内容简介:The coronavirus pandemic of 2019 and 2020 and the civil rights crisis of 2020, led by the Black Lives Matter movement, have highlighted some of the major limitations of our society today.While our economy is growing at an increasingly rapid pace, many area
The coronavirus pandemic of 2019 and 2020 and the civil rights crisis of 2020, led by the Black Lives Matter movement, have highlighted some of the major limitations of our society today.
While our economy is growing at an increasingly rapid pace, many areas of development remain under-considered. A global vision of the interconnection between the different facets of development would help to achieve a sustainable improvement in living conditions.
“[Sustainable Development is:] Development that meets the needs of the present while safeguarding Earth’s life-support system, on which the welfare of current and future generations depends.” [1]
The 17 Sustainable Development Goals
The SDGs cover seventeen interconnected goals that define global quantifiable objectives across the social , economic , and environmental dimensions of development. [1] They aim to be a universal set of indicators and a reference framework to be leveraged by the global community to motivate policies and implementation by 2030.
They were set by the United Nations in 2015 as a follow-up to the Millennium Development Goals, which provided a framework for a global partnership to reduce extreme poverty.
This new framework supports a long term transition towards more sustainable development. It fosters accountability while also promoting global collaboration . It is a tool to guide decision making but not in and of itself a prescriptive and actionable guide. As a result, member states and institutions are free to pursue policies and programs according to their own context, resources, and available scientific evidence.
Interconnections and accountability
The global goals highlight how the 17 goals are interconnected . The increase in poverty is likely to lead to a decrease in hygiene and an increase in health challenges. Meanwhile, with poverty, access to education is harder and violence may emerge more easily. Similarly, Climate Change endangers wildlife (on land and below water) and will increase the already existing inequalities between people. This explains why Griggs D, Staord-Smith M, Ganey O, et al. [1] suggested that special emphasis should be placed on earth life support systems and poverty reduction .
“For the economy itself will die if the ecosystems collapse” Leonardo Di Caprio, UN Messenger of Peace
Sustainability areas can be considered as a large organism whose parts are interconnected. The economy exists within society, which in turn exists within the global ecosystem of the Planet Earth. This framework highlights that the economy cannot exist without society, which cannot function without the environment that surrounds it.
From the vision to actual actions and lasting changes
Global goals have been defined to inspire leaders. Yet, 17 general goals with 169 targets are not in itself sufficient to trigger lasting impactful decisions . In order to move towards more sustainable countries and businesses, we need to be aware of what changes will have a lasting impact. This requires global partnerships (Goal 17) as well as centralization and easy access to information.
From the community of researchers, scientists, entrepreneurs, and innovators, a large amount of knowledge, ideas, and new products are created on a daily basis, which helps in part to achieve the SDGs. However, much work needs to be done to gather knowledge and information and to make it usable by stakeholders and decision-makers.
For example, if a decision-maker is wondering: what can I do to improve education while reducing inequalities in the population? A clear set of scientific and business evidence should be readily available to support brainstorming on the solution.
Therefore, some additional information would help stakeholders to adopt sustainable changes:
- good and centralized access to historical data relevant to the goals
- scenarios and guidance on the impact of different policies or business changes, with evidence of positive and negative impact on different goals
- centralizing and summarizing researches and patents that are relevant to the goals
With particular emphasis on the interconnection and mutual impact of the different scenarios.
Leveraging Data and Artificial Intelligence for the SDGs
Artificial Intelligence will play a major role in improving access to knowledge. In particular, text mining and natural language processing will help to make sense of the ever-increasing amount of information published daily by the research community, industry, and governments.
The Trendscanner
One such initiative, led by the United Nations Development Programme (UNDP), the RISE Research Institutes of Sweden, and the AI for Good Foundation, is Trendscanner. This visualization browser maps major current trends such as IoT, Urbanization, and Big Data to the SDGs and brings together relevant reports and documents .
The SDG Data Catalog
With this in mind, together with James H. and the AI for Good Foundation, we are building the Sustainable Development Data Catalog , an open, extensible, global database of data sets, metadata, and research networks built automatically by mining millions of published open access academic works. The SDG Data Catalog is focusing on the issue of data discovery and data sharing to accelerate progress on the Sustainable Development Goals.
The process to generates this catalog is summarized in the system design below. From open-access databases gathering research papers information, a large dataset of published papers is created (5 Million papers in our first version). Those papers are then parsed to extract datasets mentions and relevant metadata associated with it.
The extraction phase is using a flexible web-scraper that is both scalable, light-weight, and copyright aware. The process has a 65% success in PDF retrieval.
The main component of the information extraction pipeline is the named entity recognition model that identifies dataset names, dataset description, owner, number of samples, and other references and document metadata.
Both the classifiers and NER models leverage an active learning strategy [3]. They start with a few manual annotations that are gradually exploited to train the model, suggest examples for annotation where the uncertainty is highest, annotate some more instances, and improve the overall accuracy of the task. This is made possible thanks to the Prodigy software [4].
As the next steps, we wish to build query and visualization layers on top of the SDG Data Catalog. In the end, we hope to create an easily navigable Wikipedia-like experience for keeping track of useful data .
FAIRification of data
While the amount of scientific data created is enormous, many research publications do not contain detailed descriptions of the data sets used and how to access them. This has multiple explanations. First, some data sets have great scientific values but will not give credit as such, so scientists may prefer to keep them private in order to exploit their analytical potential. [5] In addition, many data sets are privately held and retained as a competitive advantage by commercial entities.
The FAIR Data Project’s Commitment Statement in the Earth, Space, and Environmental Sciences [6] has been created to address this problem. FAIR data stands for Findable , Accessible , Interoperable , and Reusable . This commitment makes it more difficult to be published in most journals in these fields without including datasets that are used in a way that supports the FAIR principles.
Making data FAIR will improve research and science in several ways. It is a move towards open-source and the availability of the most recent discoveries and will make most science publications more reproducible and comparable .
Our work to build the SDG data catalog contributes to making SDGs data FAIR .
Conclusion
If not an actionable guide toward more sustainability, at least the global goals are a common language that companies, NGOs, and governments around the world can use to frame their policies and decisions. As Scientists, Researchers, Innovators, and Entrepreneurs, it is our role to make the bridge between the goals and the decisions that will lead to the necessary changes.
以上就是本文的全部内容,希望对大家的学习有所帮助,也希望大家多多支持 码农网
猜你喜欢:本站部分资源来源于网络,本站转载出于传递更多信息之目的,版权归原作者或者来源机构所有,如转载稿涉及版权问题,请联系我们。
Spark技术内幕
张安站 / 机械工业出版社 / 2015-9-1
Spark是不断壮大的大数据分析解决方案家族中备受关注的新增成员。它不仅为分布式数据集的处理提供一个有效框架,而且以高效的方式处理分布式数据集。它支持实时处理、流处理和批处理,提供了AllinOne的统一解决方案,使得Spark极具竞争力。 本书以源码为基础,深入分析Spark内核的设计理念和架构实现,系统讲解各个核心模块的实现,为性能调优、二次开发和系统运维提供理论支持;本文最后以项目实战......一起来看看 《Spark技术内幕》 这本书的介绍吧!