Scholarlink

The proliferation of online reviews has made them a crucial source of customer opinion for products and services. However, the presence of spam reviews—fake reviews aimed at promoting or demoting certain products to generate revenue—has become a significant issue. This practice, known as review spamming, undermines the reliability of online reviews. A variety of methods have been proposed to address spam reviews. This report provides a comprehensive review of current studies on spam analysis and identification, focusing on the following aspects: Datasets and features: The methods used to detect spam reviews depend heavily on the datasets and the features extracted from them. Feature extraction techniques are crucial in identifying spam reviews effectively. Detection methods: Various methods and techniques have been proposed to solve the problem of spam detection. The choice of method significantly impacts the reliability of the detection process. Evaluation metrics: Different metrics are used to assess the effectiveness of spam detection methods. Understanding these metrics is a key to evaluating and comparing the performance of different approaches. Performance parameters: This study identifies and describes several key performance metrics that are frequently utilized to assess the effectiveness and reliability of spam detection models. Taxonomy of approaches: The report proposes a taxonomy of spam screening approaches, offering a structured way to categorize and compare different methods. Public datasets: The availability of public screening datasets is discussed, providing resources for further research and development in spam detection. Research gaps and future directions: Finally, the report identifies gaps in current research and suggests potential directions for future studies to improve the identification and handling of spam reviews. By examining these factors, this report aims to provide a comprehensive understanding of the current state of spam review detection and offer insights for future advancements in this field.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Survey on Online Spam Detection Using Different Machine Learning Techniques

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • P. Swetha,
  • D. S. Rao

摘要

The proliferation of online reviews has made them a crucial source of customer opinion for products and services. However, the presence of spam reviews—fake reviews aimed at promoting or demoting certain products to generate revenue—has become a significant issue. This practice, known as review spamming, undermines the reliability of online reviews. A variety of methods have been proposed to address spam reviews. This report provides a comprehensive review of current studies on spam analysis and identification, focusing on the following aspects: Datasets and features: The methods used to detect spam reviews depend heavily on the datasets and the features extracted from them. Feature extraction techniques are crucial in identifying spam reviews effectively. Detection methods: Various methods and techniques have been proposed to solve the problem of spam detection. The choice of method significantly impacts the reliability of the detection process. Evaluation metrics: Different metrics are used to assess the effectiveness of spam detection methods. Understanding these metrics is a key to evaluating and comparing the performance of different approaches. Performance parameters: This study identifies and describes several key performance metrics that are frequently utilized to assess the effectiveness and reliability of spam detection models. Taxonomy of approaches: The report proposes a taxonomy of spam screening approaches, offering a structured way to categorize and compare different methods. Public datasets: The availability of public screening datasets is discussed, providing resources for further research and development in spam detection. Research gaps and future directions: Finally, the report identifies gaps in current research and suggests potential directions for future studies to improve the identification and handling of spam reviews. By examining these factors, this report aims to provide a comprehensive understanding of the current state of spam review detection and offer insights for future advancements in this field.