Daniel Kohlsdorf
Data Scientist, Machine Learning Enthusiast and Programmer


I am a data scientist working at the social network Xing in Hamburg working on Recommender Systems and Machine Learning. I was also involved in organising the ACM Recommende Systems Challange 2016 and also 2017. My main interests are in Machine Learning and Data Analysis, Artificial Intelligence, Algorithms and Data Structures, Web Systems, Recommender Systems, Web Mining, Wearable Computing and Pattern Recognition.

I received a PhD in computer science from the Georgia Institute of Technology. I also worked as a graduate research assistant in the Contextual Computing Group under Thad Starner. Before joining Georgia Tech I was a student assistant in the intelligent systems research group at University Bremen. I wrote my Diploma Thesis (Diploma is a German, similar degree to a Masters) in the same research group and worked there as a researcher for a short period of time. I also worked as a software developer for mobile applications on the android and iOS platform in the Neusta Software Development Group.

During my work at Xing, in Bremen and at Georgia Tech I contributed to several research projects including: The modeling of job postings using document embeddings, learning to rank for multiple recommender systems at Xing using gradient boosting, an Algorithm that helps designing and generating gesture recognizer that show few false positives in everyday life, a glove that teaches you how to play piano, a recognizer for dolphin whistles, a pattern discovery system for dolphin communication, an easy to use computer interface for elderly people and analysis software for typing errors on mobile devices. In a student project I also contributed to an intelligent system for vulnerability analysis in computer networks. Furthermore, I worked on failure prediction for robot actions in the domain of pancake flipping.

Problem domains I am currently interested in include: recommender systems, information retrieval, document embedding and latent space analysis, gesture recognition, activity discovery (in contrast to activity recognition), context sensing on wearable computers and mining of acoustic dolphin communication.