Smart Dublin explores how AI & Social Media can help improve the city regionBack to News, Blogs & Events
Dublin City Council, through the Smart Dublin initiative, is collaborating with AI technology company, Citibeats, to better understand how citizens experience the city region. Through social media analysis, local authorities can gain important insights into how citizens feel about key civic issues.
Smart Dublin, an initiative of the four Dublin Local Authorities, uses innovative technologies and new ways of working to improve quality of life in the city region. Working with technology providers, researchers, innovators and other public sector bodies, Smart Dublin works collaboratively to solve local challenges. But what is the best way to identify local challenges? The answer: listen to what local citizens have to say!
We live in the age of information where, globally, more people have mobile phones than have access to toilet facilities. Sharing our thoughts, opinions, complaints and concerns daily, through multiple digital channels, has become common practice for huge swathes of the world’s population. When we consider the sheer scale of the information that’s communicated online, the non-stop flow can appear overwhelming. While any information one might need or want is out there, sorting through this information, identifying what’s relevant and discerning meaningful insights is where it gets tricky.
With the help of artificial intelligence, this process has become more and more possible. Smart Dublin has teamed up with Citibeats to begin exploring how local authorities can tune in to what citizens are saying on social media and use this feedback to inform their work.
Citibeats is a social intelligence and speech analytics platform. Based on natural language processing (NPL) and machine learning, Citibeats organises unstructured data – in this case, the opinions of citizens expressed via social media, in an anonymous and aggregate manner – and uses AI models to categorise and decipher the data. From this data that’s presented via visual dashboards, insights are extracted and can be used by local governments and decision makers as action items to improve city life.
Citibeats provides an important tool for local authorities, but it is important to note that these insights are indicative only of the key trends and sentiments expressed by a cohort of citizens who use social media. Local authorities employ other methods to ensure truly inclusive citizen engagement.
AI and social media applied to natural disaster mitigation and consumer protection
AI and text analytics can also be applied to solve other challenges, such as natural disaster mitigation and regulation and consumer protection in the financial sector.
After the devastating flood in Japan in 2018, Citibeats teamed up with NTT Data, a technology company, and the Japanese Ministry of Economy, Trade and Industry (METI) to interpret social media activity after a natural disaster impacted three main cities in northern Japan. The machine-learning algorithms followed the online conversation and instantly interpreted the data to reveal the issues that were of most concern.
What was discovered was that each city had different priorities. While access to healthcare was the primary concern for two of the cities, food and mobility were most urgently needed in the other. By knowing where to focus resources first through this data-driven account of people’s needs in real time, government was able to allocate aid that was needed most. Most important was the increased speed of which data was translated into action: 5,000 damaged infrastructure reports from the 2018 Japan floods were extracted 21 days earlier than it would’ve taken without the use of this technology.
In Kenya, Citibeats played a major role in strengthening the financial services sector’s regulation. As these services became more digital – so did the consumers’ method of complaints and reports of abuse. It was impossible to manually process, let alone address, the thousands of complaints that were flooding in daily in an effective manner.
Topics were identified by the Financial Sector Deepening (FSD) Kenya as investigation-worthy such as scams, blacklisting and unfair charges. Citibeats’ AI was trained to recognise these events in all text-based data found online, in call reports, etc. and alert regulators when red-flag patterns were detected. Not only did FSD Kenya report that the Citibeats alerts received were 80-90% relevant for investigation-meriting issues, but they also found that consumer complaints were flagged 45 days earlier than before the technology was put in place.
Artificial intelligence for social good
Legitimate worries regarding the use of AI and technology have arisen in recent years. Stories of misuse, bias and ineffective practices have caused concern amongst citizens and civil society organisations.
But machine learning, when applied ethically and properly, can provide major opportunities for improvements and advancement. In the healthcare field, disease-detection artificial intelligence systems are used to scan skin images for cancer diagnosis; the financial services sector uses AI to protect consumers against fraud; advanced audio-sensor data is used for much-needed environmental conservation efforts around the world.
Machine learning has proven to be unequivocally faster and more efficient than humans at identifying, processing, classifying and executing certain tasks. This is immense power that, when harnessed for social good, can greatly improve society.
With the Smart Dublin project, the goal is to develop a new tool that will help local authorities to better understand how citizens feel about key civic issues. These valuable insights will support decision-makers to ensure that civic efforts and policy reflect the needs of citizens and communities.
It’s time to harness the power of advanced technology, AI and social media for good, and to improve quality of life for citizens and communities. Have a look at some of the previous reports that Citibeats have prepared for Smartdublin on the resources page : DublinBeat
Authors: Laura Kirchner, Project Manager, Citibeats. @LauraKirchnerSa www.citibeats.net