NSF/IIS: CAREER: Mining Log Data for Computing System Management

National Science Foundation Award Number: NSF IIS-0546280 (January 1, 2006 to December 31, 2010)

Contact Information

Tao Li,  PI
School of Computer Science
Florida International University
11200 SW 8th Street,

Miami, FL 33199 U.S.A.
Office: (305) 348-6036,   Fax: (305) 348-3549

E-mail: taoli at cs.fiu.edu, URL: http://www.cs.fiu.edu/~taoli

List of Supported Students and Staff

• Dingding Wang: Ph.D. student, School of Computer Science, Florida International University

  • Summer Intern at NEC Research, 2007

• Didier Garcia: Undergraduate student, School of Computer Science, Florida International University

  • Summer Intern at IBM Almaden Research Center, 2006
  • Summer Intern at IBM Extreme Blue, 2007
  
  
• Ruskin Miller: Undergraduate student, School of Computer Science, Florida International University

• Wei Peng: Ph.D. student, School of Computer Science, Florida International University

  • The Best Graduate Student Research Award, School of Computer Science, Florida International University (2006)
  • Excellence Award, School of Computer Science, 2006-07.
  • Summer Intern at Xerox Research, 2006
  • Summer Intern at Xerox Research, 2007
  
  
• Xin Wang: Ph.D. student, School of Computer Science, Florida International University

Project Award Information

• Award Number: NSF IIS 0546280  
• Duration: January, 2006 to December, 2010
• Title: CAREER: Mining Log Data for Computing System Management
• Keywords: Data  Mining, Log Data, System Log, Computing System Management, Autonomic Computing
• Click here to see the Award information on NSF website.

Project Summary

Advancements in science and technology have led to increased complexity in computing systems. With a growing number of heterogeneous software and hardware components, computing systems are becoming more and more difficult to monitor, manage and maintain. The goal of this project is to develop an integrated framework on mining log data for automatic system management. The project is conducting research on: (1) developing new methods and tools for log data organization which create consistency and improve the ability to correlate across multiple log files; (2) developing new methods and tools for data-driven pattern discovery and problem determination; and (3) developing new methods and tools to bridge the gap between the system management applications and the intelligent techniques. The integrated framework resulting from this research will provide much better techniques for monitoring, analyzing, and adapting complex computing systems. Results from this research will be disseminated through publications and the software tools being made publicly available through a web portal. The educational component of this project includes developing a new curriculum that incorporates research into the classroom and provides women, minorities, and undergraduate students with opportunities to participate research. Florida International University (FIU) is among the top awarders of computer science degrees to Hispanic students in the USA and its history of involving the participation of underrepresented groups in the research efforts will be leveraged during the course of this project. This project will also strengthen industry research collaborations and have immediate applications to fields other than system management.

Publications and Products

• Tao Li, Shenghuo Zhu, and Mitsunori Ogihara. Text Categorization via Generalized Discriminant Analysis. Information Processing and Management, to appear as a regular paper, 2008.


• Chris Ding, Tao Li and Wei Peng. On the Equivalence Between Nonnegative Matrix Factorization and Probabilistic Latent Semantic Indexing. Computational Statistics and Data Analysis, to appear as a regular paper, 2008.


• Tao Li, Chang-Shing Perng and Sheng Ma. Guest Editorial: Special Issue on Temporal Data Mining: Theory, Algorithms and Applications. Data Mining and Knowledge Discovery, 16(1): 1-3, 2008.


• Yanfang Ye, Dingding Wang, Tao Li and Dongyi Ye. An Intelligent PE-Malware Detection System Based on Association Mining, to appear in Journal in Computer Virology, 2008.


• Tao Li. Clustering Based on Matrix Approximation. Knowledge and Information Systems Journal, to appear as a regular paper, 2008.


• Li Zhang, Tao Li, Shi Xia Liu, and Yue Pan; An Integrated System for Building Enterprise Taxonomies. Information Retrieval, To appear as a regular paper, 2008.


• Tao Li, Shenghuo Zhu, and Mitsunori Ogihara. Hierarchical Document Classification Using Automatically Generated Hierarchy. Journal of Intelligent Information Systems. To appear as a regular paper, 2007.


• Tao Li. A Unified View On Clustering Binary Data. Machine Learning,62(3): 199-215 (2006)


• Tao Li, Shenghuo Zhu, and Mitsunori Ogihara. Using Discriminant Anlaysis for Multi-class Classification: An Experiment Investigation. Knowledge and Information Systems Journal, 10(4): 453-472 (2006).


•  Wei Peng, Tao Li, and Sheng Ma. Mining Log Files for Data-Driven System Management. SIGKDD Explorations, Volume 7, issue 1, pages 44-51, June 2005.

Refereed Conference Publications

• Shenghuo Zhu, Tao Li, Zhiyuan Chen, Dingding Wang, and Yihong Gong. Dynamic Active Probing of Helpdesk Databases.  In Proceedings of  the 34th International Conference on Very Large Data Bases (VLDB 2008).


• Heng Huang, Chris Ding, Dijun Luo, Tao Li. Simultaneous Tensor Subspace Selection and Clustering: The Equivalence of High Order SVD and K-Means Clustering. In Proceedings of the The 14th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (SIGKDD 2008).


• Tao Li, Chris Ding, Yi Zhang, and Bo Shao. Knowledge Transformation from Word Space to Document Space. In Proceedings of The 31^st Annual International ACM SIGIR Conference (SIGIR 2008).


• Dingding Wang, Tao Li, Shenghuo Zhu,and Chris Ding. Multi-Document Summarization via Sentence-Level Semantic Analysis and Symmetric Matrix Factorization. In Proceedings of The 31^st Annual International ACM SIGIR Conference (SIGIR 2008).


• Wei Peng and Tao Li. Author-Topic Evolution Analysis using Three-way Non-negative Paratucker. In Proceedings of The 31^st Annual International ACM SIGIR Conference (SIGIR 2008). (Poster)


• Fei Wang, Tao Li and Changshui Zhang. Semi-Supervised Clustering via Matrix Factorization. In Proceedings of 2008 SIAM International Conference on Data Mining (SDM 2008).


• Tao Li and Chris Ding. Weighted Consensus Clustering. In Proceedings of 2008 SIAM International Conference on Data Mining (SDM 2008). 


• Tao Li, Chris Ding and Michael I. Jordan. Solving Consensus and Semi-supervised Clustering Problems Using Nonnegative Matrix Factorization. In Proceedings of 2007 IEEE International Conference on Data Mining (ICDM 2007).


• Zhongyuan Zhang, Tao Li, Chris Ding and Xiang-Sun Zhang. Binary Matrix Factorization with Applications. In Proceedings of 2007 IEEE International Conference on Data Mining (ICDM 2007).


• Wei Peng, Chris Ding, Tao Li and Tong Sun. Finding Hotspots in Document Collection. In Proceedings of the 19th IEEE International Conference on Tools with Artificial Intelligence (ICTAI 2007).


• Yanfang Ye, DingDing Wang, Tao Li, and Dongyi Ye. IMDS: Intelligent Malware Detection System. In Proceedings of ACM Int'l Conf. on Knowledge Discovery and Data Mining (SIGKDD 2007), to appear.
• Chris Ding, Rong Jin, Tao Li, and Horst D. Simon.  A Learning Framework using Green's Function and Kernel Regularization with application for recommender system. In Proceedings of ACM Int'l Conf. on Knowledge Discovery and Data Mining (SIGKDD 2007).


• Wei Peng, Charles Perng, Tao Li, and Haixun Wang.  Event Summarization for System Management.  In Proceedings of ACM Int'l Conf. on Knowledge Discovery and Data Mining (SIGKDD 2007).


• Fei Wang, Changshui Zhang, and Tao Li. Regularized Clustering for Documents. In Proceedings of the 30th Annual International ACM SIGIR Conference on Research & Development in Information Retrieval (SIGIR 2007)


• Chris Ding and Tao Li.  Adaptive Dimension Reduction Using Discriminant Analysis and K-means Clustering. In Proceedings of International Conference on Machine Learning (ICML 2007).


• Fei Wang, Changshui Zhang, and Tao Li.. Clustering with Local and Global Regularization.  In Proceedings of the 22nd National Conference on Artificial Intelligence (AAAI 2007).


• Zhiyuan Chen and Tao Li. Addressing Diverse User Preferences in SQL-Query-Result Navigation. In Proceedings of the 2007 ACM SIGMOD Conference (SIGMOD 2007), Pages 641-652.


• Tao Li and Chris Ding. The Relationships among Various Nonnegative Matrix Factorization Methods for Clustering. In Proceedings of the 2006 IEEE International Conference on Data Mining (ICDM 2006), Pages 362-371.
• Alex F. Wang, Sheng Ma, Liuzhong Yang, and Tao Li. Recommendation on Item Graphs. In Proceedings of the 2006 IEEE International Conference on Data Mining (ICDM 2006), Pages 1119-1123.


•  Wei Peng and Tao Li.  Interval Data Clustering with Applications. In Proceedings of the 18th IEEE International Conference on Tools with Artificial Intelligence (ICTAI 2006), Pages 355-362.


• Tao Li, Chengliang Zhang, and Shenghuo Zhu. Empirical Studies on Multilabel Classification. In Proceedings of the 18th IEEE International Conference on Tools with Artificial Intelligence (ICTAI 2006), Pages 86-92.


• Chris Ding, Tao Li, Wei Peng, and Haesun Park. Orthogonal Nonnegative Matrix Tri-factorizations for Clustering. In Proceedings of the Twelfth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (SIGKDD 2006), Pages 126-135.


• Chris Ding, Tao Li and Wei Peng. Nonnegative Matrix Factorization and Probabilistic Latent Semantic Indexing: Equivalence, Chi-square Statistic, and a Hybrid Method. In Proceeding of National Conference on Artificial Intelligence (AAAI-06).


• Weixiang Sun, Tao Li, Wei Peng, and Tong Sun, Incremental Workflow Mining with Optional Patterns, In proceedings of 2006 IEEE International Conference on Systems, Man, and Cybernetics (SMC 2006).


• Tao Li and Wei Peng. A Clustering Model Based on Matrix Approximation with Applications to Cluster System Log Files.  In Proceedings of the 16th European Conference on Machine Learning (ECML 2005), Pages 625-632.


• Tao Li. A General Model for Clustering Binary Data. In Proceedings of the Eleventh ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (SIGKDD 2005), Pages 188-197.


• Tao Li, Feng Liang, Sheng Ma, and Wei Peng. An Integrated Framework on Mining Logs Files for Computing System Management. In Proceedings of the Eleventh ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (SIGKDD 2005), Pages 776-781.