 As a data analyst, I have a few main tasks. One of the main tasks is name matching and other details to aid matching of identities. We get a list of names with other details and we want to see if we have these people in our database. We have developed a tool that helps us link people in events with each other. Here we can see a map where events have occurred. If we have data on people who went missing, at the same time as the event occurred, we can combine these two parts of information and it may help us find the missing person. We can filter by specific dates, both for missing persons and for events, and really see what happened, what is the circumstances of this appearance for this group of people and how they are linked. In the context of the international armed conflict between Ukraine and the Russian Federation, we have two main challenges. One is the amount of information that comes from different sources, the families, the authorities, open sources, and the second one is that the hostilities are still happening. So the caselot is dynamic. Each individual case is changing a lot and the groups of cases as well. We use the algorithms for two main reasons. The first, it's much faster. It takes a human expert about three days to process 1000 names to find possible matches. With the algorithm, it takes 30 minutes. The second reason is accuracy. The algorithm can do much more exact results than a human because it can do many more calculations and many more comparisons of the string of the family name, the first name, how we can rewrite this name in different ways and compare that. So the accuracy is very important. We don't want to miss any match because of a transliteration rule. If we have a name that in Ukrainian is Ole and Russian is Oleg, it's written the same way. But if we take the Ukrainian transliteration rules, as it's pronounced Ole, it's transliterated to Ole and in Russian transliteration rules, official rules, it will be transliterated to different letter to G, Oleg. We have to be innovative, to be faster and more efficient. In the traditional ICRC way of working, someone will just check the list one by one in the database and that would be it. But because we receive lists of sometimes hundreds of people, it would take a lot of time for someone to go through all the names one by one and checking any correspondence. With that analyst, we save a lot of time. There's a new list that just came in. I'm going to run it through our algorithm, which will help us find potential matches precisely and efficiently. Once I have a list of potential matches, my colleagues will go through it one by one and make the final decision on whether there is such a match. At the end of the day, it is a human decision.