XTXiaoying Tan

Xiaoying Tan

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Xiaoying Tan has a total of 11 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Tags and titles of videos you watched tell your gender

    by Tingting Feng, Yuchun Guo, Yishuai Chen, Xiaoying Tan, Ting Xu, Baijun Shen, Wei Zhu

    In online video systems, viewer demographic information (gender, age, etc.) is of huge commercial value for delivering targeted advertising and video recommendations, but generally not available directly. This paper targets inferring viewers' gender based on implicit watching history in the large-scale online video systems. To tackle the sparsity problem without filtering out any cold users or videos, we not only introduce video tags as features, but also use an efficient Chinese word segmentation method to extract hot key-words from video titles as features. Moreover, users' viewing behavior distribute lognormally, hence we apply a logarithmic transformation on the inference matrixes and further find key features via principal components analysis (PCA). We then solve the gender inference as a classification problem and define some modified evaluation metrics adapt to the imbalance classification problem. We compare a set of classifiers including Class prior, EM, SVM, Logistic regression, Partially supervised soft-label and belief-based mixture and find that Logistic regression is the best. The inference results show that our algorithms can obtain high F̃ 1 values for all classes. The highest value of PPTV dataset can reach nearly 0.75. And inference based on key-words results in a 14.63% increase of F̃ 1 contrast to the ratings of MovieLens.


  • Is access frequency adequate for video popularity ranking in P2P streaming systems?

    by Xiaoying Tan, Yuchun Guo, Yishuai Chen, Qiufang Ying, Wei Zhu, Changjia Chen

    Video popularity ranking is commonly used in various streaming systems. Existing popularity ranking systems usually rank videos from the perspective of access frequency and ignore other implicit feedback, due to the belief that popularity rankings with different implicit metrics are highly consistent. Based on a large collection of real-world system trace data, this paper systematically compares video popularity rankings with four different implicit metrics: access frequency, access user frequency, watching time, normalized watching time. We not only analyze the temporal dynamics of videos' popularities ranking with different metrics, but also examine the correlation between them. We have two main findings. 1) The popularities rankings with access frequency and watching time have strong correlation, but the other pairs have low consistency. 2) The temporal dynamics of daily rankings are similar, while the popularity rankings by access frequency and user number frequency are more stable than others. This paper also proposes a new top-N rank correlation coefficient and proves its efficiency. Our analysis results indicate that the implicit metric of popularity ranking needs to be carefully chosen according to specific applications.