Understanding User Intent Modeling for Conversational Recommender Systems: A Systematic Literature Review

This dataset compiles the results of a systematic literature review on user intent modeling in Natural Language Processing (NLP), with a focus on its application in conversational recommender systems. Over 13,000 papers from the past decade have been analyzed to provide a thorough understanding of the prevalent AI models used in this area. The dataset includes detailed examinations of various machine learning models such as SVM, LDA, Naive Bayes, BERT, Word2vec, and MLP, highlighting their advantages, limitations, and suitability for different scenarios in recommender systems. Additionally, the dataset encompasses a wide range of applications of user intent modeling across sectors such as e-commerce, healthcare, education, social media, and virtual assistants. It sheds light on how these models aid in delivering personalized recommendations, detecting fake reviews, providing health interventions, tailoring educational content, and enhancing user experience on social media. A key component of the dataset is a decision model, derived from the literature review, designed to assist researchers and developers in selecting the most appropriate machine learning model for specific user intent modeling tasks in recommender systems. This model addresses the challenge posed by the variety of available models and the lack of a clear classification scheme. Furthermore, the dataset includes the outcomes of two academic case studies conducted to assess the utility of the decision model. These case studies follow Yin's guidelines and provide practical insights into the application of the decision model in real-world scenarios. Researchers, developers, and practitioners in the field of NLP, AI, and recommender systems will find this dataset invaluable for navigating the complex landscape of user intent modeling. It not only synthesizes scattered research but also provides a practical tool for model selection, thereby contributing significantly to the advancement of personalized user experiences in various domains. Keywords: User Intent Modeling, NLP, Conversational Recommender Systems, Machine Learning, Systematic Literature Review, Decision Model

本数据集整合了自然语言处理（Natural Language Processing，NLP）领域中用户意图建模（User Intent Modeling）的系统性文献综述（Systematic Literature Review）成果，重点聚焦其在对话推荐系统（Conversational Recommender Systems）中的应用场景。本数据集依托对近十年间逾13000篇文献的分析，全面梳理了该领域当前流行的人工智能模型。本数据集详细剖析了多种机器学习（Machine Learning）模型，包括支持向量机（Support Vector Machine，SVM）、隐狄利克雷分配（Latent Dirichlet Allocation，LDA）、朴素贝叶斯（Naive Bayes）、BERT、Word2vec以及多层感知机（Multi-Layer Perceptron，MLP），并阐释了各类模型在推荐系统不同场景下的优势、局限与适配性。此外，本数据集还涵盖了用户意图建模在多个行业的广泛应用场景，包括电子商务、医疗健康、教育、社交媒体以及虚拟助手领域。本数据集阐明了上述模型如何助力实现个性化推荐、虚假评论检测、健康干预服务、个性化教育内容定制以及优化社交媒体平台的用户体验。本数据集的核心组成部分为一款基于本次文献综述构建的决策模型（Decision Model），旨在帮助研究人员与开发者为推荐系统中的特定用户意图建模任务遴选最适配的机器学习模型。该模型解决了当前模型种类繁多且缺乏明确分类体系所带来的选型难题。此外，本数据集还收录了两项用于评估该决策模型实用性的学术案例研究成果。这两项案例研究遵循Yin的研究规范，为决策模型在真实场景中的应用提供了切实可行的实践参考。自然语言处理、人工智能以及推荐系统领域的研究人员、开发者与从业者均可借助本数据集，高效厘清用户意图建模领域的复杂研究图景，其价值不言而喻。本数据集不仅整合了分散的研究成果，还提供了一款实用的模型选型工具，从而为各领域个性化用户体验的提升做出了重要贡献。关键词：用户意图建模、自然语言处理、对话推荐系统、机器学习、系统性文献综述、决策模型