Scribd
Upload
2 views

ABSTRACT

This study presents a fusion sentiment analysis method that combines textual analysis and machine learning to analyze online product reviews. The method involves sentiment feature extraction, polarity identification using SVM, and topic extraction with LDA, introducing a weighting method for sentiment contributions. Tested on Amazon book reviews, it provides insights for product and marketing improvements, demonstrating the effectiveness of machine learning in analyzing complex datasets.

Uploaded by

Bonagiri Roshan
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
2 views

ABSTRACT

This study presents a fusion sentiment analysis method that combines textual analysis and machine learning to analyze online product reviews. The method involves sentiment feature extraction, polarity identification using SVM, and topic extraction with LDA, introducing a weighting method for sentiment contributions. Tested on Amazon book reviews, it provides insights for product and marketing improvements, demonstrating the effectiveness of machine learning in analyzing complex datasets.

Uploaded by

Bonagiri Roshan
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 1

ABSTRACT

With the growth of e-commerce, online reviews provide valuable insights into
customer perceptions. This study proposes a fusion sentiment analysis method
combining textual analysis and machine learning to extract insights from
product reviews. The method includes three steps:

1. Sentiment Feature Extraction: Using a sentiment dictionary, extended


with semantic similarity to capture emotional information.
2. Sentiment Polarity Identification: Classifying reviews into positive or
negative sentiments using the SVM algorithm.
3. Sentiment Topic Extraction: Identifying topics related to customer
experiences using the LDA model.

A key innovation is introducing a weighting method to account for varying


sentiment contributions of words. Tested on Amazon book reviews, the method
accurately classifies sentiments and uncovers factors influencing reading
experiences. The results provide actionable insights for product and marketing
improvements. This research highlights the potential of machine learning in
analyzing complex datasets and driving data-driven decisions to optimize
business strategies.

Keywords: E-commerce, online reviews, sentiment analysis, machine learning,


sentiment dictionary, SVM algorithm, LDA model, semantic similarity,
weighting method, consumer insights, product improvement, marketing
strategy, Amazon reviews, emotional tendencies, data-driven decisions.

You might also like