
Монгол Улсын Их Сургууль
11.9K posts

Монгол Улсын Их Сургууль
@num_edu
Дэлхийн стандартад нийцсэн, Монгол Улсын хөгжил дэвшлийн тулгуур болсон үндэсний загвар судалгааны их сургууль байна.







We are excited to introduce🚀𝘿𝙞𝙫𝙚𝙧𝙨𝙞𝙩𝙮𝙊𝙣𝙚 𝘿𝙖𝙩𝙖𝙨𝙚𝙩🚀, one of the largest and most geographically diverse real world datasets for modeling everyday life behavior with smartphone sensor data. 📜 Our paper “DiversityOne: A Multi-Country Smartphone Sensor Dataset for Everyday Life Behavior Modeling” that was recently accepted at the Proceeding of the ACM on Interactive, Mobile, Wearable, and Ubiquitous Technologies (IMWUT/ @ubicomp) 2025, provides a comprehensive description of this dataset. The dataset was collected as part of the @WeNetProject. 📜 Paper: arxiv.org/pdf/2502.03347 🌐 DiversityOne Website: datascientia.disi.unitn.it/projects/diver… 📊 Dataset Catalogue: datascientiafoundation.github.io/LivePeople/ 🚀🔍DiversityOne at a Glance ✅ Unlike prior behavior modeling datasets that focus on single-country cohorts, limited sample sizes, or preprocessed sensor features, DiversityOne provides raw, multimodal smartphone sensor data from 8 countries spanning both global north and south, with a large cohort of 782 young adults. ✅ Data collected from eight countries—China 🇨🇳, India 🇮🇳, Mongolia 🇲🇳, Italy 🇮🇹, Denmark 🇩🇰, the UK 🇬🇧, Paraguay 🇵🇾, and Mexico 🇲🇽—captures diverse lifestyles and cultural variations, enabling cross-country and cross-cultural studies. ✅ Data collection from 782 participants over four weeks generated 26 smartphone sensor time series (e.g., accelerometer, GPS, app usage, screen time) and over 350,000 self-reports, providing a rich behavioral context. This makes it one of the most comprehensive and openly available time series sensor datasets. Additionally, the dataset includes psycho-social surveys and demographic information, allowing for high-resolution insights into human behavior across different cultural contexts. ✅ The dataset also serves as a testbed for AI researchers working on time series modeling, domain adaptation / generalization across countries, pushing the boundaries of ubiquitous computing, HCI, responsible AI, and machine learning research. Excited to see what the community would do with this unique dataset! Università di Trento (@UniTrento), ETH Zürich (@ETH, @ETH_en), EPFL (@EPFL, @EPFL_en), Idiap Research Institute (@Idiap_ch), IT University Copenhagen (@ITUkbh), Amrita Vishwa Vidyapeetham (@AMRITAedu), Fondazione Bruno Kessler - FBK (@FBK_research), Jilin University, IPICyT (@IPICYTciencia), The London School of Economics and Political Science (@LSEnews), National University Of Mongolia (@num_edu), Universidad Católica "Nuestra Señora de la Asunción" (@unicatolicapy), Aalborg University (@aalborg_uni)










































