Build an AI-powered Agricultural Assistant

Contact: James Hodson,  james@hodson.io

Short Description: SDG02 aims to achieve food security, improved nutrition and promote sustainable agriculture. Build an AI model that identifies patterns, similarities and differences in agriculture between developing and developed communities around the world.

Detailed Description: World hunger is on the rise, affecting 11 percent of people globally. There were an estimated 775 million undernourished people in 2014, and that number increased to 815 million by 2016. Family farming is part of the solution to the hunger problem. More than 90% of farms are run by an individual or a family and they produce about 80% of the world’s food, occupying around 70-80% of farmland. The objective of this project is to model patterns, similarities and differences in grain growth, agricultural practices and soil/climate conditions among the most hunger stricken areas of the world. The results will be used for effective decision-making at multiple levels across the agricultural value chain and to develop solutions for optimized agricultural practices of small scale farming. Projects will include Machine Learning models, data management platforms, and visualization engines to allow communities to interact with the data and assist in decision making. Successful projects will have the opportunity to present their products in front of community leaders, researchers, and policy-makers at the AI for Good Foundation Global Conference in 2019!

 

Possible Data Sets:

Data.Gov: https://catalog.data.gov/dataset?groups=agriculture8571#topic=food_navigation

Sustainable World: http://www.sustainableworld.com/data/index.html

Open Data Charter: https://opendatacharter.net/agriculture-open-data-package/

USDA: https://www.usda.gov/topics/data

Africa Information Highway: http://dataportal.opendataforafrica.org/gqzdwxe/agriculture

The World Bank: https://data.worldbank.org/indicator

Syngenta: http://opendata.syngenta.agroknow.com/the-good-growth-plan-progress-data

Monsanto: https://monsanto.com/news-releases/monsanto-boundless-collaborate-open-source-gis-contributions/

Open Climate Data: http://openclimatedata.net/

Africa Climate: http://africaclimate.opendataforafrica.org/

ISRIC: https://www.isric.online/explore/soil-geographic-databases

NASA World Wind: https://worldwind.arc.nasa.gov/

Svalbard Global See Vault: https://www.nordgen.org/sgsv/

WHO: http://www.who.int/nutrition/databases/en/

Global Nutrition Report: http://globalnutritionreport.org/the-data/dataset-and-metadata/

Grain Growth: https://pubs.usgs.gov/ds/344/html/grain.html

More data…

Build an AI-powered Poverty Eradicator

Contact: James Hodson,  james@hodson.io

Short Description: SDG01 aims to implement nationally appropriate social protection systems and measures for all. Build an AI system that identifies the needs of asymmetric demographic changes in communities.

Detailed Description: Today, 13% of the global population consists of people aged 60 or over. Although Europe has the greatest percentage (25%) of population in this age group, rapid ageing will occur in other parts of the world as well. By 2050, all regions of the world, except Africa, will have 25% or more of their populations aged 60 and above. This worldwide ageing will result in a decline in working-age populations that will cause major challenges for some economies and government budgets. Therefore, social security systems, especially in Europe, will come under serious pressure (SFA 2017 Report, p. 36).

The objective of this project is to model asymmetric demographic change in communities to identify the need to build an appropriate infrastructure for elderly and develop social programs for population living in the area. Projects will include Machine Learning models, data management platforms, and visualization engines to allow communities to interact with the data and assist in decision making. Successful projects will have the opportunity to present their products in front of community leaders, researchers, and policy-makers at the AI for Good Foundation Global Conference in 2019!

 

Possible Data Sets:

The World Bank: https://data.worldbank.org/topic/poverty

Global Partnership for Sustainable Development Data: http://www.data4sdgs.org/

OECD Data: https://data.oecd.org/inequality/poverty-rate.htm

Data.Gov: https://catalog.data.gov/dataset/demographic-statistics-by-zip-code-acfc9

Data.Gov: https://catalog.data.gov/dataset/directory-of-multi-purpose-senior-centers

Global Open Data Index: https://index.okfn.org/dataset/statistics/

Eurostat: here.

European Data Portal: https://www.europeandataportal.eu/en/homepage

The World Bank: https://data.worldbank.org/region/european-union

Data Science Central: Here.

UN DESA: http://www.un.org/en/development/desa/population/publications/dataset/index.shtml

More data…

Build an AI-powered Traffic Safety Guide

Contact: James Hodson,  james@hodson.io

Short Description: SDG03 aims to halve global deaths and injuries from road traffic accidents by 2030.

Detailed Description: Each year, more than 1.25 million people die as a result of traffic accidents, and between 20 to 50 million more suffer from non-fatal traffic related injuries. This causes substantial physical and economic losses to affected individuals, their families and nations as a whole. According to WHO, the cost of traffic accidents sums up to 3% of GDP in many countries.

Projects will include data mining, machine learning and predictive analytics models and algorithms, applied over static and possibly real-time traffic related data, to be able to detect, visualize and even predict dangerous road sections and conditions. The results will lead towards a traffic safety data management platform and visualization engine, which could be used to assist planners and policy makers to improve the infrastructure. Additionally, the same platform combined with real-time data could be used to develop services for end-users, assessing their real-time traffic safety score and adding a safety related aspect to the routing engines and navigation devices. Successful projects will have the opportunity to present their products in front of city leaders, researchers, and policy-makers at the AI for Good Foundation Global Conference in 2019, and FIA + EuroRap conference on traffic safety in 2019. Depending on results, AI4Good will assist with forming a consortium of partners to bring the research results into real-world service.

 

Available Data Sets:

– 3 years of Traffic Related Data from the whole country of Slovenia (available on request)

– List of online, real-time live traffic related data from Slovenia (use JSON formatter): http://traffic.ijs.si/API/info/getApiInfo

– Star safety rating of EU roads (We need to contact EuroRap) – they have the data and are usually willing to share

https://catalog.data.gov/dataset?q=traffic (1480 traffic-related data-sets from US, including accidents)

– Bureau of Transportation Statistics: https://www.bts.gov/browse-statistical-products-and-data

– Live accidents around Houston: http://traffic.houstontranstar.org/datafeed/datafeed_info.aspx

– Collecting your own mobility patterns data using existing mobile apps, or a library to develop your own: http://traffic.ijs.si/NextPin/aboutus.html

https://play.google.com/store/apps/details?id=net.nextpin.example.collection

– Road traffic death rates statistics: https://resourcewatch.org/data/explore/Road-Traffic-Death-Rates

 

Build an AI-powered Urban Development Architect

Contact: James Hodson,  james@hodson.io

Short Description: SDG11 aims to ensure access to inclusive, safe, resilient and sustainable housing. Build AI models for integrated human settlement planning.

Detailed Description: In the world, nearly 1 person is forcibly displaced every two seconds as a result of conflict or persecution. We are now witnessing the highest levels of displacement on record.

An unprecedented 68.5 million people around the world have been forced from home.

The objective of this project is to model displacement patterns and displacement effects on urban development in countries which are accepting a surge of immigrants coming from politically unstable territories. Suggest sustainable urban development plans, affordable housing programs, and inform policy decisions at the local, and country-wide levels. Projects will include Data Mining, Machine Learning, data management platforms, and visualization engines to allow cities to interact with the data and assist in decision making. Successful projects will have the opportunity to present their products in front of city leaders, researchers, and policy-makers at the AI for Good Foundation Global Conference in 2019!

Available Data Sets:

UN Habitat: http://urbandata.unhabitat.org/

Global Partnership for Sustainable Development Data: http://www.data4sdgs.org/

The World Bank: https://data.worldbank.org/indicator/sp.pop.totl

The World Bank: https://data.worldbank.org/indicator/EN.POP.SLUM.UR.ZS?view=chart

The World Bank: https://data.worldbank.org/products/wdi-maps

Open Government Platform India: https://data.gov.in/keywords/slums

UNHCR: http://popstats.unhcr.org/en/overview

data.world: https://data.world/datasets/immigration

iDMC:http://www.internal-displacement.org/database

Migration Data Portal: https://migrationdataportal.org/iom-data

World Inequality Database: https://wid.world/

HUD: https://www.huduser.gov/portal/home.html

IDMC: http://www.internal-displacement.org/database/displacement-data

Stanford center on Poverty and Inequality: https://inequality.stanford.edu/income-segregation-maps

More data…

Build an AI-powered System for Sanitation Improvements

Contact: James Hodson,  james@hodson.io

Short Description: SDG06 aims to ensure availability and sustainable management of water and sanitation for all. Build an AI model that identifies risks to human health caused by polluted drinking water and lack of sanitation.

Detailed Description: Today, 2.1 billion people lack access to safely managed drinking water services and 4.5 billion people lack safely managed sanitation services. A person without access to improved drinking water – for example from a protected borehole well or municipal piped supply for instance – is forced to rely on sources such as surface water, or other unprotected water sources. In the immediate environment, exposed fecal matter will be transferred back into people’s food and water resources, helping to spread serious diseases such as cholera, typhoid, infectious hepatitis, polio, etc. The lack of effective waste disposal or sewerage systems can contaminate ecosystems and contribute to disease pandemics. The objective of this project is to model disease occurrence and human mortality in areas that lack sanitation to identify where to secure the drinking water resources, suggest sanitation improvements, and inform policy decisions at the local, and country-wide levels. Projects will include Machine Learning models, data management platforms, and visualization engines to allow communities to interact with the data and assist in decision making. Successful projects will have the opportunity to present their products in front of community leaders, researchers, and policy-makers at the AI for Good Foundation Global Conference in 2019!

Possible Data Sets:

USGS: https://water.usgs.gov/owq/data.html

EU Open Data Portal: https://data.europa.eu/euodp/data/dataset/data_waterbase-water-quality

FAO UN: http://www.fao.org/nr/water/aquastat/main/index.stm

WHO: http://apps.who.int/gho/data/node.home

Waste Atlas: http://www.atlas.d-waste.com/

JPM: https://washdata.org/data

The World Bank: https://data.worldbank.org/indicator/SP.POP.TOTL

UN Data: http://data.un.org/

NASA: http://neo.sci.gsfc.nasa.gov/

More data…

Build an AI-powered Ocean Life Protector

Contact: James Hodson,  james@hodson.io

Short Description: SDG14 aims to conserve and sustainably use the oceans. Build AI models that help preserve ocean life.

Detailed Description:  8,300 million metric tons of plastics have been produced to date and 91% of that was not recycled. Studies estimate there are 15–51 trillion pieces of plastic in the world’s oceans.  Marine mammals, fish, sea turtles and seabirds ingest plastic every year. In our oceans alone, plastic debris outweighs zooplankton by a ratio of 36-to-1. A recent study found that a quarter of fish at markets in California contained plastic in their guts, mostly in the form of plastic microfibers. Plastics also affects human health. Toxic chemicals leach out of plastic and are found in the blood and tissue of nearly all of us. Exposure to them is linked to cancers, birth defects, impaired immunity, endocrine disruption and other ailments. The objective of this project is to model how animal migratory patterns cross paths with sea garbage patches to predict where the sea animal species come in contact with the highest quantity of plastic particles and how that influences the food chain. The results will help to inform policy decisions at the local, and country-wide levels. Projects will include Machine Learning models, data management platforms, and visualization engines to allow countries to interact with the data and assist in decision making. Successful projects will have the opportunity to present their products in front of country representatives, researchers, and policy-makers at the AI for Good Foundation Global Conference in 2019!

Possible Data Sets:

World Fish: https://www.worldfishcenter.org/databases

COPEPOD: https://www.st.nmfs.noaa.gov/copepod/

Data.Gov: https://catalog.data.gov/dataset/gray-whale-population-count-data

Ocean Data Viewr: http://data.unep-wcmc.org/

CEN: http://icdc.cen.uni-hamburg.de/1/daten/ocean/oscar-oceansurfacecurrent.html

NOAA: https://www.nodc.noaa.gov/oads/stewardship/data_assets.html

US Coast Guard: https://homeport.uscg.mil/Lists/Content/DispForm.aspx?&ID=211&Source=https://homeport.uscg.mil/missions/investigations/marine-casualty-pollution-investigations

https://www.data.gov/

Plastic Adrift: http://plasticadrift.org/?lat=44.4&lng=13.4&center=-5&startmon=jan&direction=fwd

NASA: http://neo.sci.gsfc.nasa.gov/

More data…

Future Energy Systems

Contact: Ashkan Yousefi, ashkan.yousefi@berkeley.edu

Description: High volumes of data are becoming available with the growth of the advanced metering infrastructure and massive deployment of IoT devices. These are expected to benefit the planning and operation of the future energy systems and to help the customers transition from a passive to an active role. In this project, we explore a novel approach using deep reinforcement learning, a hybrid approach that combines reinforcement learning with deep learning, to perform on-line optimization of schedules for building energy management systems. The learning procedure gets benefited from two methods, Deep Q-learning and Deep Policy Gradient. The hybrid approach is capable of handling multiple actions simultaneously. The large-scale Pecan Street Inc. database will be used to validate the proposed approach. The Pecan database is highly-dimensional database includes information about photovoltaic power generation, electric vehicles as well as buildings appliances. In addition, these on-line energy scheduling strategies will be used to provide real-time feedback for electricity customers and prosumers to achieve more efficient use of electricity.

PilotCity: Students are from Mars, Employers are from Venus

Contact: Derick Lee , dericklee@pilotcity.com

Description: Imagine an engine for innovation for small-to-medium sized cities… Okay now imagine the local high school students of these cities becoming the protagonist for civic transformation in their cities. PilotCity is creating an engine for innovation for cities by building career pathway systems for students to enter the workforce. We do this by hosting in-classroom project-based challenges driven by employers that lead to at-workplace work-based experiences such as internships and fellowships. How would you simulate the presence of an employer in the classroom to be the best project advisor to the student while saving time for the busy working professional? Developing on platform technologies such as voice-activated assistants such as Google Home, and/or telepresence technology such as Double Robotics – prototype a “wormhole” solution for students and employers to communicate during these multi-week project-based challenges to accelerate project advisory despite the vast connection of space and time between the classroom and the workplace.

NIST Education Super Cluster

NIST Education SuperCluster: Living Blueprint Engine for Education Innovation in Smart Cities

Contact: Derrick Lee , dericklee@pilotcity.com

Description: In the early development of smart cities, what does it take to educate a smart and connected citizen? The NIST Education SuperCluster is a consortium of educators, industry leaders, and governmental officials formulating blueprints for education innovation in smart cities under the Global City Teams Challenge, a federal smart city initiative by the National Institute of Standards & Technology (NIST) under the U.S. Department of Commerce. We are seeking for university-level student fellows to prototype a data analysis and algorithmic engine we can implement to streamline the processing of case studies to create a “living blueprint” that will inform leaders across the nation and globe of the best practices in education innovation in smart cities.

The Jokester

Contact: Brian Bordley, Skydeck: brian AT skydeck DOT vc
Description: Artificial intelligence technology is transforming many services and brain capital based industries – it’s time for comedy writing to get some help! The project should take a subset of jokes (political, raunchy, self-deprecating, etc) and a subset of video medium combining audio and visual data (laughs can be the positive reinforcing event) to create a tool to help joke writers. The project does not need to pump out a good joke every time, but should try to write something every once in a while. It could also take some data from the writer to come up. Google and Amazon already have great toolkits for this kind of thing.