Shulou(Shulou.com)11/24 Report--

CTOnews.com, March 3 (Xinhua)-- IBM and NASA's Marshall Space Flight Center recently announced a partnership to use IBM's artificial intelligence (AI) technology to discover new insights from NASA's vast amounts of earth and geospatial science data. This cooperation is the first time that AI basic modeling technology has been applied to NASA Earth observation satellite data.

CTOnews.com Note: the basic model is a type of AI model trained on a wide range of untagged data that can be used for different tasks and can be used to apply information from one situation to another.

According to IBM, the basic model has rapidly advanced the technological development in the field of AI natural language processing (NLP) in the past five years, and now IBM is exploring and creating more basic model application scenarios beyond the NPL domain.

Earth observation makes it possible for scientists to study and monitor the Earth. At the same time, this work is collecting data at an unprecedented speed and quantity, and there is an urgent need for new and innovative methods to extract knowledge from these huge data sources. The goal of the collaboration between IBM and NASA is to provide researchers with an easier way to analyze and gain insights from these large data sets. The basic modeling technology of IBM is expected to speed up the discovery and analysis of these data, so as to quickly promote human scientific understanding of the earth and response to climate-related issues.

IBM and NASA plan to develop several new technologies to extract insights from Earth observation. One of the projects will train IBM's geospatial intelligence basic model on NASA's Harmonized Landsat Sentinel-2 (HLS) data set. The HLS dataset records the data of land cover and land use change captured by Earth-orbiting satellites, and analyzes these PB satellite data to determine the geographical footprint changes of natural disasters, periodic crop yields and wildlife habitats. This basic modeling technology will help researchers to make an important analysis of the earth's environmental system.

Another output of this collaboration is expected to be an easy-to-search corpus of geoscience literature. IBM has developed a NLP model that can organize relevant literature and discover new knowledge more easily after training in nearly 300000 articles in geoscience journals. This well-trained model includes the largest AI workload ever trained on Red Hat OpenShift software, using IBM's open source multilingual question answering system PrimeQA. In addition to providing resources for researchers, new language models of geosciences can also be injected into NASA's scientific data governance and management processes.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.