[ad_1]
As a critical component of NASA open source science initiative (OSSI), science discovery engine (SDE) provides new and powerful functionality. Discovery capabilities for all of NASA’s open science data and information. The SDE beta release was announced a little more than a year ago at the 2022 American Geophysical Union Conference.Lens. Since then, the SDE team has continued to improve the interface, delivering more insightful and relevant search results and providing an engaging user experience.
Building an SDE
When NASA announced its strategic focus on open source science in 2018, one of the key recommendations from the scientific community was to develop a unified search portal.Scientists have proposed an interface that gives rise to simultaEasily access all content in NASA’s five science topic areas: Astrophysics, Biophysical Sciences, Earth Sciences, Solar Physics, and Planetary Sciences.
Our team of data scientists, subject matter experts, and developers began developing a plan to achieve this ambitious goal in early 2020. Given the diversity, distribution, and vast size of NASA’s scientific data and information, many challenges have emerged.
For example, each topic area uses its own metadata standards and vocabulary, making it difficult to create comprehensive and accurate metadata across disciplines. Additionally, much of the content related to understanding and using data is scattered across many websites and code repositories, making identifying and curating information sources a time-consuming task.
Nevertheless, the SDE team successfully collected more than 600,000 scientific documents, paving the way for researchers to pursue transformative, interdisciplinary science.
engine fine tuning
The SDE user interface is customized to meet the needs of the scientific community. Users can perform free text searches and apply text-based facets to narrow results by selected parameters. Users can also filter by information type, such as data, images, and documents. Dataset landing pages are standardized to ensure a consistent and consistent user experience.
The main goal of SDE is to make the scientific research process more efficient by helping scientists find and access the data and information they need more quickly. To this end, SDE has been responsible for finding a path to implementing and operationalizing a new technology: Insight Engines. Insight Engine software applies relevance techniques to discover, analyze, describe, and organize content and data from a variety of sources. The advantage that Insight Engine offers over traditional search engines is that it incorporates natural language processing and machine learning. By incorporating these artificial intelligence technologies into the search process, we can enhance context and tailor searches to provide users with more accurate and relevant results.
early January, the SDE team has rolled out an updated user interface to make the search process even easier and more convenient. Improvements include additional filtering options that allow users to search within a single science topic area, such as planetary science or solar physics, as well as a new acronym search feature. With these upgrades and more, SDE is promoting research efficiency through its commitment to open science principles.
looking to the future
In the coming months, the team plans to complete the first round of curating and incorporating NASA science content into SDE. In addition, the SDE team also plans to fully deploy this tool in the fall of 2024. Ultimately, SDE has begun prototyping customized search applications that meet the needs of individual scientific audiences. For example, a specialized search interface designed for users interested in environmental justice information is already in development.
The SDE team is also testing advanced search techniques enabled by emerging technologies such as large-scale language models (LLMs). These models are changing the way searches are performed, and in the future search capabilities such as SDE will need to be adapted to enable both conversational and keyword searches. The SDE team is also interested in exploring how LLM can enhance the efficiency and accuracy of curation workflows.
Since its inception, SDE has been a beacon on the path to open science, illuminating the opportunities that arise when scientific information becomes accessible to essentially everyone. This comprehensive, lightweight search interface will continue to impress and inspire researchers as they explore ways to maximize scientific progress through information discovery and collaboration.
[ad_2]
Source link
