The collection of data and the correct analysis of it is essential in order to be able to locate people, with artificial intelligence this task could become easier.
In Mexico, enforced disappearances is a serious problem that goes back many years, tracking and identifying missing people has always been a major challenge, but big strides are being made today, thanks to the use of Artificial Intelligence (AI).
Although the concept of AI has been around for more than 60 years, with the phenomenal expansion of technology and the ability to process large amounts of information, the uses in this area have increased significantly in recent years, affecting every year and better results.
Therefore, the International Meeting on Artificial Intelligence and its Applications (RIIAA) together with the National Search Commission (CNB) and the Center for Research in Mathematics (CIMAT) will hold the hackathon “Justice for the Disappeared”, a 100% online meeting, which is about generating AI-based solutions for the analysis of thousands of relevant documents from the time of political repression in the country and thus being able to clarify the fate of the victims.
It was between 1964 and 1985 when the Mexican state bitterly confronted and oppressed those whom it viewed as its internal enemies: social movements, leaders of politico-military organizations, sympathizers, and even their families and communities for whom it systematically promoted enforced disappearances .
From that moment on, there are innumerable physical documents that are gradually digitized as images, the information of which needs to be integrated and structured. Artificial intelligence is one of the technologies that makes this task easier because one of the main advantages of AI is that it allows you to manage large amounts of data efficiently, which adds value by organizing and analyzing documents in a timely manner and obtaining more accurate results.
Likewise, to identify faces, the AI ​​analyzes dozens of details, such as the proportions of the face, the dimensions of the nose, the curvature of the corners of the mouth, the distance between the pupils of both eyes, many others high quality features that will help you find missing people.
It is estimated that the use of new technologies, algorithms and deep learning models can achieve a probability of greater than 90% in finding missing persons.
The hackathon, which takes place from Friday, August 20th to Tuesday, August 24th, calls on teams of three to five people of Latin American origin to participate.
During these days the participants will face two great challenges:
- Cleaning Images and Entities: Teams need to propose and develop a preprocessing flow to improve the quality of historical images and identify the objects that make them up, their location and their classification according to similarity according to general categories such as stamps, signatures, calligraphy, Typewriting and photos.
- Text Extraction: The aim is for the participants to propose and develop an image analysis flow for extracting texts from the documents with the highest possible quality and extracting entities (people, organizations, places and dates).
It should be noted that the teams are scheduled to have meetings with mentors who will be responsible for accompanying them in planning and implementing their proposals.
At the end of the meeting, where the Ai Lab School and Talentum also work together, each challenge will be assessed and rewarded separately, and the working groups that come first in each category will be awarded prizes that will help them advance their careers in AI -Development Solutions.
About Hackathons
Hackathons are community settings where ideas are brought to life through the right ecosystem for creativity, work and entrepreneurship. Here you can create anything, there are no limits to your creativity. Among other things, teams come together to develop a product or service of minimal value. Hackers from all over the world unite in their local hackathons, this is where new tech entrepreneurs are born with success and ingenuity.
Source link
