INTELLIGENT MODEL DEVELOPMENT FOR RECOGNIZING EMOTIONS IN TEXT
DOI:
https://doi.org/10.26906/SUNZ.2020.4.077Keywords:
recognition of emotions in the text, intelligent model, semantic analysis of the text, model buildingAbstract
The topical issue of human emotions recognition in the text is considered. The offers of available products on the market that offer similar functionality are analyzed. The possibilities of implementing the model for recognizing emotions in the text are investigated. The program functions interact with the model and the requirements for the model itself are determined. The steps sequence for obtaining data for training the model and its own implementation is given. The developed architectural model is presented, the programming language use is selected and substantiated, the methods and approaches use for training the model is selected and justified, the initial data and options sources for obtaining data for training the model are considered. A model and a data set for it are proposed, which will make it possible to obtain recognition of some emotions in the text with a certain accuracy. Particular attention is paid to the possibility of developing the application and improving this approach and model in the future for the possibility of using machines to better serve people, understanding the consumer, measuring the happiness level among the population, understanding the audience mood
Downloads
References
Emotion Analysis:[Електронний ресурс].– Режим доступа: https://www.twinword.com/api/emotion-analysis.php (дата звернення 30.08.2020). – Назва з екрана.
Text analytics by Microsoft: [Електронний ресурс]: –Режим доступа: https://azure.microsoft.com/ru-ru/services/ cognitive-services/text-analytics/#features (дата звернення 14.08.2020). – Назва з екрана.
IBM Watson Tone Analyzer:[Електронний ресурс].– Режим доступа:https://tone-analyzer-demo.ng.bluemix.net(дата звернення 25.09.2020). – Назва з екрана.
MonkeyLearn sentiment analysis:[Електронний ресурс].– Режим доступа: https://monkeylearn.com/sentiment-analysis(дата звернення 30.09.2020). – Назва з екрана.
Batbaatar E. Semantic-Emotion Neural Network for Emotion Recognition From Text /E.Batbaatar, M.Li and Ryu K.H. // IEEE Access. – vol. 7. –2019.–pp. 111866-111878, doi: 10.1109/ACCESS.2019.2934529.
Gruber,N. Are GRU cells more specific and LSTM cells more sensitive in motive classification of text?/ Gruber,N.; Jockisch A. //Frontiers in Artificial Intelligence.– No 3–2020. doi:10.3389/frai.2020.00040.
Aurélien GéronHands-on Machine Learning with Scikit-Learn, Keras, and TensorFlow.– Sebastopol, CA: O'Reilly Media. 2019. –pp. 448.
Saon, G., & Picheny, M. (2017). Recent advances in conversational speech recognition using convolutional and recurrent neural networks. IBM Journal of Research and Development, 61(4/5), 1-1.
Gupta A. et al. Generative recurrent networks for de novo drug design //Molecular informatics. – 2018. – Т. 37. – No. 1-2. – С. 1700111.
Sentiment Analysis: Emotion in Text:[Електронний ресурс].– Режим доступа: https://www.kaggle.com/c/sa-emotions/data (дата звернення 01.11.2020). – Назва з екрана.
Complete Tweet Sentiment Extraction Data:[Електронний ресурс].– Режим доступа: https://www.kaggle.com/maxjon/complete-tweet-sentiment-extraction-data
The GoEmotions dataset:[Електронний ресурс].– Режим доступа: https://github.com/google-research/google-research/tree/master/goemotions (дата звернення 17.07.2020). – Назва з екрана.
Python is a programming language that lets you work more quickly and integrate your systems more effectively:[Електронний ресурс].– Режим доступа:: https://www.python.org/ (дата звернення 20.09.2020). – Назва з екрана
Ben M. Packaging data analytical work reproducibly using R (and friends) / M. Ben, C. Boettiger and L. Mullen // The American Statistician. –No 72.1. – (2018).– р. 80-88.
