2026-04-19T22:14:49Zhttps://recercat.cat/oai/requestoai:recercat.cat:2117/4606492026-04-17T04:01:50Zcom_2072_1033col_2072_452951
RECERCAT
author
Fernández Coronado, Alba
2026-04-17T04:01:50Z
2026-04-17T04:01:50Z
2026-01-27
https://hdl.handle.net/2117/460649
This thesis presents a series of improvements to the semantic search system deployed on the mango.com e-commerce platform, with the goal of enhancing retrieval accuracy, robustness, and relevance across multiple languages and query types. The work focuses on addressing key limitations of the existing semantic component within a hybrid lexical¿semantic search architecture. The main contributions include the training of a multilingual BERT uncased model to improve robustness to capitalization and diacritics, as well as fine-tuning on fashion-specific data to enhance domain understanding. In addition, limitations in handling attribute-only queries are addressed through the use of intelligent image cropping and a weighted fusion strategy that combines image embeddings with short textual metadata. Furthermore, the CLIP image encoder is fine-tuned to generate semantically richer and more discriminative visual representations, leading to higher similarity scores and improved ranking stability. Experimental results, evaluated on a manually annotated multilingual dataset, demonstrate consistent improvements across locales, increased semantic similarity scores, and more stable ranking performance, without introducing catastrophic forgetting or degrading standard query retrieval. The integration of these enhancements into the full production search pipeline provides a robust foundation for future improvements, including tighter alignment between multilingual text embeddings and the fine-tuned image encoder.
Restricted access - confidentiality agreement
Àrees temàtiques de la UPC::Informàtica::Intel·ligència artificial::Aprenentatge automàtic
Àrees temàtiques de la UPC::Economia i organització d'empreses::Comerç electrònic
Deep learning
Semantics--Data processing
Computer vision
Electronic commerce
Cerca semàntica
Cerca en comerç electrònic
Recuperació multilingüe
Sistemes de cerca híbrids
Cerca multimodal
Models visió-llenguatge
CLIP
BERT
Recuperació de moda
Adaptació al domini
Embeddings imatge-text
Semantic search
E-commerce search
Multilingual retrieval
Hybrid search systems
Multimodal search
Vision-language models
Image-text embeddings
Image-text embeddings
Image-text embeddings
Aprenentatge profund
Semàntica--Informàtica
Visió per ordinador
Comerç electrònic
Improvement and integration of a deep learning-based semantic model into a hybrid lexical-semantic search engine
Master thesis