Government & Public Sector
Case Studies
Newsfor Government & Public Sector
A UK registered charity predicts visitor flow to its building with Markov chain algorithms
A UK registered charity leveraged technology from ASI to predict visitor flow. The provider used wifi usage data to track crowd movement within the building and gather information on the amount of time people spent in each location. ASI created a model indicating the most likely routes people would take through the attraction, congestion points and locations prone to overcrowding using Markov chain algorithms. The algorithm simulated the movement of 500 different hypothetical visitors over a fifteen minute period to come up with the results.
Linked use case: Optimise product layout in stores
Alto Data Analytics identifies and evaluates unsafe sanitation around the world with image recognition technology
Alto Data Analytics leveraged AI image recognition technology to analyse open source images of living conditions according to income levels around the world from ''Dollar Street''. The aim of their project was to present the global water crisis around the water based on hygiene and sanitary conditions in which people live. Based on the system capabilities of recognising toilet facilities, the research revealed that toilets from low-income families are less recognisable and suggested that 30-50% of the world''s population are likely to living in unsafe sanitary conditions.
Linked use case: Identify with satellite imagery areas of need for targeted support
Aylesbury Vale District Council''s services team now replies to residents’ questions almost two times faster using deep learning algorithms
The Aylesbury Vale District Council (AVDC) in the UK leverages AI to improve its customer service operation. As part of its Connected Knowledge programme it implemented DigitalGenius in Salesforce ServiceCloud with the aim of lowering queries response times and reducing costs. The deep learning algorithms are trained on historical logs and provide predictive case-intelligence and AI-enabled answers. The council has managed to cut the response time in almost half, offering its residents better customer service.
Linked use case: Enhance product and service offering
Baidu builds automated cat shelter for strays using image recognition
Baidu has created a "smart" shelter for stray cats to keep them warm during the winter months. The door to the shelter is automated and distinguishes between cats and other animals, including dogs, to ensure only felines get in.
Linked use case: Diagnose known diseases from scans, images, biopsies, audio and other data
Beijing Police identify people by their walking gait from up to 50 metres away
Beijing Police are using gait recognition technology to identify individuals based on their physical features when walking. The benefit of this technology is that it can be used from further away than facial recognition technology. One of the vendors of this software is Watrix, although the technology does not work yet in real-time.
Linked use case: Identify security-related individual targets from range of data including sensors, camera feeds and suspect activity
Belgian political party sp.a creates deepfake video of US President Donald Trump
Political party sp.a in Belgium created an online video using artificial intelligence to create fake video of US President Donald Trump discussing the Paris climate agreement in order to garner signatures for the party''s climate change petition.
Linked use case: Bring famous people such as actors, singers and artists back to life visually, recreating the distinctive nuances of their voice and movements in detail
Berkeley Lab to use deep learning to analyse alternate data sources for more accurate predictions of air quality
The US Department of Energy’s Lawrence Berkeley National Laboratory (Berkeley Lab) are using deep learning methods to analyse additional information, such as satellite images and cellphone data, to more accurately predict air quality levels.
Linked use case: Accelerate data integration from multiple sources
Build Change assesses quake-hit buildings safety and suggest cost-effective way to repair them using AI
Build Change and its Founder, Elizabeth Hausler, are leveraging machine learning to aid people whose homes have been affected by earthquakes and other natural disaster. The team trained a computer to analyse photos of houses and other data in order to assess the most cost effective and safe way to repair them. In the form of an application that users could download on their phone to receive the assessment, Build Change eliminated the need for lengthy assessment visits to each house in the area of Nepal.
Linked use case: Assess building damage from weather or natural disasters with drones and satellite delivered imagery
Centers for Disease Control and Prevention reduce polio report generation time to 1 hour using machine learning to automate regional mapping of the disease
The Centers for Disease Control and Prevention investigate evolving strands of the polio virus and map its activity dgeographically. Using machine learning has reduced the manual tie involved in generating reports to 1 hour and from clustering virus strains from over 3 months to 1 week.
Linked use case: Track and predict disease vector in general population
Chinese Government hopes to improve decision support to diplomats working on Bridge and Belt initiative through investment in AI tools
In a public attempt to reinforce the value of Chinese AI investment the Government has declared the roll-out of AI tools for its diplomatic teams working on Belt and Road initiatives to support the negotiation of infrastucture projects. The aim is to support geopolitical analysis and optimise Chinese performance in negotiations.
Linked use case: Optimise public policy decisions to take into account a greater set of complex interactions
Chinese facial recognition technology mistakes businesswoman''s face on bus ad for someone actually jaywalking
The city of Ningbo, China used facial recognition to identify people crossing the road illegally, referred to as ''jaywalkers''. The system is installed at six intersections in the city and has captured more than 7,800 incidents. However, the system showed its limitations as it got confused by the face of the famous Chinese businesswoman, Dong Mingzhu in an advertisement on the side of a bus. Authorities have admitted that the technology failed and have since update the system to prevent such errors.
Linked use case: Identify persons of interest to law enforcement through facial recognition
Circle Up helps startups raise funds via crowd funding by evaluating their potential and risk using machine learning
CircleUp plans to help entrepreneurs raise capital by matching them with potential investors. They analyse companies based on 90,000 data points and classifies companies based on potential and risk. They use this to predict likelihood of breakout success using machine learning. Investors can use this data to find interesting ventures.
Linked use case: Evaluate investment opportunities in early stage companies
City of Redwood City improves parking occupancy utilisation by 250% and optimises revenue using machine vision
The City of Redwood has implemented VIMOC Technologies in two multi-stoy parking garages in the city. The solution aims at detecting vehicles and accurately monitor and report the occupancy of the parking lots to optimise utilisation of the city''s infrastructure. Using machine vision the solution is able to track vehicles entering and exiting the parking lots and has increased garage utilisation and parking -related generated revenue.
Linked use case: Predict location and availability of parking space for car drivers
Cornell University researchers increase processing efficiency 300x by developing Adaboost algorithm to automate detection of elephant rumblings
Researchers develop an algorithm to automatically detect elephant rumblings from recorded sounds at certain jungle sites in central Africa. This allows a better understanding of the elephant population density at certain times which would be difficult to obtain through visual methods. The Adaboost algorithm used was an improvement in processing efficiency of over 300x compared to manual processing, and a recall rate of at least 70%.
Linked use case: Monitor animal populations
Cray''s CosmoFlow can now predict cosmological parameters with unprecedented accuracy with a deep learning 3D convolutional neural network
Cray, the National Energy Research Scientific Computing Center at Lawrence Berkeley National Laboratory (NERSC), and Intel have developed CosmoFlow, a deep learning 3D convolutional neural network (CNN). The system is aimed at predicting cosmological parameters with great accuracy. Such parameters include density, matter density fluctuations and power law index of the density perturbation spectrum after inflation. The research team achieved the same level of accuracy as existing experiments in estimating the values of Ωm and σ8, and five times less error than previous measurements using deep learning in estimating the value of Ns.
Linked use case: Discover anomalies in data scanned from space
Crisis Text Line identifies 85% of high-risk users in need of emergency intervention by applying machine learning to texts
The Crisis Text Line allows users in need of immediate mental health help to text a counsellor. Some of those users need emergency service intervention, so machine learning has been applied to discover relations between text message content and the requirement for immediate intervention, allowing the service to better prioritise users. The system identifies 86% of users at high risk for suicide within the first few messages.
Linked use case: Recognise predictors of emergency mental health intervention
Danish emergency service dispatchers identify heart-attacks in real-time emergency calls with 95% accuracy, compared to 73% for human dispatchers, with real-time speech analysis and machine learning
Danish Emergency Service dispatchers are assisted by real-time models that can predict with 95% accuracy, regardless of who is calling or the background noise, whether someone is having a heart attack. This compares to 73% accuracy for human dispatchers who have extensive training. The models have been trained on millions of hours of call.
Linked use case: Detect potential medical events from wearable sensor data and signal emergency response
Didi to tackle traffic congestion and optimise navigation routes with deep learning
Didi has leveraged artificial intelligence to develop a city traffic management system. The company, whose AI research labs are working on deep learning, natural language processing and machine vision, has launched the Smart Transportation Brain to combine video and sensor data from its vehicles and the government for traffic congestion and navigation optimisation. The company aims to facilitate the development of smart traffic lights and monitoring systems to alleviate road congestion.
Linked use case: Predict likelihood of traffic accidents or queues and optimise traffic system to reduce the risk
DoNotPay has succesfully appealed £2 million in parking tickets using a chatbot to contest tickets based on user inputs
The DoNotPay website and Messeger chatbot uses NLP to appeal parking tickets automatically, with an initial 40% claim success rate. The site''s bot is now being implemented for other legal uses, including for refugees legal claims and making complaints against landlords.
Linked use case: Codify legal expertise for automated deployment
Drivers demonstrate Tesla''s Autopilot limitations in recreation of fatal Model X crash at exact same location
Two Tesla drivers have almost recreated the fatal accident of March 23rd in Mountain View, California. The uploaded YouTube videos show them driver their Tesla cars at the exact same location using the Autopilot feature, which fails to identify faded lane markings.
Linked use case: Visualise and recognise road layout, other vehicles and pedestrians and potential risks to empower predictive driving
Elections Canada plans to use AI to identify electoral misinformation campaigns on social media
Elections Canada plans to use AI to monitor conversations on social media with regards to elections. The intent is to identify misinformation campaigns before they spread as a safeguard.
Linked use case: Censor user generated content on social media and other platforms
FSB saves loan application time for b2b lending using algorithms to match borrowers with lenders
The FSB fundraising platform was developed to match borrowers with potential lenders based on information inputted by the borrowers regarding their business and loan expectations. Lenders are sent anonymous borrow information and then register their interest, with the available options then borrowers are provided a list of potential lenders to choose from.
Linked use case: Automate credit risk profiling to support fundraising by small businesses through crowdfunding platform
Facebook create new location address schemes for developing areas decreasing travel times by 22% in tests
Facebook researchers develop a model of predicting roadways and other human infrastructure from satellite images in order to create an accurate addressing system, particularly for areas where no consistent system exists. The goal is to aid developing countries and improve direction accuracy to help disaster relief. The automatic addressing system covers more than 80% of populated areas.
Linked use case: Create maps from satellite and other remote imagery
Foster Care Tech (FCT) finds better matching foster care families for children using machine learning
Foster care system has remained the same throughout and experiences several challenges. Foster Care Tech is using machine learning to develop a system that matches characteristics of a child with that of a family to find better matches improving outcomes for everyone. Children placed using the system experienced fewer disruptions with 22.5% better placement stability.
Linked use case: Identify the right match between foster child and foster family
Funding Circle provides the only access to finance for 21% of its SME users SMEs by using machine learning in loan underwriting
Funding Circle enables small and medium businesses to apply for business to business loans through its streamlined application process, which includes machine learning for part of the underwriting process.
Linked use case: Automate credit risk profiling to support fundraising by small businesses through crowdfunding platform