Software

My GitHub profile: https://github.com/fabiopetroni

GENRE (2021)

The GENRE (Generarive ENtity REtrieval) system as presented in Autoregressive Entity Retrieval implemented in pytorch.

GitHub repository: https://github.com/facebookresearch/GENRE

KILT (2020)

A Benchmark for Knowledge Intensive Language Tasks.

Website: kiltbenchmark.com

GitHub repository: https://github.com/facebookresearch/KILT

SAFEtorch (2020)

Pytorch implemenation of the SAFE neural network.

GitHub repository: https://github.com/facebookresearch/SAFEtorch

BLINK (2019)

BLINK is an Entity Linking python library that uses Wikipedia as the target knowledge base.

GitHub repository: https://github.com/facebookresearch/BLINK

LAMA (2019)

LAMA is a set of connectors to pre-trained language models.

GitHub repository: https://github.com/facebookresearch/LAMA

attr2vec (2018)

TensorFlow implementation of the attr2vec model, based on the following publication:

F. Petroni, V. Plachouras, T. Nugent and J. L. Leidner: "attr2vec: : Jointly Learning Word and Contextual Attribute Embeddings with Factorization Machines." In: Proceedings of the 16th Annual Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT), 2018.

If you use the application please cite the paper.

GitHub repository: https://github.com/thomsonreuters/attr2vec

libFM - BPR extension (2015)

An extended version of libFM that includes the support for the BPR optimization criterion, with a multi-thread training procedure.

If you use the application please cite the following paper:

F. Petroni, L. Del Corro and R. Gemulla (2015): "CORE: Context-Aware Open Relation Extraction with Factorization Machines". EMNLP, 2015.

GitHub repository: https://github.com/fabiopetroni/libfm_with_BPR.

Source code (C++) libFM_BPR.src.zip, GPL v3 license.

Manual manual_libFM_with_BPR_extension.pdf

CORE (2015)

The source code and data used in the following publication:

F. Petroni, L. Del Corro and R. Gemulla (2015): "CORE: Context-Aware Open Relation Extraction with Factorization Machines". EMNLP, 2015.

If you use the application please cite the paper.

GitHub repository: https://github.com/fabiopetroni/CORE.

Source code (Scala) CORE.src.zip, GPL v3 license.

VGP (2015)

A software package for one-pass Vertex-cut balanced Graph Partitioning, based on the publication:

F. Petroni, L. Querzoni, K. Daudjee, S. Kamali and G. Iacoboni: "HDRF: Efficient stream-based partitioning for power-law graphs". CIKM, 2015.

If you use the application please cite the paper.

GitHub repository: https://github.com/fabiopetroni/VGP.

Source code (Java) VGP.src.zip, GPL v3 license.

Executable jar VGP.jar.zip, GPL v3 license.

HDRF GraphLab PowerGraph 2.2 patch HDRF has been integrated in GraphLab PowerGraph!

CrawlFacebookPostTrees (2014)

A crawler for Facebook post trees through web browser automation.

If you use the application please cite the following paper:

Alessandro Bessi, Fabio Petroni, Michela Del Vicario, Fabiana Zollo, Aris Anagnostopoulos, Antonio Scala, Guido Caldarelli, Walter Quattrociocchi: "Viral Misinformation: The Role of Homophily and Polarization". In Proceedings of the 24th International Conference on World Wide Web Companion (WWW), 2015.

GitHub repository: https://github.com/fabiopetroni/CrawlFacebookPostTrees.

Source code (Java) CrawlFacebookPostTrees src.zip, GPL v3 license.

Executable jar CrawlFacebookPostTrees.jar.zip, GPL v3 license.

GASGD (2014)

A multithread simulator for distributed asynchronous matrix completion, based on the publication:

F. Petroni and L. Querzoni: "GASGD: stochastic gradient descent for distributed asynchronous matrix completion via graph partitioning". In: Proceedings of the 8th ACM Conference on Recommender systems (RecSys), 2014.

If you use the application please cite the paper.

GitHub repository: https://github.com/fabiopetroni/GASGD.

Source code (Java) GASGD.src.zip, GPL v3 license.

Executable jar GASGD.jar.zip, GPL v3 license.

LCBM (2014)

A fast and lightweight collaborative filtering algorithm for binary ratings.

If you use LCBM please cite the following paper:
F. Petroni, L. Querzoni, R. Beraldi, M. Paolucci: "LCBM: Statistics-Based Parallel Collaborative Filtering." In: Proceedings of the 17th International Conference on Business Information Systems (BIS), 2014.

GitHub repository: https://github.com/fabiopetroni/LCBM.

Hadoop MapReduce source code (Java) LCBM.src_mapreduce.zip, GPL v3 license.

Hadoop MapReduce executable jar LCBM_mapreduce.jar.zip, GPL v3 license.

Source code (Java) LCBM.src.zip, GPL v3 license.

Executable jar LCBM.jar.zip, GPL v3 license.

RecSysMF (2013)

Matrix Factorization for Recommender System with Parallel Stochastic Gradient Descent (with locks).

If you use RecSysMF please cite the following paper:
F. Petroni and L. Querzoni: "GASGD: stochastic gradient descent for distributed asynchronous matrix completion via graph partitioning." In: Proceedings of the 8th ACM Conference on Recommender systems (RecSys), 2014.

GitHub repository: https://github.com/fabiopetroni/RecSysMF.

Source code (Java) RecSysMF.src.zip, GPL v3 license.

Executable jar RecSysMF.jar.zip, GPL v3 license.

HSienaDemo (2012)

Part of my master thesis, a demo for the self-reconfiguring procedure of a pub/sub network presented in:

F. Petroni and L. Querzoni: "HSIENA: a hybrid publish/subscribe system". DESEC-LCCI 2012.

GitHub repository: https://github.com/fabiopetroni/HSienaDemo.

Source code (Java) hsiena_demo.src.zip, GPL v3 license.

Executable jar hsiena_demo.jar.zip, GPL v3 license.

Artist Driver (2010)

Part of my bachelor thesis, a web application constituted by a music manager linked to youtube videos (built with Google App Engine).

Unfortunately, the application is no longer maintained.