

Research
About us
The Applied Research group at eBay Israel R&D center explores and provides solutions for deeply complex problems by conducting research in areas of Data Science, Machine Learning, NLP, Recommendation Systems, Information Retrieval, and Computer Vision. We work closely with the engineering and product teams to bring state of the art technology to production environments. The group focuses on a variety of topics, all with a deep impact on eBay’s core business. For example, using ML to make eBay’s data more structured, generating high quality content for eBay’s catalog, offering intelligent pricing suggestions, and guiding sellers on how to bid on eBay’s promoted listings.
Our researchers are owning the full lifecycle of their products: from understanding the business problem, through conducting the research and finding the right ML approach, to production deployment. Think, full-stack researcher / data scientist. Moreover, we see great importance in publishing our work in scientific top-tier conferences (WWW, WSDM, SIGIR, CIKM, EMNLP, IJCAI, SIGMOD, VLDB and others).
Our team closely collaborates with leading Israeli universities:
The Technion, Tel Aviv University, The Hebrew University of Jerusalem,
and Ben-Gurion University of the Negev.




Meet the team

Research Scientist

Research Scientist

Research Scientist

Research Scientist

Team lead

Research coordinator

Team Lead

Intern scientist

Research Scientist

Research Scientist

Research Scientist

Research Scientist

Research Scientist

Research Scientist

Research Scientist

Team Lead

Researchers

Research Scientist

Research Scientist

Research Scientist

Research Scientist

Research Scientist

Research Scientist

Team Lead

Team Lead
Publications
Demonstrating TabEE: Tabular Embedding Explanations [link]
Copul, R., Frost, N., Milo, T., & Razmadze, K.
Proceedings of the VLDB Endowment, 17(12), 4285-4288
CaFA: Cost-aware, Feasible Attacks With Database Constraints Against Neural Tabular Classifiers [link]
Ben-Tov, Matan, Daniel Deutch, Nave Frost, and Mahmood Sharif
IEEE Symposium on Security and Privacy (SP) 2024, pp. 227-227. IEEE Computer Society, 2024
Predicting Fact Contributions from Query Logs with Machine Learning. [link]
Arad, Dana, Daniel Deutch, and Nave Frost
EDBT, pp. 704-716. 2024
Evaluating Anomaly Explanations Using Ground Truth [link]
Antwarg Friedman, Liat, Chen Galed, Lior Rokach, and Bracha Shapira
AI 5, no. 4 (2024): 2375-2392
Personalized Ordering of Recommendation-Modules on an E-Commerce Homepage [link]
Roitman, Haggai, Alex Nus, and Yotam Eshel
Companion Proceedings of the ACM on Web Conference 2024, pp. 879-882. 2024
Partially Interpretable Models with Guarantees on Coverage and Accuracy [link]
Frost, Nave, Zachary Lipton, Yishay Mansour, and Michal Moshkovitz
International Conference on Algorithmic Learning Theory, pp. 590-613. PMLR, 2024
TabEE: Tabular Embeddings Explanations [link]
Copul, Roni, Nave Frost, Tova Milo, and Kathy Razmadze
Proceedings of the ACM on Management of Data 2, no. 1 (2024): 1-26
Banzhaf Values for Facts in Query Answering [link]
Abramovich, Omer, Daniel Deutch, Nave Frost, Ahmet Kara, and Dan Olteanu
Proceedings of the ACM on Management of Data 2, no. 3 (2024): 1-26
Unsupervised Search Algorithm Configuration using Query Performance Prediction [link]
Roitman, Haggai
Companion Proceedings of the ACM on Web Conference 2024, pp. 658-661. 2024
Pricing the Nearly Known-When Semantic Similarity is Just not Enough [link]
Fuchs, Gilad, Pavel Petrov, Ido Ben-Shaul, Matan Mandelbrod, Oded Zinman, Dmitry Basin, and Vadim Arshavsky.
eCom@ SIGIR. 2023
Exploring the Approximation Capabilities of Multiplicative Neural Networks for Smooth Functions
Ben-Shaul, Ido, Tomer Galanti, and Shai Dekel
Transactions on Machine Learning Research (2023).
the Nearly Known – When Semantic Similarity is Just not Enough
Gilad Fuchs, Pavel Petrov, Ido Ben-Shaul, Matan Mandelbrod, Oded Zinman, Dmitry Basin and Vadim Arshavski.Pricing
ECOM23@SIGIR23
Comparative Generalization Bounds for Deep Neural Networks
Galanti, Tomer, Liane Galanti, and Ido Ben-Shaul
Transactions on Machine Learning Research (2023)
Is It Out Yet? Automatic Future Product Releases Extraction from Web Data
Fuchs, Gilad, Ido Ben-Shaul, and Matan Mandelbrod
Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing: Industry Track. 2022
The Tip of the Buyer: Extracting Product Tips from Reviews [link]
S Hirsch, S Novgorodov, I Guy, A Nus
ACM Transactions on Internet Technology, 2023
Lot or not: Identifying multi-quantity offerings in e-commerce [link]
G Lavee, I Guy
ACM Transactions on Internet Technology, 2023
Promoting Tail Item Recommendations in E-Commerce [link]
T Didi, I Guy, A Livne, A Dagan, L Rokach, B Shapira
Proceedings of the 31st ACM Conference on User Modeling, Adaptation and Personalization 2023
Shop by image: characterising visual search in e-commerce [link]
A Dagan, I Guy,S Novgorodov
Information Retrieval Journal 26 (1), 2, 2023
Reverse Engineering Self-Supervised Learning
Ben-Shaul, Ido, et al
arXiv e-prints (2023): arXiv-2305
LearnShapley: Learning to Predict Rankings of Facts Contribution Based on Query Logs
Arad, D., Deutch, D. and Frost, N.
Proceedings of the 31st ACM International Conference on Information & Knowledge Management (pp. 4788-4792)
Product Titles-to-Attributes As a Text-to-Text Task [link]
Gilad Fuchs, Yoni Acriche
Proceedings of the Fifth Workshop on e-Commerce and NLP (ECNLP 5). 2022
Automatic image selection for online product catalogs [link]
Arnon Dagan,Ido Guy,Alexander Nus,Raphael Bryl,Noa Shimoni Barzilai, Avinoam OmerY, an Radovilsky, Einav Itamar, Gadi Mikles
US Patent 11,699,101, 2023
Query modality recommendation for e-commerce search [link]
A Dagan, S Novgorodov, I Guy
US Patent App. 17/569,876, 2023
System, method, and medium to select a product title [link]
A Dagan, A Zhicharevich
US Patent 11,580,589 2023
Visual quality performance predictors [link]
I Guy, S Novgorodov, A Dagan
US Patent App. 17/448,083, 2023
Is It Out Yet? Automatic Future Product Releases Extraction from Web Data [link]
Fuchs, Gilad, Ido Ben-Shaul, and Matan Mandelbrod.
Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing: Industry Track. 2022
LearnShapley: Learning to Predict Rankings of Facts Contribution Based on Query Logs [link]
Arad, D., Deutch, D. and Frost, N., 2022, October.
Proceedings of the 31st ACM International Conference on Information & Knowledge Management (pp. 4788-4792).
Product Titles-to-Attributes As a Text-to-Text Task [link]
Gilad Fuchs, Yoni Acriche
Proceedings of the Fifth Workshop on e-Commerce and NLP (ECNLP 5). 2022
An Image is Worth a Thousand Terms? Analysis of Visual E-Commerce Search
SIGIR 2021 – July 2021, Montreal, Canada
PreSizE: Predicting Size in E-Commerce using TransformersA
SIGIR 2021 – July 2021, Montreal, Canada
Automatic Form Filling with Form-Bert
Improving Constrained Search Results By Data Melioration
ConCaT: Construction of Category Trees from Search Queries in ECommerce
Uri Avron, Shay Gershtein, Ido Guy, Tova Milo, Slava Novgorodov
ICDE’ 21 April 2021 , Chania, Greece
Category Recognition in E-Commerce using Sequence-to-Sequence Hierarchical Classification
Generating Tips from Product Reviews
Sharon Hirsch, Slava Novgorodov, Ido Guy, Alexander Nus
WSDM’21 – March 2021, Jerusalem, Israel
Event-driven Query Expansion
Descriptions from the Customers: Comparative Analysis of Review-based Product Description Generation Methods
Intent-Driven Similarity in E-Commerce Listings
E-Commerce Dispute Resolution Prediction
tdGraphEmbed: Temporal Dynamic Graph-Level Embedding
CONCIERGE: Improving Constrained Search Results by Data Melioration
Ido Guy, Tova Milo, Slava Novgorodov, Brit Youngmann
VLDB’20 – August 2020, Tokyo, Japan
Product Bundle Identification using Semi-Supervised Learning
Hen Tzaban, Ido Guy, Asnat Greenstein-Messica, Arnon Dagan, Lior Rokach, Bracha Shapira
Query Reformulation in E-Commerce Search
Sharon Hirsch, Ido Guy, Alexander Nus, Arnon Dagan, Oren Kurland
SIGIR’20 – July 2020, Xi’an, China
MC3: A System for Minimization of Classifier Construction Cost
Minimization of Classifier Construction Cost for Search Queries