Museums have tons of data, and AI could make it more accessible

Ice core in the freezer, dinosaurs on display, fish in jars, birds in boxes, human residues and ancient artifacts from long-lasting civilizations that ever see-Mujium collections are all more and more full.

These collections are treasury traves that recall the natural and human history of the planet, and they help scientists in different fields, such as geology, fossil science, anthropology and more. What you see on a museum visit is only a piece of miracles held in their collection.

Museums usually want to provide physical or digital content of their collections for teachers and researchers. However, employees of each collection have their own way of organizing data, so navigating these collections can prove to be challenging.

Digital copies of museum samples or collection require information about physical objects, systematic and distribution Incredible dataAnd this data can feed in machine learning models or other artificial intelligence To answer big questions,

Currently, even within a single research domain, various repository needs to be navigated to find the right data. AI can help in order to organize large amounts of data from various collections and to remove information to answer specific questions.

But using AI is not a perfect solution. A set of shared practices and systems for data management between museums can improve data cursion and share AI the necessary to do its work. These practices can help both humans and machines make new discoveries with these valuable collections.

In form of Information scientist Who studies scientists’ research data management and opinions, I have seen that the world’s physical collection is infrastructure objects and a patchwork quilts for them. Affiliated metabolism,

AI equipment can do amazing things, such as Create 3D models Of Digitized edition Among the items in the museum collection, but only if sufficiently streamlined data about that item is available. To see how AI museum can help in collection, my team of researchers started Focus groups Along with those who managed the museum collection. We asked what they are doing to use their collection by both humans and AI.

Collection manager

When an item comes in a museum collection, the collection managers are those who describe the characteristics of that item and generate data about it. That data called metadata allows others to use it and the collector’s name, geographical location, which was collected, and, in the case of geological samples, this is that apoch. For samples from an animal or plant, it may include Its classificationWhich is a set of Latin names that classify it.

All together, this information adds up to an arbitrary zodiac of data.

But the combination of data in the domain with different standards is really difficult. Fortunately, the collection managers are working to standardize their procedures Beyond discipline and for Many types of samplesThe grant has helped science communities to create equipment for standardization.

In biological collection, Specify the equipment The managers allow the managers to quickly classify samples with the drop-down menu with standards for taxonomy and other parameters to continuously describe the samples.

There is a common metadata standard in biology Darwin coreSimilar well -installed metadata and equipment are present in all sciences to take actual items and pour them into as easy as possible.

Special tool collection managers such as these and metadata make data from their objects re -purpose for research and educational purposes.

all the small things

My team and I held 10 focus groupWith a total of 32 participants from many physical sample communities. These include Collection Manager in DisciplineNronae, archeology, botany, geology, ichthyology, entomology, Herpatology and pelionutology.

Each participant answered questions about how they use access, organized, stored and used data from their collection in an attempt to prepare their materials for AI. While human subjects need to be agreed to study, most species do not. Therefore, an AI can collect and analyze data from inhuman physical collection without privacy or consent concerns.

We found that Collection managers of different fields and institute have a lot of different practices, when it comes to preparing its physical collection for AI. Our results show that Matadata records to standardize the types of managers and in the collection they can make items more accessible and usable in ways to store it.

Additional research projects such as our study may help collection managers to build infrastructure that they will need to prepare their data machine. Human expertise can help notify AI devices that create new discoveries based on the old treasures in the museum collection.

This article has been reinstated Conversation Under a Creative Commons License. read the Original article,