Qingxia Liu Basic Info Participants 120 including VIPs

Qingxia Liu

Basic Info • Participants – 120 (including VIPs) – from mainland China (100), Hong Kong, Australia, Singapore, Japan, Qatar, Bangladesh • Review Process – Research Track • 146 submissions, 67 accepted (45%) • 89 reviewers from 13 countries – Demo Track • 7 submissions, 10 accepted (70%) • 8 reviewers from 5 countries – Industry Track • 5 submissions, 3 accepted (60%) • 8 reviewers from China and USA • Contents – – workshop (4), tutorial (3), industry session (1) keynote (3), panel (1), research session (18) distinguished young lecturer series, demo, APWeb Data Challenge

Research Sessions Text analysis : 2, 11 Topic model: 5 Semantics: 7 Semantic Web and Knowledge Base: 10 • location-based service: 15 • graph processing: 16 • • On The Marriage of SPARQL and Keywords Peng (Peking University) Lei Zou* (Peking University)* Dongyan Zhao (Peking University) • • • Social prediction: 1 Learning from Data: 3 Hardware, Schema and Query: 4 Recommendation System: 9, 14 Metric Learning: 6 Query Processing: 8 Mapreduce: 12 social network: 13 Influence maximization: 18

Best Paper • Best Paper: Hybrid-LSH for Spatio-Textual Similarity Queries – Mingdong Zhu * (Northeastern University, China)* Derong Shen (Northeastern university) Ling Liu (Georgia Institute of Technology, USA), Ge Yu (Northeastern university), • Runner-Up: Trustworthy Collaborative Filtering through Downweighing Noise and Redunancy – Qiuxiang Dong, Zhi Guang and Zhong Chen – Peking University

On Coherent Indented Tree Visualization of RDF Graphs • Reviews (4. 33) – strong accept – accept Average Score of Accepted Papers 25 20 15 10 5 0 • Presentation – Session 10: semantic web – Questions • for what application • RDF/XML is already a tree • selection of starting resource 4. 5 • • • 4. 33 4 5 (strong accept) 4 (accept) 3 (neutral) 3. 67 • • 3. 5 3. 33 3 2 (reject) 1 (strong reject)

Tutorials • Indexing for Querying Metric Spaces – Dr. Yunjun Gao, Zhejiang University • Influence computation in spatial databases – Dr. Muhammad Aamir Cheema, Monash University, Australia • Providing Retrievability Guarantees in Cloud. Based Storage Services – Dr. Yin Yang, Hamad Bin Khalifa University (HBKU) • On Repairing Structural Problems In Semistructured Data – Dr. Shanshan Ying, Advanced Digital Sciences Center, Singapore

Tutorials • Influence computation in spatial databases – Dr. Muhammad Aamir Cheema, Monash University, Australia – Rk. NN • Reverse k Nearest Neighbors • 关键用户所在区域 • pruning-verification • On Repairing Structural Problems In Semistructured Data – Dr. Shanshan Ying, Advanced Digital Sciences Center, Singapore – data quality, XML data

Tutorials • Influence computation in Data Quality in Numbers spatial databases $600 billion/year – Dr. Muhammad Aamir 80% of the. Cheema, time Monash 16% growing of data quality tools University, Australia – Rk. NN <tutorial> Reverse k Nearest <title> Neighbors Data Quality Issues on Semi-structured Data <speaker> 关键用户所在区域 <name>Shanshan </name> <name> Ying </name> pruning-verification </speaker> <venue> <name> APWeb </name> <year> 2015 </year> <location> Guangzhou</location> </venue> </tutorial> • • On Repairing Structural Problems In Semistructured Data – Dr. Shanshan Ying, Advanced Digital Sciences Center, Singapore – data quality, XML data • tag missing – substitution-based heuristic • structural anomalies – unexpected elements detection

Keynotes • Big Data – Security and Privacy – Elisa Bertino • Purdue University, USA – Use of data for security • cyber security homeland protection, health care • Personilized Web News Filtering and Summarization – Xindong Wu • University of Vermont, USA; • Hefei Univ. of Technology （合肥 业大学） • Querying Big Data: Bridging Theory and Practice – Wenfei Fan • University of Edinburgh, UK

Keynote 2 • Personilized Web News Filtering and Summarization on the Web – Xindong Wu （吴信东） • University of Vermont, USA; • Hefei Univ. of Technology （合肥 业大学） – Technical interests • deduction -> induction (data generation) • 专家系统 (1988) ->DB知识获取 (1995)->知识发现 – Big Data Characteristics: HACE theorem • Big Data starts with large-volume, Heterogeneous, Autonomous sources with distributed and decentralized control, and seeks to explore Complex and Evolving relationships among data • Wu X, Zhu X, Wu G Q, et al. Data mining with big data[J]. Knowledge and Data Engineering, IEEE Transactions on, 2014, 26(1): 97 -107. – No. 1 most downloaded paper in IEEE XPLORE between Jan. ’ 14 and June’ 15 – No. 2 in July’ 15, No. 3 in Aug’ 15

Keynotes • Querying Big Data: From Theory to Practice – Wenfei Fan （樊文飞） • University of Edinburgh, UK • 北京航空航天大学 – Fundamental issues • computational tractability, bounded evaluability – Effective techniques to compute exact answers • compression, using views, bounded incremental evaluation – Approximate QA • query-driven approximation • data driven approximation

Panel: Cradle of Bubble? Where Is Big Data Age Going? • Zhifeng Hao （郝志峰） • Elisa Bertino • – Guangdong University of Technology – 张量方法 • – University of New South Wales, Australia – next generation tech. – Purdue University, USA – 从多媒体自动抽取数据 • • Audaque Data Tech. @深圳 Jeffrey Xu Yu （于旭） • • Wenfei Fan （樊文飞） • tech: unstructured data + model + KG buisness model : data to money policy – The Chinese University of Hong Kong, Hong Kong – How to quickly get answers 数据整合技术 从不同view进行DM – University of Edinburgh, UK – 数据一致性、质量（评价） Arbor Yabo Xu （徐亚波） – Sun Yat-Sen University, China – Yeezhao. com – Industry want： Kang Chen （陈康） – Guangzhou Research Institute, China Telecom Co. – 企业关注的技术 Xuemin Lin （林学民） • sampling auto-approximate Xiaofang Zhou （周晓方） – University of Queensland, Australia – new data management problems: • Redundancy, reduction

Panel by Xueming Lin

Finally • Next Year – 苏州大学 Next Year

Thank You~