Using AI and Social Media to Detect Noisy Areas
See how we can use AI and social media to detect noisy areas.
Join the DZone community and get the full member experience.Join For Free
Noise disturbances are far and away the most common form of anti-social behavior reported to the police and local authorities, and I'm sure we've all experienced the blight of loud parties. For officials, however, it's likely that the number of reported incidents is a fraction of the total number.
Researchers from the Universidad Politécnica de Madrid have developed an AI-driven tool that they believe will provide officials with a more accurate representation of noise concerns.
The work, which was documented in a recently published paper, mines social network data before performing a text analysis to autonomously detect and then map complaints about noise pollution. What's more, the system was also able to accurately predict the onset of noisy events to try and help officials to plan their resources and intervene more effectively.
Traditionally, a mixture of surveys and official reports have been used to gauge noise concerns in a neighborhood, but these offer something of a blunt instrument that is not all that effective. It also fails to reflect specific events, and while various online reporting tools have been developed in recent years, usage remains low.
Social media on the other hand, is a commonly used platform for venting frustration about noisy neighbors.
"For years, companies have been applying machine learning and natural language processing techniques to find out the opinion of clients about their products and services in social media in order to improve sales. However, this technological trend has not been applied in city management, missing social media posts, which can provide real-time data about issues in a city," the researchers explain.
The team developed a text analysis tool that can detect complaints made on social media automatically before then classifying them according to their origin. The tool relies on a combination of machine learning and language analysis to not only detect noise pollution more effectively but to forecast events ahead of time.
While the tool has a wide range of possible uses outside of noise pollution, such as in understanding citizens reactions to urban planning decisions, the team believes it can have the biggest impact in the short-term in helping to root out unsocial behavior.
The team is now working with collaborators, such as Télécom ParisTech, to further develop the technology and test it in a live environment, after which it can hopefully begin to be deployed in a practical setting.
Published at DZone with permission of Adi Gaskell, DZone MVB. See the original article here.
Opinions expressed by DZone contributors are their own.