School of Engineering Department of Computer Science and Engineering 96 Department of Computer Science and Engineering AI meets Big Data: User Analytics and Personalized Recommendation Based on Location Data Supervisor: CHAN Gary Shueng Han / CSE Student: PRINTZ Maximilian / COGBM Course: UROP1100, Fall This report specifies my participation in designing a highly accurate classification-based proximity detection model, which is a winning approach in an international open challenge (TC4TL challenge). After introducing the structure of our proximity detection model, this report describes 3 major experiments conducted after the competition to further learn about what factors drive a successful proximity detection model. Brief descriptions of each experiment are given, and their results are explained through comprehensive charts and diagrams. Given the recent major challenges caused by the Covid-19 pandemic, developing an effective and efficient proximity detection model for contact tracing is critical. AI meets Big Data: User Analytics and Personalized Recommendation Based on Location Data Supervisor: CHAN Gary Shueng Han / CSE Student: WU Qi / COSC Course: UROP1100, Fall Among the many indoor localization methods, wifi-based indoor localization is one of the most popular ones. In this article, I will try to post my understanding of wifi-based indoor localization and some research data. Since I was not on campus, relevant activities are conducted off-lab. I spent the first few week on relevant papers and learning about artificial intelligence after discussing with Amy. Considering I have no background knowledge, I started with the simpler ones, ideally to review wideep’s experiments. The first part of the article is the understanding of the concepts related to wifi-based indoor localization, the second part is mainly some data, and the third part is my feeling of taking the UROP for the first time. AI meets Big Data: User Analytics and Personalized Recommendation Based on Location Data Supervisor: CHAN Gary Shueng Han / CSE Student: ZHANG Weiwen / DSCT Course: UROP1100, Fall Crowd counting has been a hot topic with enormous potential in real-world application. Multi-camera crowd counting is one of branches of crowds counting and its general approach is in field of computer vision and deep learning. This project is based on the approach that to train a CNN model to generate density maps with synthetic image datasets and count the number of people by integrating on a density map. And this report is mainly concentrating on proposing a novel method to generate synthetic images in Unity engine, in which we can add more variety into the dataset to imitate crowds in real-world scenario. The initial result and unity project can be seen on the website: https://github.com/KevynUtopia/multi-camera-crowdcounting_UROP.