Abstract
Loss functions play a critical role in evaluating the performance of a model trained on specific parameters of the dataset. In simpler terms, loss functions serve as a penalty for a bad prediction to improve the prediction with its testing values. However, in the case of a prediction on an imbalanced dataset, the loss function must also be modified so that the weights of the loss that occurred due to misclassification do not change. In contrast, the weights in case of loss due to correct classification were reduced. Different developers have brought up a variety of loss functions to curb this problem that occurred due to an imbalanced dataset, one of them being the focal loss function. This focal loss function will be compared with the normal cross-entropy loss function, widely used in the evaluation. The final comparison between the two-loss functions will help determine whether the change of loss function can create how much difference in the model performance.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Malavolta I, Ruberto S, Soru T, Terragni V (2015) Hybrid mobile apps in the google play store: an exploratory investigation. In: 2nd ACM international conference on mobile software engineering and systems, pp. 56–59
Viennot N, Garcia E, Nieh J (2014) A measurement study of google play. ACM SIGMETRICS Perform Eval Rev 42(1), 221–233
McIlroy S, Shang W, Ali N, Hassan AE (2017) Is it worth responding to reviews? Studying the top free apps in Google Play. IEEE Softw 34(3):64–71
Shashank S, Naidu B (2020) Google play store apps—data analysis and ratings prediction. Int Res J Eng Technol (IRJET) 7:265–274
Arxiv A Longitudinal study of Google Play page, https://arxiv.org/abs/1802.02996, Accessed 21 Dec 2021
Patil HP, Atique M (2015) Sentiment analysis for social media: a survey. In: 2nd international conference on information science and security (ICISS), pp. 1–4
Zainuddin N, Selamat, A.: Sentiment analysis using support vector machine. In: International conference on computer, communications, and control technology (I4CT) 2014, pp. 333–337
Dubey A, Rasool A (2021) Efficient technique of microarray missing data imputation using clustering and weighted nearest neighbor. Sci Rep 11(1)
Li X, Wang X, Liu H (2021) Research on fine-tuning strategy of sentiment analysis model based on BERT. In: International conference on communications, information system and computer engineering (CISCE), pp. 798–802
Mohammadian S, Karsaz A, Roshan YM (2017) A comparative analysis of classification algorithms in diabetic retinopathy screening. In: 7th international conference on computer and knowledge engineering (ICCKE) 2017, pp. 84–89
Latif R, Talha Abdullah M, Aslam Shah SU, Farhan M, Ijaz F, Karim A (2019) Data scraping from Google Play Store and visualization of its content for analytics. In: 2nd international conference on computing, mathematics and engineering technologies (iCoMET) 2019, pp. 1–8
Day M, Lin Y (2017) Deep learning for sentiment analysis on Google Play consumer review. IEEE Int Conf Inf Reuse Integr (IRI) 2017:382–388
Abdul Khalid KA, Leong TJ, Mohamed K (2016) Review on thermionic energy converters. IEEE Trans Electron Devices 63(6):2231–2241
Regulin D, Aicher T, Vogel-Heuser B (2016) Improving transferability between different engineering stages in the development of automated material flow modules. IEEE Trans Autom Sci Eng 13(4):1422–1432
Li D, Qian J (2016) Text sentiment analysis based on long short-term memory. In: First IEEE international conference on computer communication and the internet (ICCCI) 2016, pp. 471–475 (2016)
Lin T, Goyal P, Girshick R, He K, Dollár P (2020) Focal loss for dense object detection. IEEE Trans Pattern Anal Mach Intell 42(2):318–327
Arxiv A Sentimental Education: Sentiment Analysis Using Subjectivity Summarization Based on Minimum Cuts, https://arxiv.org/abs/cs/0409058, Accessed 21 Dec 2021
Sfu Webpage Methods for Creating Semantic Orientation Dictionaries, https://www.sfu.ca/~mtaboada/docs/publications/Taboada_et_al_LREC_2006.pdf, Accessed 21 Dec 2021
Sudhir P, Suresh VD (2021) Comparative study of various approaches, applications and classifiers for sentiment analysis. Glob TransitS Proc 2(2):205–211
Gillioz A, Casas J, Mugellini E, Khaled OA (2020) Overview of the transformer-based models for NLP tasks. In: 15th conference on computer science and information systems (FedCSIS) 2020, pp. 179–183
Zhou Y, Li M (2020) Online course quality evaluation based on BERT. In: 2020 International conference on communications, information system and computer engineering (CISCE) 2020, pp. 255–258
Truong TL, Le HL, Le-Dang TP (2020) Sentiment analysis implementing BERT-based pre-trained language model for Vietnamese. In: 7th NAFOSTED conference on information and computer science (NICS) 2020, pp. 362–367 (2020)
Kano T, Sakti S, Nakamura S (2021) Transformer-based direct speech-to-speech translation with transcoder. IEEE spoken language technology workshop (SLT) 2021, pp. 958–965
Arxiv Comparing BERT against traditional machine learning text classification, https://arxiv.org/abs/2005.13012, Accessed 21 Dec 2021
Arxiv A Comparison of LSTM and BERT for Small Corpus, https://arxiv.org/abs/2009.05451, Accessed 21 Dec 2021
Arxiv BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding, https://arxiv.org/abs/1810.04805, Accessed 21 Dec 2021
Naseer M, Asvial M, Sari RF (2021) An empirical comparison of BERT, RoBERTa, and Electra for fact verification. In: International conference on artificial intelligence in information and communication (ICAIIC) 2021, pp. 241–246
Ho Y, Wookey S (2020) The real-world-weight cross-entropy loss function: modeling the costs of mislabeling. IEEE Access 8:4806–4813
Zhou Y, Wang X, Zhang M, Zhu J, Zheng R, Wu Q (2019) MPCE: a maximum probability based cross entropy loss function for neural network classification. IEEE Access 7:146331–146341
Yessou H, Sumbul G, Demir B (2020) A Comparative study of deep learning loss functions for multi-label remote sensing image classification. IGARSSIEEE international geoscience and remote sensing symposium 2020, pp. 1349–1352
Liu L, Qi H (2017) Learning effective binary descriptors via cross entropy. In: IEEE winter conference on applications of computer vision (WACV) 2017, pp. 1251–1258 (2017)
Riquelme N, Von Lücken C, Baran B (2015) Performance metrics in multi-objective optimization. In: Latin American Computing Conference (CLEI) 2015, pp. 1–11
Dubey A, Rasool A (2020) Clustering-based hybrid approach for multivariate missing data imputation. Int J Adv Comput Sci Appl (IJACSA) 11(11):710–714
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Vyas, P., Sharma, M., Rasool, A., Dubey, A. (2023). Comparative Study of Loss Functions for Imbalanced Dataset of Online Reviews. In: Singh, P., Singh, D., Tiwari, V., Misra, S. (eds) Machine Learning and Computational Intelligence Techniques for Data Engineering. MISP 2022. Lecture Notes in Electrical Engineering, vol 998. Springer, Singapore. https://doi.org/10.1007/978-981-99-0047-3_11
Download citation
DOI: https://doi.org/10.1007/978-981-99-0047-3_11
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-99-0046-6
Online ISBN: 978-981-99-0047-3
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)