Prediction reliability of deep learning based systems allows users to confirm if the prediction is reliable to real applications, which is the key to the success of recommendation systems. Current research on deep learning based recommendation systems focused on estimating model reliability, but seemed to be missing data contributions to the prediction reliability. This paper proposed a novel framework to estimate the prediction reliability for deep learning-based methods through combining the model reliability and the data reliability. The proposed framework has been validated in a case study based on named entity recognition (NER) that is from an Intuit recommendation task. It employed two NER datasets: WNUT and GMB to examine detailed performance of the proposed framework, where, specifically, we proposed a novel evaluation metric to comprehensively evaluate the performance. Experimental results demonstrated that, compared to model reliability only, combining data reliability with model reliability will significantly improve performance, as well as enhance the prediction interpretability.

In recent years, Online Social Networks (OSN) have become popular content-sharing environments. With the emergence of smartphones with high-quality cameras, people like to share photos of their life moments on OSNs. The photos, however, often contain private information that people do not intend to share with others (e.g., their sensitive relationship). Solely relying on OSN users to manually process photos to protect their relationship can be tedious and error-prone. Therefore, we designed a system to automatically discover sensitive relations in a photo to be shared online and preserve the relations by face blocking techniques. We first used the Decision Tree model to learn sensitive relations from the photos labeled private or public by OSN users. Then we defined a face blocking problem and developed a linear programming model to optimize the tradeoff between preserving relationship privacy and maintaining the photo utility. In this paper, we generated synthetic data and used it to evaluate our system performance in terms of privacy protection and photo utility loss.