Image of ericbarns on Pixabay
A team of researchers from the Polytechnic University of Madrid (UPM) has developed a text analysis system that, applied to the comments published on social networks, is able to automatically detect complaints about acoustic pollution and classify them according to their origin.The system combines artificial intelligence-specifically machine learning-with various language analysis techniques.In addition, allows predicting the occurrence of noisy events, which can help urban managers to design early interventions to avoid inconvenience and health problems for citizens.The study has been developed in collaboration with Telecom Paristech.
Related Acoustic pollution It affects 25% of that of the European population.Of the 500 million people in the European Union, it is estimated that 125 are exposed to noise levels higher than those recommended by the World Organization Health: This causes public health problems and losses in the quality of life of the population, especially in urban settings, associated with the lack of rest and stress generated by this annoying situation.
acoustic pollution and digital revolution
With the arrival of the digital revolution, the population has changed the way of communicating and using technology.Today 4200 million people have an Internet connection, there are more than 3000 million users who actively use media social communication and it is estimated that each user has an average of 5.5 accounts on these platforms. The users of these networks provide their opinions and feelings on a multitude of topics: politics, television, products and of course the environment, including here acoustic pollution.
As Luis Gasco, researcher of the research group in Instrumentation and Applied Acoustics (I2A2) of the UPM, points out, "for years now companies have been applying machine learning and natural language processing techniques to know the valuation of their clients.about their brands and products on social networks with the intention of improving their sales.
However, this technological trend has not been replicated in urban management , losing the information of a communication channel used by thousands of citizens, such as social networks, and which can provide data in real time about problems in the city quickly."
This is precisely what the project research team has developed, or a text analysis system that is capable of automatically detecting complaints about acoustic contamination and classifying them according to their origin.
language analysis
For this they have used the latest techniques in artificial intelligence, such as machine learning, and various language analysis techniques.Additionally, through statistical techniques they have designed a prediction system that allows to know the occurrence of an annoying noisy event from the temporal evolution of the number of complaints and specific words used in these.
The applications of the system developed by the UPM researchers are not limited to the field of acoustic contamination since, as indicated by those responsible for the project, "the same methodology could be applied for the detection of problems of another one in a city , from damage to urban furniture to the opinion of citizens about changes in urban planning in a city such as, for example, the semi-pedestrianization of the Gran Via in Madrid, something that would be of great help for the creation of more inclusive municipal ordinances ".
goodbye to polls
Traditionally, surveys have been the tool used to know the citizen's perception of the acoustic environment in urban environments.These means have some significant disadvantages such as low citizen participation and the cost associated with their development and analysis.
In addition, do not allow a manageability of the problems detected, or of specific noisy events that, however, can cause great discomfort to the population. On the other hand, in recent years they have appeared System of online citizen participation that allow faster interaction with urban managers, but which, however, are not widely used by the population, probably because they use a specific channel in which they do not feel comfortable.
Source: UPM
Comments
Post a Comment