Master projects are research-oriented projects with 3-6 students for the duration of one semester (12 credits). Descriptions of all HPI master projects are here as pdf files.
- Sommer 2024: DBStrange: Exploring the Multiverse of Entity Resolution Datasets
- Winter 2023/24: Constrainted-based Schema Matching
Results published at EDBT 2025 - Winter 2022/23: CAST: Classifying Time Series Anomalies
- Summer 2022: Music Walks
- Winter 2021/22: Wikipedia Cleanup: Recognizing Stale Data
Results published at EDBT 2023 - Summer 2021: Generating Art using GANs
Results published at IJCAI 2022 - Winter 2020/21: Discovering Change Dependencies
Results published at EDBT 2022 - Summer 2020: Multimodal Analysis for Cultural Data
- Winter 2019/20: Mapping and Understanding the Evolution of Language
Results published at LWDA 2020 - Summer 2019: What’s in a Life? Representing Vitae in 1000 Dimensions
- Summer 2018: Vandalism Detection in Wikipedia Table Revisions
- Summer 2017: Hate Speech Detection
Results published at COLING Workshop TRAC 2018 - Winter 2016/17: Profiling Dynamic Data
Results published at EDBT 2019 - Winter 2015/16: Learning to Note: Intelligent Support for Document Annotation using Semi-Supervised Learning (with EPIC-group)
Results published at SMBM 2016 - Summer 2015: Approximate Data Profiling
Results published at CIKM 2016 and at BTW 2017 - Winter 2014/15: Metadata Trawling
- Summer 2014: Finding Relevant Tweets for News
Results published at WWW 2015 (Poster Track) - Winter 2013/14: Piggyback Profiling: Metadata for Query Results
Results published at LWDA 2018 - Winter 2013/14: Joint Data Profiling
Results published at EDBT 2016 - Winter 2012/13: FactScore: Global Relevance Scores for DBpedia Facts
Results published at DESWeb 2014 workshop - Winter 2011/12: LuSim: Similarity Search with Apache Lucene
Results published at DBRank 2012 workshop - Summer 2011: Duplikaterkennung auf GPUs
Results published at BTW 2013 (Runner up for Best Paper Award)