Reading Time: 3 minutes
We interview Alejandro Martín, Assistant Professor at the Computing Systems Department of the Polytechnic University of Madrid (UPM). Member of the Applied Intelligence and Data Analysis research group, one of the winners of the call for solutions by Digital Future Society: under the question “How can technology help fight disinformation optimising the fact-checking processes?”.Know more about the call Tech Against Disinformation
The UPM proposal is a system / platform for the automated detection, monitoring and analysis of false claims on Social Networks. The solution combines AI-based multilingual natural language processing (NLP), social media analytics, and information tracking on social media. The solution implements a multilingual architecture recognizing languages such as: Spanish, English, Catalan, Basque or Galician, among many other languages. This allows you to compare information in different languages without translation.
Once a verifier has detected a rumor or a false claim, the solution helps to track its presence in the RRSS, tracing all its history, from the first appearances in social networks to its current state. The solution is independent of any social network, easily adapting to any of them.
The solution has a TRL of 5. It has been designed and developed by researchers from the Applied Intelligence and Data Analysis research group, from the Department of Computer Systems Engineering of the Polytechnic University of Madrid.
Alejandro, would you explain to us a bit of your proposal and how it works? What is the objective for your solution, and how is it supposed to be used?
Well, thank you for your questions. Our main goal is to help fact-checkers do their work. We know that there is a lot of misinformation in all networks. So our goal here is to help them verify information more intuitively using AI techniques, combine the role of computer science, and apply it to a specific solution, for instance, automatically verifying new claims in Twitter or Telegram.
For instance, we can incorporate mood analysis and compare sentences in different languages to also examine, for example, if the feeling of a claim is negative, or positive which is very similar to the so-called automated fact-checking. So that's our goal: to help checkers.
Since you've been working around the topic, how worrying do you think misinformation is nowadays?
I think we have a concerned population about this problem, but we can also see that there are not enough measures to tackle this problem. There is a lack of regulation on the Internet. We can watch messages sent from other countries, and there is no revelation between countries who can send a message or who can send a message to thousands of people. So I think the main problem is the lack of regulation.
We can also see that messaging services are used to share information without any kind of control or verification. So in the field of social networks, there's a lot of work to do. We need their participation to analyse the information; it could be beneficial for the population.
Where do you think technology stands in all this information issue? Why do you find it essential to use technology to tackle these issues?
Well, I think the best example is to see the checker’s work and to see how difficult it is to analyse such a vast amount of misinformation that spreads every day on social networks. We need to provide faster AI techniques to help them make decisions more intuitively and efficiently. So, for instance, if a hoax is identified, we can provide tools so that in the future, if we see that hoax again, we can automatically detect it and inform the user.
So I think in this line, computer science is going to be helpful for these matters.
According to your experience, how would an optimal fact-checking process work?
It's going to be a process where new technologies will play a fundamental role in automatically finding new hoaxes and taking a proactive approach in detecting, as soon as possible, one piece that was initially a piece of information that wasn’t true. So I think that the links can help define that kind of trend in a social network.
And the last question, who do you think is the final responsible actor who needs to do something to avoid becoming a misinformed society?
Well, I think this is a question that has a lot of answers, and it's pretty complicated to answer. Journalism provides us with good information that if we trust classical journals, we will see information that can be charged. On the contrary, social networks are playing the role of spreading disinformation, misinformation and data with no kind of verification. So I think one of the main problems lies in the lack of verification, the misinformation spread in social networks. And of course, as I said before, the lack of regulation is one of the main problems that we should address in the following years.