CRAFT: Time Series Forecasting with Cross-Future Behavior Awareness

2025-05-20

AI Summary

To address high uncertainty in time-series forecasting caused by limited historical information, this paper introduces the novel concept of “Cross-Future Behavior” (CFB)—dynamic signals that occurred prior to the current timestamp yet implicitly influence future trajectories. We propose an end-to-end CFB modeling framework: a Koopman-operator-based prediction module, a dual-trend mining mechanism (incorporating internal local trend modeling and hierarchical external trend guidance), and a demand-constrained loss function to calibrate predictive distributions. Evaluated on large-scale offline benchmarks and real-world online A/B tests, our model consistently outperforms state-of-the-art baselines. The source code and datasets are publicly released.

Technology Category

• Machine Learning: Time-Series/Data Streams
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Planning, Routing, and Scheduling: Model-Based Reasoning

Application Category

• Web Mining and Content Analysis: Large pretrained models with web data
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• User Modeling, Personalization and Recommendation: User modeling for targeted and personalized online advertising

Abstract

The past decades witness the significant advancements in time series forecasting (TSF) across various real-world domains, including e-commerce and disease spread prediction. However, TSF is usually constrained by the uncertainty dilemma of predicting future data with limited past observations. To settle this question, we explore the use of Cross-Future Behavior (CFB) in TSF, which occurs before the current time but takes effect in the future. We leverage CFB features and propose the CRoss-Future Behavior Awareness based Time Series Forecasting method (CRAFT). The core idea of CRAFT is to utilize the trend of cross-future behavior to mine the trend of time series data to be predicted. Specifically, to settle the sparse and partial flaws of cross-future behavior, CRAFT employs the Koopman Predictor Module to extract the key trend and the Internal Trend Mining Module to supplement the unknown area of the cross-future behavior matrix. Then, we introduce the External Trend Guide Module with a hierarchical structure to acquire more representative trends from higher levels. Finally, we apply the demand-constrained loss to calibrate the distribution deviation of prediction results. We conduct experiments on real-world dataset. Experiments on both offline large-scale dataset and online A/B test demonstrate the effectiveness of CRAFT. Our dataset and code is available at https://github.com/CRAFTinTSF/CRAFT.

Problem

• Address uncertainty in time series forecasting with limited past data
• Utilize cross-future behavior to predict time series trends
• Overcome sparse and partial flaws in cross-future behavior analysis

Innovation

• Uses Cross-Future Behavior (CFB) for trend mining
• Employs Koopman Predictor for key trend extraction
• Applies demand-constrained loss for distribution calibration