Title: Keywords are not always the key: A metadata field analysis for natural language search on open data portals Show Chinese title

Title: 关键词并不总是关键：对开放数据门户自然语言搜索的元数据字段分析

Abstract: Open data portals are essential for providing public access to open datasets. However, their search interfaces typically rely on keyword-based mechanisms and a narrow set of metadata fields. This design makes it difficult for users to find datasets using natural language queries. The problem is worsened by metadata that is often incomplete or inconsistent, especially when users lack familiarity with domain-specific terminology. In this paper, we examine how individual metadata fields affect the success of conversational dataset retrieval and whether LLMs can help bridge the gap between natural queries and structured metadata. We conduct a controlled ablation study using simulated natural language queries over real-world datasets to evaluate retrieval performance under various metadata configurations. We also compare existing content of the metadata field 'description' with LLM-generated content, exploring how different prompting strategies influence quality and impact on search outcomes. Our findings suggest that dataset descriptions play a central role in aligning with user intent, and that LLM-generated descriptions can support effective retrieval. These results highlight both the limitations of current metadata practices and the potential of generative models to improve dataset discoverability in open data portals. Show Chinese abstract

Abstract: 开放数据门户对于提供公众访问开放数据集至关重要。 然而，它们的搜索界面通常依赖于基于关键词的机制和有限的元数据字段。 这种设计使得用户难以通过自然语言查询找到数据集。 当用户不熟悉特定领域的术语时，元数据常常不完整或不一致，这使问题更加严重。 在本文中，我们研究了各个元数据字段如何影响对话式数据集检索的成功，并探讨大型语言模型（LLM）是否能弥合自然查询与结构化元数据之间的差距。 我们使用真实世界数据集上的模拟自然语言查询进行受控消融实验，以评估不同元数据配置下的检索性能。 我们还将元数据字段“描述”的现有内容与LLM生成的内容进行比较，探讨不同的提示策略如何影响质量和对搜索结果的影响。 我们的研究结果表明，数据集描述在与用户意图对齐方面起着核心作用，而LLM生成的描述可以支持有效的检索。 这些结果突显了当前元数据实践的局限性，以及生成模型在提高开放数据门户中数据集可发现性的潜力。