Tätigkeitsgebiet: Softwaretechnik, Web Engineering, Künstliche Intelligenz
Büro: 04-1.326 (Parkstadt Mülheim)
Telefon: +49 208 88254-817
E-Mail:
Fatih Gedikli ist Informatiker und Professor für Künstliche Intelligenz und Big Data am Institut für Informatik der Hochschule Ruhr West.
Prof. Dr. Fatih Gedikli verfügt über weitreichende Expertise in Softwaretechnik und Web Engineering. Seine Forschung konzentriert sich auf die Entwicklung von Empfehlungssystemen, die in unserer informationsüberfluteten Welt eine wesentliche Entscheidungshilfe für den Menschen darstellen. Durch den Einsatz maschineller Lernverfahren verbesserte er die Genauigkeit dieser Systeme und stellte die Wichtigkeit von Erklärungsschnittstellen für das Verständnis und die Akzeptanz seitens der Anwendenden heraus.
Als begeisterter Verfechter für Big-Data-Analysen widmet sich Prof. Dr. Fatih Gedikli intensiv der Entwicklung innovativer Lösungen und Werkzeuge, die das automatische Erkennen und Interpretieren komplexer Muster in umfangreichen Datensätzen ermöglichen. Ein zentraler Aspekt seiner Forschung ist der Einsatz von Natural Language Processing (NLP)-Techniken zur detaillierten Analyse von Nachrichtentexten mit dem Ziel, tiefergehende Einblicke zu gewinnen und wertvolle Informationen effizient aus der Fülle an Textdaten zu extrahieren.
Forschung
- Recommender Systems
- News Recommendation and Analytics
- Natural Language Processing (NLP)
- Artificial Intelligence: Applications and Theory
Lehre
- Grundlagen der Informatik und Programmierung
- Softwaretechnik
- Webtechnologien und mobile Anwendungen
- Empfehlungssysteme
- Mobile Computing
Publikationen
Aktuelle und ausgewählte Publikationen
- Recommender Systems, Semantics-based
F. Gedikli, D. Jannach
Book chapter in: Encyclopedia of Social Network Analysis and Mining (ESNAM). Alhajj, R. and Rokne, J. (Eds.), Springer, forthcoming 2026 (preprint) - The Importance of Recommender Systems: A Key Technology of the World Wide Web
F. Gedikli
Expert blog post for Frontnow GmbH, September 22, 2023 (link) - Explaining BERT model decisions for near-duplicate news article detection based on named entity recognition
A. Stockem Novo, F. Gedikli
IEEE 17th International Conference on Semantic Computing (ICSC), 278-281, 2023 - Named Entities as Key Features for Detecting Semantically Similar News Articles
A. Stockem Novo, F. Gedikli
International Journal of Semantic Computing (IJSC) 17 (4), 2023 - NewsRecs: A Mobile App Framework for Conducting and Evaluating Online Experiments for News Recommender Systems
N. Janzen, F. Gedikli
15th International Conference on Agents and Artificial Intelligence (ICAART), 267-275, 2023 - KI-basierter Wissensgraph: Veranschaulichung von Menschenrechtsverletzungen und Umweltsünden in globalen Lieferketten
F. Gedikli
Embedded Software Engineering, Online-Fachmagazin für Embedded Software Entwicklung und professionelles Software Engineering, September 29, 2022 (link) - Semi-Automated Identification of News Story Chains: A New Dataset and Entity-based Labeling Method
F. Gedikli, A. Stockem Novo, D. Jannach
Workshop on News Recommendation and Analytics (INRA) @ RecSys, 29-42, 2021 - How should I explain? A comparison of different explanation types for recommender systems
F. Gedikli, D. Jannach, M. Ge
International Journal of Human-Computer Studies (IJHCS) 72 (4), 367-382, 2014 - Improving recommendation accuracy based on item-specific tag preferences
F. Gedikli, D. Jannach
ACM Transactions on Intelligent Systems and Technology (TIST) 4 (1), 1-19, 2013 - What recommenders recommend–an analysis of accuracy, popularity, and sales diversity effects
D. Jannach, L. Lerche, F. Gedikli, G. Bonnin
User Modeling, Adaptation, and Personalization (UMAP): 21th International Conference, 25-37, 2013 - Accuracy improvements for multi-criteria recommender systems
D. Jannach, Z. Karakaya, F. Gedikli
Proceedings of the 13th ACM conference on electronic commerce (EC), 674-689, 2012
Externe Publikationslisten
Wissenschaftliche Mitarbeiter
Noah Janzen
Büro: 04-1.267 (Parkstadt Mülheim)
AUSGEWÄHLTE PUBLIKATIONEN
2012
11.Ge, Mouzhi; Jannach, Dietmar; Gedikli, Fatih; Hepp, Martin
Effects of the Placement of Diverse Items in Recommendation Lists Proceedings Article
In: ICEIS (2), S. 201–208, 2012.
BibTeX | Schlagwörter:
@inproceedings{Ge2012a,
title = {Effects of the Placement of Diverse Items in Recommendation Lists},
author = {Mouzhi Ge and Dietmar Jannach and Fatih Gedikli and Martin Hepp},
year = {2012},
date = {2012-01-01},
booktitle = {ICEIS (2)},
pages = {201--208},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
2011
10.Kubatz, Marius; Gedikli, Fatih; Jannach, Dietmar
LocalRank-Neighborhood-based, fast computation of tag recommendations Proceedings Article
In: International Conference on Electronic Commerce and Web Technologies, S. 258–269, 2011.
BibTeX | Schlagwörter:
@inproceedings{Kubatz2011,
title = {LocalRank-Neighborhood-based, fast computation of tag recommendations},
author = {Marius Kubatz and Fatih Gedikli and Dietmar Jannach},
year = {2011},
date = {2011-01-01},
booktitle = {International Conference on Electronic Commerce and Web Technologies},
pages = {258--269},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
9.Ge, Mouzhi; Gedikli, Fatih; Jannach, Dietmar
Placing High-Diversity Items in Top-N Recommendation Lists Proceedings Article
In: ITWP@ IJCAI, 2011.
BibTeX | Schlagwörter:
@inproceedings{Ge2011,
title = {Placing High-Diversity Items in Top-N Recommendation Lists},
author = {Mouzhi Ge and Fatih Gedikli and Dietmar Jannach},
year = {2011},
date = {2011-01-01},
booktitle = {ITWP@ IJCAI},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
8.Gedikli, Fatih; Bagdat, Faruk; Ge, Mouzhi; Jannach, Dietmar
RF-REC: Fast and accurate computation of recommendations based on rating frequencies Proceedings Article
In: 2011 IEEE 13th Conference on Commerce and Enterprise Computing, S. 50–57, 2011.
BibTeX | Schlagwörter:
@inproceedings{Gedikli2011,
title = {RF-REC: Fast and accurate computation of recommendations based on rating frequencies},
author = {Fatih Gedikli and Faruk Bagdat and Mouzhi Ge and Dietmar Jannach},
year = {2011},
date = {2011-01-01},
booktitle = {2011 IEEE 13th Conference on Commerce and Enterprise Computing},
pages = {50--57},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
7.Gedikli, Fatih; Ge, Mouzhi; Jannach, Dietmar
Explaining online recommendations using personalized tag clouds Artikel
In: i-com Zeitschrift für interaktive und kooperative Medien, Bd. 10, Nr. 1, S. 3–10, 2011.
BibTeX | Schlagwörter:
@article{Gedikli2011a,
title = {Explaining online recommendations using personalized tag clouds},
author = {Fatih Gedikli and Mouzhi Ge and Dietmar Jannach},
year = {2011},
date = {2011-01-01},
journal = {i-com Zeitschrift für interaktive und kooperative Medien},
volume = {10},
number = {1},
pages = {3--10},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
6.Gedikli, Fatih; Ge, Mouzhi; Jannach, Dietmar
Understanding recommendations by reading the clouds Proceedings Article
In: International Conference on Electronic Commerce and Web Technologies, S. 196–208, 2011.
BibTeX | Schlagwörter:
@inproceedings{Gedikli2011b,
title = {Understanding recommendations by reading the clouds},
author = {Fatih Gedikli and Mouzhi Ge and Dietmar Jannach},
year = {2011},
date = {2011-01-01},
booktitle = {International Conference on Electronic Commerce and Web Technologies},
pages = {196--208},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
2010
5.Gedikli, Fatih; Jannach, Dietmar
Neighborhood-restricted mining and weighted application of association rules for recommenders Proceedings Article
In: International Conference on Web Information Systems Engineering, S. 157–165, 2010.
BibTeX | Schlagwörter:
@inproceedings{Gedikli2010,
title = {Neighborhood-restricted mining and weighted application of association rules for recommenders},
author = {Fatih Gedikli and Dietmar Jannach},
year = {2010},
date = {2010-01-01},
booktitle = {International Conference on Web Information Systems Engineering},
pages = {157--165},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
4.Gedikli, Fatih; Jannach, Dietmar
Rating items by rating tags Proceedings Article
In: Proceedings of the 2010 Workshop on Recommender Systems and the Social Web at ACM RecSys, S. 25–32, 2010.
BibTeX | Schlagwörter:
@inproceedings{Gedikli2010a,
title = {Rating items by rating tags},
author = {Fatih Gedikli and Dietmar Jannach},
year = {2010},
date = {2010-01-01},
booktitle = {Proceedings of the 2010 Workshop on Recommender Systems and the Social Web at ACM RecSys},
pages = {25--32},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
3.Gedikli, Fatih; Jannach, Dietmar
Recommending based on rating frequencies: Accurate enough? Proceedings Article
In: Proceedings of the 8th Workshop on Intelligent Techniques for Web Personalization & Recommender Systems at UMAP'10 (ITWP'10), S. 65–70, 2010.
BibTeX | Schlagwörter:
@inproceedings{Gedikli2010b,
title = {Recommending based on rating frequencies: Accurate enough?},
author = {Fatih Gedikli and Dietmar Jannach},
year = {2010},
date = {2010-01-01},
booktitle = {Proceedings of the 8th Workshop on Intelligent Techniques for Web Personalization & Recommender Systems at UMAP'10 (ITWP'10)},
pages = {65--70},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
0000