M2TRec: Metadata-aware Multi-task Transformer for Large-scale and Cold-start free Session-based Recommendations
Session-based recommender systems (SBRSs) have shown superior performance over
conventional methods. However, they show limited scalability on large-scale industrial …
conventional methods. However, they show limited scalability on large-scale industrial …
Transformers with multi-modal features and post-fusion context for e-commerce session-based recommendation
Session-based recommendation is an important task for e-commerce services, where a
large number of users browse anonymously or may have very distinct interests for different …
large number of users browse anonymously or may have very distinct interests for different …
Beyond ndcg: behavioral testing of recommender systems with reclist
As with most Machine Learning systems, recommender systems are typically evaluated
through performance metrics computed over held-out data points. However, real-world …
through performance metrics computed over held-out data points. However, real-world …
BERT goes shopping: Comparing distributional models for product representations
F Bianchi, B Yu, J Tagliabue - arXiv preprint arXiv:2012.09807, 2020 - arxiv.org
Word embeddings (eg, word2vec) have been applied successfully to eCommerce products
through~\textit {prod2vec}. Inspired by the recent performance improvements on several …
through~\textit {prod2vec}. Inspired by the recent performance improvements on several …
Reasonable scale machine learning with open-source metaflow
J Tagliabue, H Bowne-Anderson, V Tuulos… - arXiv preprint arXiv …, 2023 - arxiv.org
As Machine Learning (ML) gains adoption across industries and new use cases,
practitioners increasingly realize the challenges around effectively developing and iterating …
practitioners increasingly realize the challenges around effectively developing and iterating …
You do not need a bigger boat: Recommendations at reasonable scale in a (mostly) serverless and open stack
J Tagliabue - Proceedings of the 15th ACM Conference on …, 2021 - dl.acm.org
We argue that immature data pipelines are preventing a large portion of industry
practitioners from leveraging the latest research on recommender systems. We propose our …
practitioners from leveraging the latest research on recommender systems. We propose our …
EvalRS: a rounded evaluation of recommender systems
Much of the complexity of Recommender Systems (RSs) comes from the fact that they are
used as part of more complex applications and affect user experience through a varied …
used as part of more complex applications and affect user experience through a varied …
Building a serverless Data Lakehouse from spare parts
J Tagliabue, C Greco, L Bigon - arXiv preprint arXiv:2308.05368, 2023 - arxiv.org
The recently proposed Data Lakehouse architecture is built on open file formats,
performance, and first-class support for data transformation, BI and data science: while the …
performance, and first-class support for data transformation, BI and data science: while the …
Zero party data between hype and hope
A Polonioli - Frontiers in big Data, 2022 - frontiersin.org
Zero Party Data (ZPD) is a hot topic in the context of privacy-aware personalization, as the
exponential growth of consumer data collected by retailers has made safeguarding data …
exponential growth of consumer data collected by retailers has made safeguarding data …
Text classification for predicting multi-level product categories
In an online shopping platform, a detailed classification of the products facilitates user
navigation. It also helps online retailers keep track of the price fluctuations in a certain …
navigation. It also helps online retailers keep track of the price fluctuations in a certain …