Scholarlink

<p>In the twenty-first century, chatbots have become one of the most widely used technologies in our lives. They have rapidly evolved from simple rule-based systems to advanced AI-powered tools, now commonly applied in customer service, healthcare, education, and online shopping. This review paper looks closely at the different types of algorithms used to build chatbots. These include rule-based, retrieval-based, generative, transformer-based, and hybrid models. This paper focuses on how these algorithms help chatbots understand user intent, manage conversations, and generate replies. The paper also explores how natural language processing (NLP)and machine learning methods from older models like Naive Bayes and SVM to more advanced ones like LSTM, GRU, and transformer models improve chatbot performance. Various evaluation methods are discussed, including automatic metrics such as BLEU, ROUGE, METEOR, and Perplexity, as well as human-based evaluations assessing fluency, relevance, and engagement. This review further highlights key challenges, such as maintaining conversational context, addressing ethical concerns, supporting multiple languages, and protecting user privacy. By providing a comprehensive summary of chatbot algorithms and previous research, this study aims to support further advancements in the field.</p>

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

Reviewing chatbot algorithms: methods for intelligent dialogue systems

  • 期刊论文

  • 发表日期：

    2025年9月8日

查看全文
  • Md Tousif Hasan Lavlu,
  • Afnanul Hassan,
  • Sanjida Akhtar,
  • Salsa- Bil- Labiba,
  • Hossen Asiful Mustafa

摘要

In the twenty-first century, chatbots have become one of the most widely used technologies in our lives. They have rapidly evolved from simple rule-based systems to advanced AI-powered tools, now commonly applied in customer service, healthcare, education, and online shopping. This review paper looks closely at the different types of algorithms used to build chatbots. These include rule-based, retrieval-based, generative, transformer-based, and hybrid models. This paper focuses on how these algorithms help chatbots understand user intent, manage conversations, and generate replies. The paper also explores how natural language processing (NLP)and machine learning methods from older models like Naive Bayes and SVM to more advanced ones like LSTM, GRU, and transformer models improve chatbot performance. Various evaluation methods are discussed, including automatic metrics such as BLEU, ROUGE, METEOR, and Perplexity, as well as human-based evaluations assessing fluency, relevance, and engagement. This review further highlights key challenges, such as maintaining conversational context, addressing ethical concerns, supporting multiple languages, and protecting user privacy. By providing a comprehensive summary of chatbot algorithms and previous research, this study aims to support further advancements in the field.