TY - JOUR

T1 - Assessing the Poetry of a Text and Its Emotional Content Using a Hybrid Approach

AU - Mengliev, Davlatyor

AU - Urinboeva, Nazokat

AU - Sharipov, Sirojbek

AU - Polatova, Sevinch

AU - Atakhanov, Mukhammadjon

AU - Khamraeva, Saida

AU - Boltayev, Nodirbek

PY - 2024/11/27

Y1 - 2024/11/27

N2 - This article presents a hybrid approach for recognizing poetic texts in the Uzbek language, which combines a dictionary approach with modern sentiment analysis methods implemented using artificial intelligence tools such as SpaCy. The article describes in detail the process of counting the number of poetic words and determining their share in the text, if exceeded (35% or more), the text is classified as poetic or containing elements of poetry. To evaluate the effectiveness of the algorithm, a number of experiments were conducted, where the results showed that the algorithm achieves an accuracy of 95% when identifying poetic words, while in mood detection tasks the accuracy is slightly lower - 89%. The authors included information about the limitations of the algorithm in the form of the size of the dictionary required to detect words. However, ways to further expand the vocabulary base to improve the accuracy of the analysis have been proposed.

AB - This article presents a hybrid approach for recognizing poetic texts in the Uzbek language, which combines a dictionary approach with modern sentiment analysis methods implemented using artificial intelligence tools such as SpaCy. The article describes in detail the process of counting the number of poetic words and determining their share in the text, if exceeded (35% or more), the text is classified as poetic or containing elements of poetry. To evaluate the effectiveness of the algorithm, a number of experiments were conducted, where the results showed that the algorithm achieves an accuracy of 95% when identifying poetic words, while in mood detection tasks the accuracy is slightly lower - 89%. The authors included information about the limitations of the algorithm in the form of the size of the dictionary required to detect words. However, ways to further expand the vocabulary base to improve the accuracy of the analysis have been proposed.

UR - https://www.scopus.com/record/display.uri?eid=2-s2.0-85212092585&origin=inward&txGid=522e346849ba3c257489f6216f7120c7

UR - https://www.mendeley.com/catalogue/e0231041-6ca3-3b5d-b761-96fd8b3e9d4a/

U2 - 10.1063/5.0241412

DO - 10.1063/5.0241412

M3 - Conference article

VL - 3244

JO - AIP Conference Proceedings

JF - AIP Conference Proceedings

SN - 0094-243X

IS - 1

M1 - 030060

T2 - 2024 International Scientific Conference on Modern Problems of Applied Science and Engineering

Y2 - 2 May 2024 through 3 May 2024

ER -