227)
Attieh J, Tekli J (2023) Supervised term-category feature weighting for improved text classification. Knowl-Based Syst 261:110215
Kaize D, Jianling W, Jundong L, Dingcheng L, Huan L (2020) Be more with less: hypergraph attention networks for inductive text classification. In EMNLP (1), pages 4927–4936. ACL
Bojanowski P, Grave E, Joulin A, Mikolov T (2017) Enriching word vectors with subword information. Trans Assoc Comput Linguist 5:135–146
Mohit I, Varun M, Jordan LB-G, Hal D (2015) Deep unordered composition rivals syntactic methods for text classification. In ACL 1:1681–1691
Zhang Y, Jin R, Zhou ZH (2010) Understanding bag-of-words model: a statistical framework. Int J Mach Learn Cybern 1(1–4):43–52. https://doi.org/10.1007/s13042-010-0001-0
Wang S, Manning CD (2012) Baselines and bigrams: simple, good sentiment and topic classification. In 50th annual meeting of the association for computational linguistics, ACL 2012 - proceedings of the conference (Vol. 2, pp. 90–94).
Kim Y (2014) Convolutional neural networks for sentence classification. In EMNLP 2014 - 2014 conference on empirical methods in natural language processing, proceedings of the conference (pp. 1746–1751). Association for computational linguistics (ACL). https://doi.org/10.3115/v1/d14-1181
Tai KS, Socher R, Manning CD (2015) Improved semantic representations from tree-structured long short-Term memory networks. In ACL-IJCNLP 2015 - 53rd annual meeting of the association for computational linguistics and the 7th international joint conference on natural language processing of the Asian federation of natural language processing, proceedings of the conference (Vol. 1, pp. 1556–1566). Association for computational linguistics (ACL). https://doi.org/10.3115/v1/p15-1150
Liu P, Qiu X, Xuanjing H (2016) Recurrent neural network for text classification with multi-task learning. In IJCAI international joint conference on artificial intelligence (Vol. 2016-January, pp. 2873–2879). International joint conferences on artificial intelligence.
Wang Y, Huang M, Zhao L, Zhu X (2016) Attention-based LSTM for aspect-level sentiment classification. In EMNLP 2016-conference on empirical methods in natural language processing, proceedings (pp. 606–615). Association for Computational Linguistics (ACL). https://doi.org/10.18653/v1/d16-1058
Yao L, Mao C, Luo Y (2019) Graph convolutional networks for text classification. In 33rd AAAI conference on artificial intelligence, AAAI 2019, 31st innovative applications of artificial intelligence conference, IAAI 2019 and the 9th AAAI symposium on educational advances in artificial intelligence, EAAI 2019 (pp. 7370–7377). AAAI Press. https://doi.org/10.4000/books.aaccademia.4577
Peng H, Li J, He Y, Liu Y, Bao M, Wang L, Yang Q (2018) Large-scale hierarchical text classification with recursively regularized deep Graph-CNN. In the web conference 2018-proceedings of the World Wide Web conference, WWW 2018. Association for Computing Machinery, Inc. (pp. 1063–1072) https://doi.org/10.1145/3178876.3186005
Zhang Y, Yu X, Cui Z, Wu S, Wen Z, Wang L (2020) Every document owns its structure: inductive text classification via graph neural networks. Associat Computat Linguist (ACL). https://doi.org/10.18653/v1/2020.acl-main.31
Joulin A, Grave E, Bojanowski P, Mikolov T (2017) Bag of tricks for efficient text classification. EACL 2:427–431
Shen D, Wang G, Wang W, Min MR, Qinliang S, Zhang Y, Li C, Henao R, Carin L (2018) Baseline needs more love: on simple wordembedding-based models and associated pooling mechanisms. ACL 1:440–450
Peng Zhou, Zhenyu Qi, Suncong Zheng, Jiaming Xu, Hongyun Bao, and Bo Xu. 2016. Text classification improved by integrating bidirectional LSTM with two-dimensional max pooling. In COLING, pages 3485–3495. ACL
Siwei L, Liheng X, Kang L, Jun Z (2015) Recurrent convolutional neural networks for text classification. In AAAI, pages 2267–2273. AAAI Press
Santiago G-C, Eduardo CG-M (2020) Comparing BERT against traditional machine learning text classification. CoRR, abs/2005.13012
Yukio O, Nels EB, Masahiko Y (1998) Keygraph: Automatic indexing by co-occurrence graph based on building construction metaphor. In ADL, pages 12–18. IEEE Computer Society
Lu Z, Jiandong D, Yi X, Yingyao L, Shuigeng Z (2021) Weakly-supervised text classification based on keyword graph. In EMNLP (1), pages 2803–2813. Association for Computational Linguistics
Hamilton WL (2020) Graph representation learning. Springer International Publishing, Cham
Jian T, Meng Q, Qiaozhu M (2015) PTE: predictive text embedding through large-scale heterogeneous text networks. In KDD, pages 1165– 1174. ACM
Thomas N. Kipf and Max Welling (2017) Semisupervised classification with graph convolutional networks. In ICLR (Poster). OpenReview.net.
Kingma DP, Ba J (2014). Adam: a method for stochastic optimization. arXiv preprint arXiv:1412.6980
Hamilton WL, Ying R, Leskovec J (2017) Inductive representation learning on large graphs. In: advances in neural information processing systems (Vol. 2017-December, pp. 1025–1035). Neural information processing systems foundation
Humayun MA, Yassin H, Shuja J, Alourani A, Abas PE (2022) A transformer fine-tuning strategy for text dialect identification. Neural Comput Appl 35:1–10
Naser A, Aydemir O (2022) Classification of pleasant and unpleasant odor imagery EEG signals. Neural Comput Appl. https://doi.org/10.1007/s00521-022-08171-8
Wang Z, Bai Y, Zhou Y, Xie C (2022) Can CNNs Be More Robust Than Transformers? arXiv preprint arXiv:2206.03452
Fey M, Lenssen JE, Weichert F, Leskovec J (2021). Gnnautoscale: scalable and expressive graph neural networks via historical embeddings. In: international conference on machine learning (pp. 3294–3304). PMLR
Haonan L, Huang SH, Ye T, Xiuyan G (2019) Graph star net for generalized multi-task learning. arXiv preprint arXiv:1906.12330
Pham P, Nguyen LT, Pedrycz W, Vo B (2022) Deep learning, graph-based text representation and classification: a survey, perspectives and challenges. Artific Intell Rev 12:1–35
Galke L, Scherp A (2022) Bag-of-words vs. graph vs. sequence in text classification: Questioning the necessity of text-graphs and the surprising strength of a wide MLP. In Proceedings of the 60th annual meeting of the association for computational linguistics (Volume 1: Long Papers) (pp. 4038–4051)
Brown T, Mann B, Ryder N, Subbiah M, Kaplan JD, Dhariwal P, Amodei D (2020) Language models are few-shot learners. Adv Neural Inf Process Syst 33:1877–1901
Rajpurkar P, Zhang J, Lopyrev K, Liang P (2016) Squad: 100,000+ questions for machine comprehension of text. arXiv preprint arXiv:1606.05250
Wang A, Singh A, Michael J, Hill F, Levy O, Bowman SR (2018) GLUE: a multi-task benchmark and analysis platform for natural language understanding. arXiv preprint arXiv:1804.07461
Poulinakis K, Drikakis D, Kokkinakis IW, Spottswood SM (2023) Machine-learning methods on noisy and sparse data. Mathematics 11(1):236
Du J, Huang Y, Moilanen K (2020) Pointing to select: a fast pointer-LSTM for long text classification. In: proceedings of the 28th international conference on computational linguistics (pp. 6184–6193)
Lin, M., & Qiang Chen, S. Y. (2012). Imagenet classification with deep convolutional neural networks.
Wang K, Han SC, & Poon J (2022) InducT-GCN: Inductive graph convolutional networks for text classification. arXiv preprint arXiv:2206.00265
Jia X, Wang L (2022) Attention enhanced capsule network for text classification by encoding syntactic dependency trees with graph convolutional neural network. Peer J Computer Science 7:e831. https://doi.org/10.7717/peerj-cs.831
Ai W, Wang Z, Shao H, Meng T, Li K (2023) A multi-semantic passing framework for semi-supervised long text classification. Appl Intell. https://doi.org/10.1007/s10489-023-04556-x
Acknowledgements
The authors would like to express their gratitude for Research Office (Zayed University) for financing and providing the tools needed to complete this work. Research Incentive Fund (RIF) Grant Activity Code: R22083-Zayed University, UAE.
Funding
This work has been supported by Research Incentive Fund (RIF) Grant Activity Code: R22083—Zayed University, UAE.
Author information
Authors and Affiliations
Contributions
HIA has been a key contributor in conception, design, analyzing, drafting the results of all experiments, and revising the final version of the manuscript. AAA has been a key contributor in conception and design, implementing the approach, analyzing the results of all experiments, and the preparation, writing, and revising of the final version of the manuscript. SDR has contributed by analyzing the results, drafting the manuscript, and reviewing the final version of this manuscript.
Corresponding author
Ethics declarations
Conflict of interest
The authors declare that they have no competing interests.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Abdalla, H.I., Amer, A.A. & Ravana, S.D. BoW-based neural networks vs. cutting-edge models for single-label text classification. Neural Comput & Applic 35, 20103–20116 (2023). https://doi.org/10.1007/s00521-023-08754-z
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00521-023-08754-z