Leman Akoglu
Leman Akoglu is a machine learning researcher at Carnegie Mellon University whose work focuses on graph mining and the detection of anomalies and fraud. After a computer science degree at Bilkent in 2007, she coauthored OddBall in 2010 to find unusual nodes in large social networks. She completed a CMU doctorate under Christos Faloutsos in 2012 and held an assistant professorship at Stony Brook before joining Heinz College in 2016. Her later papers introduced FoMo-0D in 2024 and OUTFORMER in 2026, foundation models for detecting outliers in tabular data. She directs the Data Analytics Techniques Algorithms (DATA) Lab at Heinz and became an Amazon Scholar in 2023.
- Now at
- Carnegie Mellon University and Amazon
- Works
- 194
- Citations
- 9,371
- h-index
- 44
- Badges
Publication numbers from OpenAlex, as of 27 October 2026.
Timeline
-
2007
A degree at Bilkent
She earns a bachelor's degree in computer science from Bilkent University in Turkey in 2007.
-
2010
The OddBall algorithm
With Mary McGlohon and Christos Faloutsos, she coauthors OddBall: Spotting Anomalies in Weighted Graphs in 2010. The paper presents fast algorithms to find unusual nodes in large social networks and wins the PAKDD 2010 best paper award.
-
2012
Doctorate and Stony Brook
She completes a PhD in Carnegie Mellon's Computer Science Department in 2012, advised by Christos Faloutsos. In 2012, she becomes an assistant professor in Stony Brook University's Department of Computer Science.
-
2013–2015
Army and NSF awards
She receives the US Army Research Office Young Investigator award in 2013. In 2015, she receives a National Science Foundation CAREER award.
-
2016
Joining Heinz College
In 2016, she joins Carnegie Mellon University's Heinz College of Information Systems and Public Policy as an assistant professor.
-
April 2021
The SDM/IBM research award
She receives the SDM/IBM Early Career Data Mining Research Award from the Society for Industrial and Applied Mathematics in April 2021.
-
August 2023
An Amazon Scholar role
In August 2023, she begins research and collaborations as an Amazon Scholar.
-
2024
FoMo-0D for outlier detection
In 2024, she coauthors FoMo-0D with Yuchen Shen and Haomin Wen. Pretrained on synthetic data, the foundation model identifies outliers in tabular data without labeled examples, further training or hyperparameter tuning for a new task.
-
2026
The OUTFORMER model
With Xueying Ding, Haomin Wen and Simon Klüttermann, she coauthors From Zero to Hero in 2026, introducing OUTFORMER. The model trains on synthetic labeled datasets and uses a new task's training data as context to identify outliers.
-
As of October 2026
Associate professor at Heinz
In October 2026, she is the Heinz College Dean's Associate Professor of Information Systems at Carnegie Mellon University and directs the Data Analytics Techniques Algorithms (DATA) Lab. Her research focuses on graph mining and anomaly, fraud and event detection.
Career
- NowCarnegie Mellon UniversityDean's Associate Professor of Information Systems, Director, Data Analytics Techniques Algorithms (DATA) Lab
- 2023–2026AmazonAmazon Scholar
- 2012–2016Stony Brook UniversityAssistant Professor
Education
- 2012Carnegie Mellon UniversityPhD
- 2007Bilkent UniversityBachelor's degree in computer science
Sources and credits
Photo: We have no published photo of Leman Akoglu yet. Send us a free-licensed photo.
- Leman Akoglu - Carnegie Mellon Database Group
- Leman Akoglu and Mary McGlohon win the ‘best paper’ award at PAKDD 2010 - Carnegie Mellon Database Group
- Christos Faloutsos awarded PAKDD’s “Most Influential Paper Award” - Amazon Science
- Leman Akoglu | Carnegie Mellon University Computer Science Department
- Leman Akoglu | Carnegie Mellon University's Heinz College
- Leman Akoglu - College of Engineering at Carnegie Mellon University
- Leman Akoglu, Heinz College
- [2409.05672] FoMo-0D: A Foundation Model for Zero-shot Tabular Outlier Detection
- [2602.03018] From Zero to Hero: Advancing Zero-Shot Foundation Models for Tabular Outlier Detection


