Abstract
Accurate emotion classification for online reviews is vital for business organizations to gain deeper insights into markets. Although deep learning has been successfully implemented in this area, accuracy and processing time are still major problems preventing it from reaching its full potential. This paper proposes an Enhanced Leaky Rectified Linear Unit activation and Weighted Loss (ELReLUWL) algorithm for enhanced text emotion classification and faster parameter convergence speed. This algorithm includes the definition of the inflection point and the slope for inputs on the left side of the inflection point to avoid gradient saturation. It also considers the weight of samples belonging to each class to compensate for the influence of data imbalance. Convolutional Neural Network (CNN) combined with the proposed algorithm to increase the classification accuracy and decrease the processing time by eliminating the gradient saturation problem and minimizing the negative effect of data imbalance, demonstrated on a binary sentiment problem. All work was carried out using supervised deep learning. The results for accuracy and processing time are obtained by using different datasets and different review types. The results show that the proposed solution achieves better classification performance in different data scenarios and different review types. The proposed model takes less convergence time to achieve model optimization with seven epochs against the current convergence time of 11.5 epochs on average. The proposed solution improves accuracy and reduces the processing time of text emotion classification. The solution provides an average class accuracy of 96.63% against a current average accuracy of 91.56%. It also provides a processing time of 23.3 milliseconds compared to the current average processing time of 33.2 milliseconds. Finally, this study solves the issues of gradient saturation and data imbalance. It enhances overall average class accuracy and decreases processing time.
Similar content being viewed by others
References
Bengio Y (2000) Gradient-based optimization of Hyperparameters. Neural Comput 12(8):1889–1900. https://doi.org/10.1162/089976600300015187
Bose R (2011) Discovering business intelligence from the subjective web data. Int J Bus Intell Res 2:1–16
Chatterjee A, Gupta U, Chinnakotla MK, Srikanth R, Galley M, Agrawal P (2019) Understanding emotions in text using deep learning and big data. Comput Hum Behav 93:309–317. https://doi.org/10.1016/j.chb.2018.12.029
Chen T, Xu R, He Y, Wang X (2017) Improving sentiment analysis via sentence type classification using BiLSTM-CRF and CNN. Expert Syst Appl 72(C):221–230. https://doi.org/10.1016/j.eswa.2016.10.065
Collobert R, Weston J, Bottou L, Karlen M, Kavukcuoglu K, Kuksa P (2011) Natural Language Processing (Almost) from Scratch. J. Mach. Learn. Res. 12:2493–2537
Colnerič N, Demšar J (2020) Emotion recognition on twitter: comparative study and training a unison model. IEEE Trans Affect Comput 11(3):433–446. https://doi.org/10.1109/TAFFC.2018.2807817
Dhaoui C, Webster Cynthia M, Tan Lay P (2017) Social media sentiment analysis: lexicon versus machine learning. J Consum Mark 34(6):480–488. https://doi.org/10.1108/JCM-03-2017-2141
Gu X, Gu Y, Wu H (2017) Cascaded convolutional neural networks for aspect-based opinion summary. Neural Process Lett 46(2):581–594. https://doi.org/10.1007/s11063-017-9605-7
Hanafy M, Khalil MI, and Abbas HM (2018) "Combining Classical and Deep Learning Methods for Twitter Sentiment Analysis," in Artificial Neural Networks in Pattern Recognition, Cham, L. Pancioni, F. Schwenker, and E. Trentin, Eds. Springer International Publishing, pp. 281–292.
Hara K, Saitoh D, and Shouno H (2015) "Analysis of function of rectified linear unit used in deep learning. [Online]. Available: https://doi.org/10.1109/IJCNN.2015.7280578.
Hinton G, Srivastava N, Krizhevsky A, Sutskever I and Salakhutdinov R (2012) Improving neural networks by preventing co-adaptation of feature detectors. arXiv.org
IMDB Dataset (n.d.) [Online]. Available: https://www.kaggle.com/lakshmi25npathi/imdb-dataset-of-50k-movie-reviews
Jing Y, Guanci Y (2018) Modified Convolutional Neural Network Based on Dropout and the Stochastic Gradient Descent Optimizer. Algorithms 11(3):28. https://doi.org/10.3390/a11030028
Kalchbrenner N, Grefenstette E, Blunsom P (2014) A convolutional neural network for modelling sentences
Kiritchenko S, Zhu X, Mohammad SM (2014) Sentiment analysis of short informal texts. J Artif Int Res 50(1):723–762
Kratzwald B, Ilic S, Kraus M, Feuerriegel S, and Prendinger H (2018) Deep learning for affective computing: text-based emotion recognition in decision support. arXiv.org. 115. https://doi.org/10.1016/j.dss.2018.09.002.
Kratzwald B, Ilić S, Kraus M, Feuerriegel S, Prendinger H (2018) Deep learning for affective computing: Text-based emotion recognition in decision support. Decis Supp Syst 115:24–35. https://doi.org/10.1016/j.dss.2018.09.002
Kraus M, Feuerriegel S (2017) Decision support from financial disclosures with deep neural networks and transfer learning. Decis Support Syst 104:38–48
Lin D, Li L, Cao D, Lv Y, Ke X (2018) Multi-modality weakly labeled sentiment learning based on explicit emotion signal for Chinese microblog. Neurocomputing 272:258–269. https://doi.org/10.1016/j.neucom.2017.06.078
Liu B (2020) Text sentiment analysis based on CBOW model and deep learning in big data environment. J Ambient Intell Human Comput 11(2):451–458. https://doi.org/10.1007/s12652-018-1095-6
Macêdo D, Zanchettin C, Oliveira ALI, Ludermir T (2019) Enhancing batch normalized convolutional networks using displaced rectifier linear units: a systematic comparative study. Expert Syst Appl 124:271–281. https://doi.org/10.1016/j.eswa.2019.01.066
Mahmoudi N, Docherty P, Moscato P (2018) Deep neural networks understand investors better. Decis Support Syst 112:23–34
Minglei L, Qin L, Yunfei L, Lin G (2017) Inferring affective meanings of words from word embedding. T-AFFC 8(4):443–456. https://doi.org/10.1109/TAFFC.2017.2723012
Mohammad SM, Kiritchenko S, Zhu X (2013) "NRC-Canada : building the state-of-the-art in sentiment analysis of tweets," presented at the proceedings of the seventh international workshop on semantic evaluation exercises, 2013/07/01
Mou L et al. (2016) "How Transferable are Neural Networks in NLP Applications?," arXiv.org
Mundra S, Sen A, Sinha M, Mannarswamy S, Dandapat S, and Roy S (2017) "Fine-Grained Emotion Detection in Contact Center Chat Utterances," in Advances in Knowledge Discovery and Data Mining, Cham, J. Kim, K. Shim, L. Cao, J.-G. Lee, X. Lin, and Y.-S. Moon, Eds.: Springer International Publishing, pp. 337–349
Pang B, Lee L (2004) "A sentimental education: sentiment analysis using subjectivity summarization based on minimum cuts," presented at the Proceedings of the 42nd Annual Meeting on Association for Computational Linguistics, Barcelona, Spain Online Available. https://doi.org/10.3115/1218955.1218990
Phan D-A, Matsumoto Y, Shindo H (2018) Autoencoder for Semisupervised multiple emotion detection of conversation transcripts. T-AFFC 12:1–691. https://doi.org/10.1109/TAFFC.2018.2885304
Phillips P, Sui Y, Liu B, Yang M, Cheng H (2018) Classification of Alzheimer’s disease based on eight-layer convolutional neural network with leaky rectified linear unit and max pooling. J Med Syst 42(5):1–11. https://doi.org/10.1007/s10916-018-0932-7
Poernomo A, Kang D-K (2018) Biased dropout and Crossmap dropout: learning towards effective dropout regularization in convolutional neural network. Neural Netw 104:60–67. https://doi.org/10.1016/j.neunet.2018.03.016
Qian S, Liu H, Liu C, Wu S, Wong HS (2018) Adaptive activation functions in convolutional neural networks. Neurocomputing 272:204–212. https://doi.org/10.1016/j.neucom.2017.06.070
Randhawa S, Alsadoon A, Prasad PWC, Al-Dala’in T, Dawoud A, Alrubaie A (2021) Deep learning for liver tumour classification: enhanced loss function. Multimedia Tools Appl 80(3):4729–4750. https://doi.org/10.1007/s11042-020-09900-8
Rao G, Huang W, Feng Z, Cong Q (2018) LSTM with sentence representations for document-level sentiment classification. Neurocomputing 308:49–57. https://doi.org/10.1016/j.neucom.2018.04.045
Srivastava N, Hinton G, Krizhevsky A, Sutskever I, Salakhutdinov R (2014) Dropout: a simple way to prevent neural networks from overfitting. J Mach Learn Res 15(1):1929–1958
Steno P, Alsadoon A, Prasad PWC, Al-Dala’in T, Alsadoon OH (2020) A novel enhanced region proposal network and modified loss function: threat object detection in secure screening using deep learning. J Supercomput. https://doi.org/10.1007/s11227-020-03418-4
Stojanovski D, Strezoski G, Madjarov G, Dimitrovski I, Chorbev I (2018) Deep neural network architecture for sentiment analysis and emotion identification of twitter messages. Multimed Tools Appl 77(24):32213–32242. https://doi.org/10.1007/s11042-018-6168-1
Sun X (2020) A novel approach to generate a large scale of supervised data for short text sentiment analysis. Multimed Tools Appl 79(9–10):5439–5459. https://doi.org/10.1007/s11042-018-5748-4
Sun X, Zhang C (2020) Detecting anomalous emotion through big data from social networks based on a deep learning method. Multimed Tools Appl 79(13–14):9687–9687. https://doi.org/10.1007/s11042-018-5665-6
Sun X, Li C, Ren F (2016) Sentiment analysis for Chinese microblog based on deep neural networks with convolutional extension features. Neurocomputing 210:227–236. https://doi.org/10.1016/j.neucom.2016.02.077
Tang D, Qin B, Feng X, Liu T (2015) Effective LSTMs for target-dependent sentiment classification
Tang Y (2015) "Deep Learning using Linear Support Vector Machines," arXiv.org
Vinay Kumar J, Shishir K, Prabhat M (2018) Sentiment recognition in customer reviews using deep learning. Int J Enterprise Inform Syst (IJEIS) 14(2):77–86. https://doi.org/10.4018/IJEIS.2018040105
Wu H, Gu X, Gu Y (2017) Balancing between over-weighting and under-weighting in supervised term weighting. Inf Process Manage 53(2):547–557. https://doi.org/10.1016/j.ipm.2016.10.003
Xiong S, Lv H, Zhao W, Ji D (2018) Towards twitter sentiment classification by multi-level sentiment-enriched word embeddings. Neurocomputing 275:2459–2466. https://doi.org/10.1016/j.neucom.2017.11.023
Xu R, Chen T, Xia Y, Lu Q, Liu B, Wang X (2015) Word Embedding Composition for Data Imbalances in Sentiment and Emotion Classification. Cogn Comput 7(2):226–240. https://doi.org/10.1007/s12559-015-9319-y
Yoo S, Song J, Jeong O (2018) Social media contents based sentiment analysis and prediction system. Expert Syst Appl 105:102–111. https://doi.org/10.1016/j.eswa.2018.03.055
Yoon K (2014) "convolutional neural networks for sentence classification," ed. Cornell University Library, arXiv.org, Ithaca
Yu L, Zhang W, Wang J, Yu Y (2016) SeqGAN: sequence generative adversarial nets with policy gradient
Zhang Y and Wallace B (2016) A Sensitivity Analysis of (and Practitioners' Guide to) Convolutional Neural Networks for Sentence Classification. arXiv.org.
Zhang Y, Zhang Z, Miao D, Wang J (2019) Three-way enhanced convolutional neural networks for sentence-level sentiment classification. Inf Sci 477:55–64. https://doi.org/10.1016/j.ins.2018.10.030
Zhang Z, Zou Y, Gan C (2018) Textual sentiment analysis via three different attention convolutional neural networks and cross-modality consistent regression. Neurocomputing 275:1407–1415. https://doi.org/10.1016/j.neucom.2017.09.080
Zhao H, Liu F, Li L, Luo C (2018) A novel softplus linear unit for deep convolutional neural networks. Appl Intell 48(7):1707–1720. https://doi.org/10.1007/s10489-017-1028-7
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Yang, H., Alsadoon, A., Prasad, P.W.C. et al. Deep learning neural networks for emotion classification from text: enhanced leaky rectified linear unit activation and weighted loss. Multimed Tools Appl 81, 15439–15468 (2022). https://doi.org/10.1007/s11042-022-12629-1
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-022-12629-1