Experiences Developing a User-centric Presentation of Provenance for a Web-based Science Data Analysis Tool

Printer-friendly version

Presented at the EGU General Assembly 2011

Abstract:

Web-based science analysis and processing tools allow users to access, analyze, and generate visualizations of data while alleviating users from having to directly manage complex data processing operations. These tools provide value by streamlining the data analysis process, but usually shield users from details of the data processing steps, algorithm assumptions, caveats, etc. Correct interpretation of the final analysis requires user understanding of how data has been generated and processed and what potential biases, anomalies, or errors may have been introduced. By providing services that leverage data lineage provenance and domain-expertise, expert systems can be built to aid the user in understanding data sources, processing, and the suitability for use of products generated by the tools.

We describe our experiences developing a semantic, provenance-aware, expert-knowledge advisory system applied to an existing web-based Earth science data analysis tool (e.g. Giovanni from NASA/GSFC). First, we introduce our use cases for provenance and science metadata capture, reasoning, and visualization to the end user. Then we introduce our integrated semantic data model, which is comprised of provenance, data processing, and science domain ontologies. We describe how we developed an initial set of expert rules, to reason over our data model and recognize conditions in the processing provenance that could lead to anomalies or errors in the processing results. Finally we will show how knowledge from the semantic data model and inferences of the advisory expert ruleset is presented to the user to assist in user understanding of the suitability of products generated by the analysis tool.

Related Projects:

MDSA LogoMulti-Sensor Data Synergy Advisor (MDSA)
Principal Investigator: Peter Fox
Description: Augment Giovanni, the Goddard online tool for data access, visualization and analysis, with semantic web technologies and ontologies to support data inter-comparisons from different sensors or models. Data provenance (i.e. the essential data parameter details, quality and production caveats) will be added to enable researchers to make valid data comparisons and draw quantitative conclusions on specific analysis (e.g. ocean fertilization due to acid rain). In the resulting Giovanni framework, the dataset variable characteristics and related quality can be encoded so that inter-comparison rules can be derived.

Related Research Areas:

Data Science
Lead Professor: Peter Fox
Description: Science has fully entered a new mode of operation. Data science is advancing inductive conduct of science driven by the greater volumes, complexity and heterogeneity of data being made available over the Internet. Data science combines of aspects of data management, library science, computer science, and physical science using supporting cyberinfrastructure and information technology. As such it is changing the way all of these disciplines do both their individual and collaborative work.

Data science is helping scienists face new global problems of a magnitude, complexity and interdisciplinary nature whose progress is presently limited by lack of available tools and a fully trained and agile workforce.

At present, there is a lack formal training in the key cognitive and skill areas that would enable graduates to become key participants in escience collaborations. The need is to teach key methodologies in application areas based on real research experience and build a skill-set.

At the heart of this new way of doing science, especially experimental and observational science but also increasingly computational science, is the generation of data.

Concepts: eScience
Knowledge Provenance
Lead Professor: Deborah L. McGuinness
Description: Knowledge Provenance
Concepts: Provenance,
Semantic eScience
Lead Professor: Peter Fox
Description:
Science has fully entered a new mode of operation. E-science, defined as a combination of science, informatics, computer science, cyberinfrastructure and information technology is changing the way all of these disciplines do both their individual and collaborative work.
As semantic technologies have been gaining momentum in various e-Science areas (for example, W3C's new interest group for semantic web health care and life science), it is important to offer semantic-based methodologies, tools, middleware to facilitate scientific knowledge modeling, logical-based hypothesis checking, semantic data integration and application composition, integrated knowledge discovery and data analyzing for different e-Science applications.
Partially influenced by the Artificial Intelligence community, the Semantic Web researchers have largely focused on formal aspects of semantic representation languages or general-purpose semantic application development, with inadequate consideration of requirements from specific science areas. On the other hand, general science researchers are growing ever more dependent on the web, but they have no coherent agenda for exploring the emerging trends on the semantic web technologies. It urgently requires the development of a multi-disciplinary field to foster the growth and development of e-Science applications based on the semantic technologies and related knowledge-based approaches.

Concepts: eScience
X-informatics
Lead Professor: Peter Fox
Description: In the last 2-3 years, Informatics has attained greater visibility across a broad range of disciplines, especially in light of great successes in bio- and biomedical-informatics and significant challenges in the explosion of data and information resources. Xinformatics is intended to provide both the common informatics knowledge as well as how it is implemented in specific disciplines, e.g. X=astro, geo, chem, etc. Informatics' theoretical basis arises from information science, cognitive science, social science, library science as well as computer science. As such, it aggregates these studies and adds both the practice of information processing, and the engineering of information systems.
Concepts: , eScience