Artificial intelligence and one of its promising areas, machine learning, have been widely used by the research community to turn massive, diverse, and even heterogeneous health care data sources into high quality facts and knowledge, providing leading capabilities to robust pattern discovery. However, the application of machine learning strategies on big and complex medical datasets is computationally expensive, and it consumes a very large amount of logical and physical resources, such as data storage, CPU, and memory. Additionally, and from the implementation perspective, most big data machine learning algorithms are complex, and their implementations are available for few environments. These operational restrictions cause various difficulties for utilization of big data analytics, and even more, they create challenges to establish novel experiments and develop new research ideas.
Sophisticated big data analytics-as-a-Service platforms for efficient data analyses is becoming more valuable as the amount of data generated daily in the health care literature exceeds the boundaries of normal processing capabilities. The objective of the bigdas@KDD2017 is to provide a professional forum for data scientists, researchers, and engineers across the world to present their latest research findings, innovations, and developments in turning big data health care analytics into fast, easy-to-use, scalable, and highly available services over the Internet. This workshop is aimed at data science practitioners working at the intersection of big data machine learning, Software as a Service (SaaS) platforms, Internet of Things (IoT), and health informatics. It will highlight current trends and insights for the future of health data analytics, which is bigger and smarter.
The first workshop on Big data analytics-as-a-Service: Architecture, Algorithms, and Applications in Health Informatics is taking place on August 14, 2017 (in conjunction with KDD 2017) in Halifax, Nova Scotia, Canada. The workshop will consist of a combination of invited keynote speakers, panel discussion, and paper/poster presentations. We allocate significant time for open discussions on sharing best practices and future directions.
Suggested topics include (but are not limited to) the following with the focus of health informatics application area:
Big data machine learning algorithms
Big data semi-supervised learning, active learning, inductive inference, organizational learning, evolutional learning, transfer learning, manifold learning, probabilistic and relational learning
Big data deep learning
Big data decision support systems
Big data scientific visualization
Big temporal data mining
Big data time series and sequential pattern mining
Big data clinical/biomedical text analytics
Automatic semantic annotation of medical content
Large-scale classification, clustering, and interpretation of biomedical images and videos
Genetic data analytics, mining big gene databases and biological databases
Gold Standards
Feature engineering considerations and selection
Algorithm considerations and selection
Analysis selection criteria
Systems Architecture
Infrastructures for big data analytics
Scalable and high throughput systems for large-scale data analytics
Performance evaluation or comparative study of big data analytics tools, such as DataMelt, RapidMiner, Orange, Rattle, Apache Spark MLlib, Apache Mahout, etc.
Performance evaluation or comparative study of Machine Learning as a Service platforms, such as BigML, Microsoft Azure, Amazon Machine Learning, Google Cloud Prediction API, IBM Watson Analytics, etc.
Integration PaaS (iPaaS) supporting Big Data applications and services
Application of cloud computing to big data analytics
Big data analytics-as-a-Service
Big data machine learning-as-a-Service
Turning big data health informatics into WWW services
Big data deep learning-as-a-Service
Big data infrastructure-as-a-Service
08月14日
2017
会议日期
注册截止日期
留言