What if you could tell, with a certain degree of certainty, which U.S. states were the most corrupt? Even better, what if you could tell which ones were going to be mired in corruption even before any allegations, accusations, and convictions happen? That’s the vision shared by two researchers from the University of Valladolid in Spain, inspiring them to develop a fully-working model with the use of artificial neural networks in order to predict which Spanish provinces were more prone to corruption after a number of years.

The study, which was conducted by Félix J. López-Iturriaga from the University of Valladolid School of Business and Economics and Iván Pastor Sanz from the National Research University‘s Higher School of Economics in Moscow, Russia, was published in the journal Social Indicators Research. The study is titled, “Predicting Public Corruption with Neural Networks: An Analysis of Spanish Provinces.”

According to a report on the study, the researchers relied on a number of known factors in order to conduct their research. It is said that they considered data in relation to things like real estate tax, seemingly ridiculous price increases in housing, the establishment of new banking institutions, and the creation of new companies, among other things. Based on their research, it appeared that these things and more have been known to induce public corruption, and that when you add them all up, the overall result should be carefully examined in order to find what exactly is going on.

According to Ivan Pastor, one of the study’s co-authors, their study also issues further confirmation that the longer a politician stays in office, the higher the chances of him or her becoming susceptible to corruption. “As might be expected, our model confirms that the increase in the number of years in the government of the same political party increases the chances of corruption, regardless of whether or not the party governs with majority,” said Pastor.

He later added: “Anyway, fortunately, for the next years this alert system predicts less indications of corruption in our country. This is mainly due to the greater public pressure on this issue and to the fact that the economic situation has worsened significantly during the crisis.” Evidently, if the algorithm is able to detect and predict increased chances of corruption, it can also tell when or where it will be reduced.

The use of neural networks for this particular study is quite novel and could open the door to similar types of research. If the researchers there could use artificial intelligence and a deep learning neural net in order to come up with predictions regarding corruption in the government, there is a possibility that it could also be used for other industries. For now, the researchers are working with other organizations, such as Transparency International, in order to improve the results of the technology they used. Perhaps its use in predicting corruption in the medical industry or the pharmaceutical industry can also be possible.

Read more about interesting uses for artificial intelligence in Robots.news.

Sources include:

AgenciaSinc.se

Link.Springer.com