Interactive Query Answering on Knowledge Graphs with Soft Entity Constraints

2025-08-19

AI Summary

Existing knowledge graph (KG) query methods predominantly rely on first-order logic, limiting their ability to model real-world soft constraints—such as attribute preferences or category biases—that are inherently fuzzy and context-dependent. To address answer unreachability caused by missing KG edges, this paper formally introduces the novel task of *interactive query answering with soft entity constraints*. We propose the Neural Query Re-ranking (NQR) framework, which jointly leverages embedding-based representation learning and an incremental interactive feedback mechanism. Given user-provided positive and negative preference examples, NQR dynamically refines answer ranking while preserving the original query’s expressiveness and robustness. Crucially, it seamlessly integrates ambiguous semantic constraints into the retrieval process. Extensive experiments on an extended benchmark dataset demonstrate that NQR significantly improves answer relevance and accurately captures soft constraint semantics.

Technology Category

• Data Mining & Knowledge Management: Intelligent Query Processing
• Natural Language Processing: Question Answering
• Knowledge Representation and Reasoning: Preferences

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Querying, indexing, and retrieval in Web-related graphs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Methods for query answering over incomplete knowledge graphs retrieve entities that are likely to be answers, which is particularly useful when such answers cannot be reached by direct graph traversal due to missing edges. However, existing approaches have focused on queries formalized using first-order-logic. In practice, many real-world queries involve constraints that are inherently vague or context-dependent, such as preferences for attributes or related categories. Addressing this gap, we introduce the problem of query answering with soft constraints. We propose a Neural Query Reranker (NQR) designed to adjust query answer scores by incorporating soft constraints without disrupting the original answers to a query. NQR operates interactively, refining answers based on incremental examples of preferred and non-preferred entities. We extend existing QA benchmarks by generating datasets with soft constraints. Our experiments demonstrate that NQR can capture soft constraints while maintaining robust query answering performance.

Problem

• Answering queries with vague soft constraints on knowledge graphs
• Incorporating context-dependent preferences without disrupting original answers
• Handling incomplete knowledge graphs through interactive query refinement

Innovation

• Neural Query Reranker adjusts scores with soft constraints
• Interactive refinement using incremental preference examples
• Extends benchmarks with generated soft constraint datasets