A Comparison Study on Sentiment Analysis and Emotion Detection in Marketing Analytics Based on AI Approach
Publish Year: 1402
نوع سند: مقاله کنفرانسی
زبان: English
View: 72
This Paper With 7 Page And PDF Format Ready To Download
- Certificate
- من نویسنده این مقاله هستم
استخراج به نرم افزارهای پژوهشی:
شناسه ملی سند علمی:
CRIAL01_101
تاریخ نمایه سازی: 7 مرداد 1403
Abstract:
Sentiment analysis is a method to determine the sentiment or opinion expressed in text data, enabling companies to understand customer feedback, brand perception, and make data-driven decisions. There exist various approaches to handling emotions, one of which involves the utilization of deep learning algorithms and machine learning techniques.This study explores sentiment analysis using machine learning and deep learning algorithms.We compare different algorithms, especially SVM, which is a powerful method that can deal with complex and high-dimensional data, with other methods such as naive Bayes, decision tree, and random forest .According to the paper, all of the techniques utilized in the study have proven to yield impressive accuracy and F۱ scores exceeding ۷۵%. Combining PV-DBOW or PV-DM with SVM or Logistic Regression has been found to yield the best outcomes, achieving an accuracy of approximately ۸۷% and an F۱ score of ۸۱%. The paper also highlights that PV-DBOW, in conjunction with Logistic Regression, classifies certain data differently, possibly due to an imbalance in the data that favors negative sentiment. To enhance performance and outcomes, the paper proposes several future research directions
Keywords:
Authors
Mobina Hajimohammadi
BS.c Student, Department of Computer Engineering, Karaj Branch, Islamic Azad University, Karaj, Iran
Maryam Jamalpour
MS.c Student, Department of Computer Engineering, Karaj Branch, Islamic Azad University, Karaj, Iran
Behnaz Nahvi
Assistant Professor, Department of Computer Engineering, Karaj Branch, Islamic Azad University, Karaj, Iran