Konstantinos Bougiatiotis

I develop efficient methods for learning and reasoning over graphs, with applications in biomedical discovery and financial document analysis.

I am a Ph.D. candidate in Computer Science and a research associate at the Institute of Informatics and Telecommunications, NCSR “Demokritos”. I am part of the Document Intelligence Centre of Excellence (DICE), working with Dr. Georgios Paliouras.

My research spans multi-relational graph representations, tensor rank estimation, and the use of graph-based tools to support small language models.

Featured research

FastRank

Estimating tensor rank from spectral energy, without repeatedly computing tensor decompositions.

Explore the research →

Bag of Paths pipeline from a multi-relational graph through multi-hop matrices to node, edge, and graph features.

Prime Adjacency Matrices & Bag of Paths

Compact multi-relational graph representations and path features for node, edge, and graph prediction.

Explore the research →

All projects, datasets, and collaborations →

News

Oct 05, 2026	New publication in Artificial Intelligence in Medicine: our benchmark of deep learning and path analysis studies multi-type drug–drug interaction prediction on a disease-specific literature knowledge graph. Bag of Paths contributes interpretable path-based features to this biomedical application.
Aug 30, 2026	FastRank is now available in the AISTATS 2026 proceedings. Explore the method and code for fast tensor rank estimation.
Jul 14, 2026	Two preprints explore graph-based tools for small language models: molecular property prediction and knowledge graph grounding for reasoning. Both were presented at Causal NeSy @ ESWC 2026.

Selected Publications

FastRank: Fast Tensor Rank Approximation based on Spectral Energy

Konstantinos Bougiatiotis, and Georgios Paliouras

In Proceedings of the 29th International Conference on Artificial Intelligence and Statistics, May 2026

Bib Paper PDF Code Website

@inproceedings{bougiatiotis2026fastrank,
  title = {FastRank: Fast Tensor Rank Approximation based on Spectral Energy},
  author = {Bougiatiotis, Konstantinos and Paliouras, Georgios},
  booktitle = {Proceedings of the 29th International Conference on Artificial Intelligence and Statistics},
  series = {Proceedings of Machine Learning Research},
  volume = {300},
  pages = {1873--1881},
  year = {2026},
  month = may,
}

Identifying drug repurposing candidates for rare neuro-muscular disorders, using different AI methods on the literature knowledge graph

Fotios Papadimas, Stavroula Svolou, Konstantinos Bougiatiotis, and 4 more authors

JAMIA Open, Apr 2026

DOI Bib Code Data

@article{papadimas2026repurposing,
  title = {Identifying drug repurposing candidates for rare neuro-muscular disorders, using different AI methods on the literature knowledge graph},
  author = {Papadimas, Fotios and Svolou, Stavroula and Bougiatiotis, Konstantinos and Aisopos, Fotis and Krithara, Anastasia and Paliouras, Georgios and {SIMPATHIC Consortium}},
  journal = {JAMIA Open},
  volume = {9},
  number = {2},
  pages = {ooag050},
  year = {2026},
  month = apr,
  doi = {10.1093/jamiaopen/ooag050},
}

From Primes to Paths: Enabling Fast Multi-Relational Graph Analysis

Konstantinos Bougiatiotis, and Georgios Paliouras

Data & Knowledge Engineering, Jan 2026

DOI arXiv Bib Code Website

@article{bougiatiotis2024primespathsenablingfast,
  title = {From Primes to Paths: Enabling Fast Multi-Relational Graph Analysis},
  author = {Bougiatiotis, Konstantinos and Paliouras, Georgios},
  journal = {Data & Knowledge Engineering},
  volume = {163},
  pages = {102554},
  year = {2026},
  month = jan,
  doi = {10.1016/j.datak.2026.102554},
}

A benchmark study comparing deep learning and path analysis approaches for multi-class link prediction on a disease-specific literature knowledge graph

Marios Vottas, Konstantinos Bougiatiotis, Fotis Aisopos, and 1 more author

Artificial Intelligence in Medicine, 2026

Abs DOI Bib Paper Website

Identifying and understanding multi-type drug-drug interactions and their associated toxicities is crucial for preventing adverse outcomes, especially in vulnerable populations, such as cancer patients, who may already have compromised health statuses. This paper provides a benchmark study that compares various commonly-used Machine Learning-based approaches, such as graph embeddings and path analysis, aiming at predicting different types of drug-drug interaction links in a disease-specific knowledge graph, constructed from biomedical literature. These approaches are evaluated based on an external interaction dataset collected from Drugbank. We first introduce a simplified taxonomy of multi-type interactions and then compare the models’ performance over the different drug-drug interaction types. Overall, the graph embeddings approach outperforms the competition, while the path-based analysis allows for an interpretation of predictions, using the most important features of each path.
@article{VOTTAS2026103542, title = {A benchmark study comparing deep learning and path analysis approaches for multi-class link prediction on a disease-specific literature knowledge graph}, author = {Vottas, Marios and Bougiatiotis, Konstantinos and Aisopos, Fotis and Krithara, Anastasia}, journal = {Artificial Intelligence in Medicine}, pages = {103542}, year = {2026}, issn = {0933-3657}, doi = {10.1016/j.artmed.2026.103542}, url = {https://www.sciencedirect.com/science/article/pii/S0933365726001946}, keywords = {Knowledge graph, Link prediction, Graph embeddings, Drug-drug interactions}, }

A Multimodal Alignment-Based Anomaly Detection Method for Bankruptcy Prediction

Andreas Sideras, Konstantinos Bougiatiotis, Elias Zavitsanos, and 2 more authors

In Proceedings of the 6th ACM International Conference on AI in Finance, Nov 2025

DOI Bib Code Data

@inproceedings{sideras2025mabad,
  title = {A Multimodal Alignment-Based Anomaly Detection Method for Bankruptcy Prediction},
  author = {Sideras, Andreas and Bougiatiotis, Konstantinos and Zavitsanos, Elias and Paliouras, Georgios and Vouros, George},
  booktitle = {Proceedings of the 6th ACM International Conference on AI in Finance},
  pages = {53--61},
  year = {2025},
  month = nov,
  doi = {10.1145/3768292.3770380},
}