Context-Aware Users' Preference Models by Integrating Real and Supposed Situation Data

Chihiro Ono, Yasuhiro Takishima, Yoichi Motomura, Hideki Asoh, Yasuhide Shinagawa, Michita Imai, Yuichiro AnzaiIEICE Trans. Information and Systems 2008 · DOI · OpenReview

Introduction

Context-aware recommender systems aim to provide personalized suggestions based on a user's immediate environment and state. For instance, a system might suggest a light salad on a hot afternoon or a warm stew on a cold winter night. However, building these models is notoriously difficult due to the challenge of obtaining high-quality data for every possible context.

The Data Acquisition Bottleneck

To train effective preference models, researchers typically require large datasets. However, collecting "real situation data"—where users are actually in the specific state being measured (e.g., genuinely exhausted or hungry)—is logistically complex and expensive.

The challenge of acquiring large-scale real situation data compared to supposed data.

As a result, many systems rely on "supposed situation data," where subjects imagine their preferences in a given scenario. While easier to collect at scale, this imagined data often fails to capture the nuances of real-world behavior. This paper addresses how to integrate a small amount of real data with a larger volume of supposed data to create models that are both robust and accurate.

Proposed Integration Methods

The authors explore three strategies for data integration, focusing on two methods that allow for the "fast model updates" essential for mobile applications:

  1. Model Modification (Method 1): This method adjusts the internal logic of a Bayesian network. The Conditional Probability Tables (CPTs)—the mathematical rules of the model—are calculated as a weighted sum of real and supposed data. A weight parameter ($\alpha$) allows the model to prioritize real-world evidence while using supposed data to fill in the gaps.
  2. Inference Result Modification (Method 2): This approach maintains two separate models—one for real data and one for supposed data. The final recommendation is a weighted average of the predictions from both models.
Comparison of Method 1 (Model Modification) and Method 2 (Inference Result Modification).

While a baseline approach of merging all data and retraining is possible, it is computationally expensive. In tests, the proposed methods updated in just 2 seconds, compared to over 25 minutes for the baseline.

Experimental Results

The methods were evaluated using food preference surveys. Dataset 1 included 23,760 records of supposed data from 1,500 subjects. Dataset 2 included 480 real records and 240 supposed records from a smaller group of 12 subjects.

Key findings include:

  • Accuracy Boost: Integrated models significantly outperformed those trained on only one data type, achieving lower Mean Squared Error (MSE).
  • Optimal Weighting: For Method 1, a weight of $\alpha = 0.2$ (20% real data) provided the best results, demonstrating that even a small amount of real-world evidence can significantly correct the biases of imagined data.
  • Method 1 vs. Method 2: Modifying the model's internal CPTs generally yielded better performance than simply averaging inference results, as it better captures the complex dependencies within the preference network.

Conclusion

By integrating a small "dose" of real-world veracity with a large volume of supposed data, developers can overcome the data acquisition bottleneck in context-aware systems. This approach maintains high accuracy while ensuring the computational efficiency required for real-time mobile applications. Future work will focus on automatically estimating optimal weighting parameters and testing these methods in domains beyond food recommendations.

Summary updated

Abstract

This paper proposes a novel approach of constructing statistical preference models for context-aware personalized applications such as recommender systems. In constructing context-aware statistical preference models, one of the most important but difficult problems is acquiring a large amount of training data in various contexts/situations. In particular, some situations require a heavy workload to set them up or to collect subjects capable of answering the inquiries under those situations. Because of this difficulty, it is usually done to simply collect a small amount of data in a real situation, or to collect a large amount of data in a supposed situation, i.e., a situation that the subject pretends that he is in the specific situation to answer inquiries. However, both approaches have problems. As for the former approach, the performance of the constructed preference model is likely to be poor because the amount of data is small. For the latter approach, the data acquired in the supposed situation may differ from that acquired in the real situation. Nevertheless, the difference has not been taken seriously in existing researches. In this paper we propose methods of obtaining a better preference model by integrating a small amount of real situation data with a large amount of supposed situation data. The methods are evaluated using data regarding food preferences. The experimental results show that the precision of the preference model can be improved significantly.
BibTeX
@article{ono2008context,
  author = {Chihiro Ono and Yasuhiro Takishima and Yoichi Motomura and Hideki Asoh and Yasuhide Shinagawa and Michita Imai and Yuichiro Anzai},
  title = {{Context-Aware Users' Preference Models by Integrating Real and Supposed Situation Data}},
  journal = {IEICE Trans. Information and Systems},
  year = {2008},
  doi = {10.1093/ietisy/e91-d.11.2552},
}
Figure