Dialnet will apply artificial intelligence to analyze sentiment in documents

Logroño (EFE).- The Dialnet Foundation faces the challenge of applying artificial intelligence (AI) in sentiment analysis to verify how another scientist cites and to measure these documents qualitatively, not just quantitatively.

This was detailed on Thursday, in statements to journalists, by the Managing Director of the Dialnet Foundation, Elena López Tamayo, who is participating in the “Dialnet Global” conference, which brings together in Logroño the main agents of science in Spain and Latin America.

Dialnet is one of the largest bibliographic portals in the world, containing 9 million reference documents and 3,000,000 doctoral dissertations, available to the entire scientific community and society in general in a free and open way, with more than 3 million registered users.

This portal was created at the University of La Rioja (UR) in 1999 as a collaborative project, and 15 years later, it chose to develop the Dialnet Global project, whose main goal is to turn Spanish into the second scientific language after English. .

“This Dialnet data should be visible in intelligent information systems that allow drawing conclusions and analyzing sentiment on how one scientist cites another, whether it is something positive or only neutral, which will be achieved as technology develops.” “, explained Lopez Tamayo.

A window to the world

For his part, the Rector of the University of Uruguay, Juan Carlos Ayala, has chosen to create a personal network of all scientific institutions in Spanish which allows them to work together, which is why around 200 people from Spain and Spain are participating in the conference. Latin America, representing universities, libraries and other entities.

He pointed out that Dialnet was born in a small university as “a window into the world of the Spanish university system as a whole” and would allow the idea of ​​“the possibility of disseminating and disseminating science in the Spanish language” to be consolidated.

Ayala emphasized that what really mattered to the world a few years ago were publications in the pure sciences, but the social sciences, legal sciences and humanities were “a bit abandoned.”

For this reason, Dialnet was able to show those areas in Spanish that “until now remained hidden,” he added.

In this process, artificial intelligence systems will be used to generate “advanced knowledge”, which will allow “diving” into the 9 million documents stored in this portal, which will make it possible to identify a specific topic of interest to a specific community of scientists or researchers. He works.

AI to expand the rules

Coordinator of the content working group of the universities cooperating with Dialnet, José María García Avilés, pointed out that so far, publications that have already been digitized have been entered into this portal, but there are a lot of old bibliographic collections that have been documentary and scholarly.

This value has a “very short duration” in the case of pure science, but much longer in the humanities, so AI can help a lot in “expanding the knowledge base in a simple, agile and fast way.”

On the other hand, the Director of the Scientific Production Area of ​​the University of Valladolid, Juan García Serna, pointed out that one of the great functions of Dialnet CRIS (abbreviation in English for Current Research Information System) is to show, in a friendly, useful and quick way, all the scientific production of universities and check their interconnectedness.

“The challenge that this portal must face is to ensure better dissemination of science to the entire community, and not to focus on publishing only,” he said.