Irish Dhesyrie I. Gagante, Mark Dave T. Limpin. 4 0
An enhancement of fuzzy support vector machine applied to sentiment analysis in students faculty evaluation system ofPamantasan ng Lungsod ng Maynila. 6 6 - - - - - - - - . - . - 0 . - . - 0 .
Undergraduate Thesis : (Bachelor of Science major in Computer Science) : Pamantasan ng Lungsaod ng Maynila, 2024.
5
ABSTRACT: This study addresses fundamental challenges in sentiment analysis by enhancing the effectiveness of Fuzzy Support Vector Machines (FSVM). We propose novel methodologies to improve FSVM's interpretability, generality, and accuracy in sentiment classification. By integrating Bayesian rules, employing feature selection techniques, and optimizing fuzzy membership functions within a fuzzy logic frameqork, our approach aims to overcome the limitations of traditional FSVM. Through thorough testing and evaluation, our enhanced FSVM algorithm demonstrates superior adherence to predefined parameters, prioritizing precision and reliability in sentiment analysis tasks compares to existing methods. Specifically, we focus on refining feature selection to enhance model adaptability, incorporating fuzzy membership functions to optimize performance, and introducing a probabilistics framework for more accurate classification. Our findings showcase the effectiveness of these enhancements in delivering dependable sentimnent analysis results, outperforming existing approaches. While acknowledging potential influences of data quality and dataset characteristics, our study contributes to advancing text analysis techniques, particularly in sentiment analysis, with broader applications across various sectors.
5
2 = =
2
2 --0------
6 --0-- 2 --------
0 2 --
--20------
--------20--
--------20--
----2
/ 2
/ 2
/
/
An enhancement of fuzzy support vector machine applied to sentiment analysis in students faculty evaluation system ofPamantasan ng Lungsod ng Maynila. 6 6 - - - - - - - - . - . - 0 . - . - 0 .
Undergraduate Thesis : (Bachelor of Science major in Computer Science) : Pamantasan ng Lungsaod ng Maynila, 2024.
5
ABSTRACT: This study addresses fundamental challenges in sentiment analysis by enhancing the effectiveness of Fuzzy Support Vector Machines (FSVM). We propose novel methodologies to improve FSVM's interpretability, generality, and accuracy in sentiment classification. By integrating Bayesian rules, employing feature selection techniques, and optimizing fuzzy membership functions within a fuzzy logic frameqork, our approach aims to overcome the limitations of traditional FSVM. Through thorough testing and evaluation, our enhanced FSVM algorithm demonstrates superior adherence to predefined parameters, prioritizing precision and reliability in sentiment analysis tasks compares to existing methods. Specifically, we focus on refining feature selection to enhance model adaptability, incorporating fuzzy membership functions to optimize performance, and introducing a probabilistics framework for more accurate classification. Our findings showcase the effectiveness of these enhancements in delivering dependable sentimnent analysis results, outperforming existing approaches. While acknowledging potential influences of data quality and dataset characteristics, our study contributes to advancing text analysis techniques, particularly in sentiment analysis, with broader applications across various sectors.
5
2 = =
2
2 --0------
6 --0-- 2 --------
0 2 --
--20------
--------20--
--------20--
----2
/ 2
/ 2
/
/