Arrow Research search

Author name cluster

Donna Xu

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

2 papers
1 author row

Possible papers

2

IS Journal 2019 Journal Article

A Data-Analytics Approach for Enterprise Resilience

  • Donna Xu
  • Ivor W. Tsang
  • Eng K. Chew
  • Cosimo Siclari
  • Varun Kaul

Enterprise resilience plays an important role to prevent business services from disruptions caused by human-induced disasters such as failed change implementations and software bugs. Traditional expert-centric approach has difficulty to maintain continued critical business functions because the disasters can often only be handled after their occurrence. This paper introduces a data-analytics approach, which leverages system monitoring data for the enterprise resilience. With the power of data mining and machine learning techniques, we build an intelligent business analytics system to detect the potential disruptions proactively, and to assist the operational team for enterprise resilience enhancement. We demonstrate the effectiveness of our approach on a real enterprise system monitoring dataset in simulation.

Details DOI

AAAI Conference 2019 Conference Paper

Label Embedding with Partial Heterogeneous Contexts

  • Yaxin Shi
  • Donna Xu
  • Yuangang Pan
  • Ivor W. Tsang
  • Shirui Pan

Label embedding plays an important role in many real-world applications. To enhance the label relatedness captured by the embeddings, multiple contexts can be adopted. However, these contexts are heterogeneous and often partially observed in practical tasks, imposing significant challenges to capture the overall relatedness among labels. In this paper, we propose a general Partial Heterogeneous Context Label Embedding (PHCLE) framework to address these challenges. Categorizing heterogeneous contexts into two groups, relational context and descriptive context, we design tailor-made matrix factorization formula to effectively exploit the label relatedness in each context. With a shared embedding principle across heterogeneous contexts, the label relatedness is selectively aligned in a shared space. Due to our elegant formulation, PHCLE overcomes the partial context problem and can nicely incorporate more contexts, which both cannot be tackled with existing multi-context label embedding methods. An effective alternative optimization algorithm is further derived to solve the sparse matrix factorization problem. Experimental results demonstrate that the label embeddings obtained with PHCLE achieve superb performance in image classification task and exhibit good interpretability in the downstream label similarity analysis and image understanding task.

PDF Details