Interactive Information Need Prediction with Intent and Context

2025-01-05

AI Summary

Accurately modeling users’ information needs prior to search remains challenging due to the lexical gap between user intent and document content, as well as context overload in long documents. Method: We propose an explicit-intent–implicit-context co-driven information need prediction framework. Users interactively select multi-granular contextual units (e.g., paragraphs, sentences, or terms) and provide lightweight, structured intent prompts (e.g., “how”, “applications”), enabling the system to either generate natural-language questions or retrieve answers directly. The method jointly leverages generative language models (for question generation) and retrieval models (for answer extraction), supporting large-context inputs and low-overhead interaction. Contribution/Results: Experiments demonstrate significant improvements in prediction accuracy and robustness under long-context settings; minimal intent prompts effectively bridge the lexical gap without requiring full queries; and the framework exhibits strong practicality and deployment feasibility.

Technology Category

• Natural Language Processing: Question Answering
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: ML for personalized search and recommendations
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

The ability to predict a user's information need would have wide-ranging implications, from saving time and effort to mitigating vocabulary gaps. We study how to interactively predict a user's information need by letting them select a pre-search context (e.g., a paragraph, sentence, or singe word) and specify an optional partial search intent (e.g.,"how","why","applications", etc.). We examine how various generative language models can explicitly make this prediction by generating a question as well as how retrieval models can implicitly make this prediction by retrieving an answer. We find that this prediction process is possible in many cases and that user-provided partial search intent can help mitigate large pre-search contexts. We conclude that this framework is promising and suitable for real-world applications.

Problem

• Information Need Prediction
• Information Retrieval Efficiency
• Vocabulary Gap Mitigation

Innovation

• Interactive Prediction
• Generative Language Models
• Retrieval Models