Research Scientist


Hey there!

Enrico Palumbo front picture

I'm Enrico Palumbo and I am research scientist (PhD). I currently work for Amazon Alexa in the Natural Language Understanding field.

My research interests lie at the intersection among Recommender Systems, Natural Language Understanding and Semantics on topics such as:

  • knowledge graph embeddings for recommender systems
  • sequence-aware recommender systems
  • machine learning approaches to entity matching
  • measures of entity relatedness for entity linking
  • sentiment analysis on social media data

In my free time, I play the guitar and sing in a band called Alephant.


We are builders!

new york skyline


  • Alexa: is a voice-first digital assistant created by Amazon. I work in the research and development of the Natural Language Understanding model.
  • 2019

  • Tinderbook: is a book recommendation application based on my knowledge graph embeddings algorithm entity2rec. It provides book suggestions given a single book that you like!
  • 2018

  • CEDUS: is a smart city project that generates a set of software components to effectively collect, integrate and visualize city data to support decision making processes. The team is in charge of the impact assessment and go-to-market activities, i.e. of the evaluation of the societal, economic and environmental impacts of the CEDUS technologies.
  • 2017

  • PasTime: creation and development of a recommender system that provides personalized suggestions of places and events in a city. We learn to recommend tourist paths by feeding sequences of tourist check-ins in Points-of-Interests to a Recurrent Neural Network.
  • 2016

  • 3cixty: creation and development of a machine learning based approach to automatically identify duplicates during the integration of places and events data from heterogeneous sources for the creation of city knowledge bases.
  • 2015 - 2017

  • Snowball: creation and development of a data-driven Decision Support System based on Multi-Criteria Decision Making to support crisis managers in dealing with cascading effects


Open source repos

hand covering the sun
  • entity2rec: provides top-N item recommendations from knowledge graphs.
  • entity2vec: creates property-specific embeddings of knowledge graphs applying node2vec on property-specific subgraphs
  • Stacked Threshold Based Entity Matching (STEM): performs entity matching by stacking a supervised learner on top of an ensemble of threshold-based classifiers. Stacking enhances both precision and recall at the same time.
  • SentiMe++: stacking on top of an ensemble of state-of-the-art sentiment classifiers to enhance performance.


Considerate la vostra semenza: fatti non foste a viver come bruti, ma per seguir virtute e canoscenza

Enrico Palumbo picture


  • Palumbo, Enrico, et al. Semantic Diversity for Natural Language Understanding Evaluation in Dialog Systems. Proceedings of the 28th International Conference on Computational Linguistics: Industry Track, 2020
  • Palumbo, Enrico. Knowledge Graph Embeddings for Recommender Systems (PhD thesis dissertation)
  • Palumbo, Enrico. Knowledge Graph Embeddings for Recommender Systems (PhD thesis dissertation - main contributions)
  • Palumbo, Enrico, et al. entity2rec: Property-specific knowledge graph embeddings for item recommendation. Expert Systems with Applications (2020): 113235.
  • 2019

  • Palumbo, Enrico, et al. Tinderbook: Fall in Love with Culture. European Semantic Web Conference. Springer, Cham, 2019.
  • 2018

  • Monti, Diego, et al. Sequeval: An offline evaluation framework for sequence-based recommender systems. Information 10.5 (2019): 174.
  • Monti, Diego, et al. Semantic Trails of City Explorations: How Do We Live a City. arXiv preprint arXiv:1812.04367 (2018).
  • Monti, Diego, et al. An ensemble approach of recurrent neural networks using pre-trained embeddings for playlist completion." Proceedings of the ACM Recommender Systems Challenge 2018. 2018. 1-6.
  • Giammusso Sara, Guerriero Mario, Lisena Pasquale, Palumbo Enrico & Troncy Raphaël. Predicting The Emotion of Playlists Using Track Lyrics. In 19th International Society for Music Information Retrieval Conference (ISMIR), Late-Breaking Demo Track, Paris, France, September 23-27, 2018.
  • Monti D., Palumbo E., Rizzo G., Lisena P., Troncy R., Fell M., Cabrio E., Morisio M. An Ensemble Approach of Recurrent Neural Networks using Pre-Trained Embeddings for Playlist Completion RecSys Challenge, Vancouver, 2018
  • Carducci G., Rizzo G., Monti D., Palumbo E., Morisio M. TwitPersonality: Computing Personality Traits from Tweets Using Word Embeddings and Supervised Learning. Information. 2018 May 18;9(5):127.
  • Caroleo B.R.D., Palumbo E., Osella M. Lotito A., Rizzo G., Ferro E.G., Attanasio A., Chiusano S.A., Zuccaro, G., Leone M., De Gregorio D. A Knowledge-Based Multi-Criteria Decision Support System Encompassing Cascading Effects For Disaster Management, International Journal of Information Technology & Decision Making, 2018
  • Palumbo E., Rizzo G., Troncy R., Baralis E., Osella M., Ferro E. Translational Models for Item Recommendation, ESWC2018, Heraklion, Greece, 2018
  • Palumbo E., Rizzo G., Troncy R., Baralis E., Osella M., Ferro E. Knowledge Graph Embeddings with node2vec for Item Recommendation, Poster & Demo sessions, ESWC2018, Heraklion, Greece, 2018 (best poster paper award)
  • Palumbo E., Rizzo G., Troncy R., Baralis E., Osella M., Ferro E. An Empirical Comparison of Knowledge Graph Embeddings for Item Recommendation , Deep Learning for Knowledge Graphs and Semantics Workshop at ESWC2018, Heraklion, Greece, 2018
  • Palumbo E., Rizzo G., Troncy R. STEM: Stacked Threshold-based Entity Matching for Knowledge Base Generation , Semantic Web Journal, Special Issue on Machine Learning for Knowledge Base Generation, 2018
  • 2017

  • Palumbo E., Rizzo G., Troncy R., Baralis E. Predicting Your Next Stop-over from Location-based Social Network Data with Recurrent Neural Networks , RecTour Workshop at RecSys2017, Como, Italy, 2017
  • Palumbo E., Rizzo G., Troncy R. entity2rec: Learning User-Item Relatedness from Knowledge Graphs for Top-N Item Recommendation. In 11th ACM Conference on Recommender Systems (RecSys) , Como, Italy, 2017
  • Troncy, Raphaël, et al. 3cixty: Building Comprehensive Knowledge Bases For City Exploration." Journal of Web Semantics (2017).
  • Palumbo E., Sygkounas E., Troncy R., Rizzo G. (2017) SentiME++ at SemEval-2017 Task 4A: Stacking State-of-the-Art Classifiers to Enhance Sentiment Classification. In International Workshop on Semantic Evaluation (SemEval), Vancouver, Canada
  • 2016

  • Palumbo, Enrico, Giuseppe Rizzo, and Raphael Troncy. "An Ensemble Approach to Financial Entity Matching for the FEIII 2016 Challenge." Proceedings of the Second International Workshop on Data Science for Macro-Modeling (DSMM'16). No. 14. ACM, 2016.
  • Palumbo, Enrico, et al. "ICDSST 2016 on Decision Support Systems Addressing Sustainability & Societal Challenges SnowBall DSS: an ensemble Multi-Criteria Decision Support System encompassing cascading effects for disaster management." (2016).
  • 2015

  • Palumbo, Enrico, and Walter Allasia. "Semantic Similarity Between Images: A Novel Approach Based on a Complex Network of Free Word Associations." International Conference on Similarity Search and Applications. Springer, Cham, 2015.
  • Allasia, Walter, and Enrico Palumbo. "A complex network model of semantic memory impairments." Multimedia & Expo Workshops (ICMEW), 2015 IEEE International Conference on. IEEE, 2015.
  • Contacts

    Get in touch!

    SWSA Dissertation Award