Đăng nhập
 
Tìm kiếm nâng cao
 
Tên bài báo
Tác giả
Năm xuất bản
Tóm tắt
Lĩnh vực
Phân loại
Số tạp chí
 

Bản tin định kỳ
Báo cáo thường niên
Tạp chí khoa học ĐHCT
Tạp chí tiếng anh ĐHCT
Tạp chí trong nước
Tạp chí quốc tế
Kỷ yếu HN trong nước
Kỷ yếu HN quốc tế
Book chapter
Tạp chí quốc tế 2024
Số tạp chí 1(2024) Trang: 66-72
Tạp chí: ICMLSC '24: Proceedings of the 2024 8th International Conference on Machine Learning and Soft Computing

Interpreting and explaining complex models such as ensemble machine learning models for opinion mining is essential to increase the level of transparency fairness and reliability of positive and negative opinion prediction results. Although ensemble learning models offer significant benefits, their lack of interpretability poses a major challenge in understanding the rationale behind their prediction, creating a complex problem related to the interpretation of the model. There is also limited research on developing ensemble learning models that describe the internal function and behavior of the model. In this paper, we propose a new approach for opinion mining with random density forest interpretation to provide explanatory power in opinion mining. Using the Local Interpretable Model-agnostic Explanation (LIME), we further interpret the random density forest model leading to the prediction of opinion polarization in opinion mining according to specific domains related to online reviews of restaurants and hotels. It has demonstrated accurate results in terms of the contribution of opinion features in mining the overall opinion. In addition, we also compared the probability density of opinion feature words and were interested in the contribution of essential features to the results. Model prediction using the SHAPLEY value, based on the interaction value of opinion feature words, has shown the level of influence in predicting positive or negative opinion polarization results. Empirical results show that the proposed system tries to explain efficiency.

Các bài báo khác
Số tạp chí 2(2024) Trang: 523-544
Tác giả: Cao Nhất Linh
Tạp chí: Revista Jurídica unicuritiba
Số tạp chí Volume XI(2024) Trang: 618-623
Tạp chí: International Journal of Research and Scientific Innovation (IJRSI) (ISSN 2321-2705)
Số tạp chí 2024(2024) Trang: 15
Tác giả: Lê Thanh Tùng
Tạp chí: Communications in Optimization Theory
Số tạp chí 5(2024) Trang: 234-250
Tạp chí: Journal of Research on English and Language Learning


Vietnamese | English






 
 
Vui lòng chờ...