Carlos Bobed's Home Page

Latest BsC Thesis defended so far …

admin — Wed, 18 Feb 2026 09:23:34 +0000

The latest studed I have supervised (Javier Cuesta) defended his thesis last december. The line initiated in his thesis titled “Development of a semantic search system based on natural language processing and clustering of contextual vectors” is going to be continued and we are already preparing a paper version for an international conference. I want to ack and thank Samuel García as external advisor, whose help, despite not being listed as official advisor, has been really useful in the development of this project (and, of course, is going to be working with us in the continuation … ).

What a batch of BSc Thesis have been defended!

admin — Wed, 03 Sep 2025 07:42:34 +0000

During the last defense period, four students I was supervising defended their BSc thesis. Alain Villagrasa presented his thesis “Malware recognition based on contextual analysis of execution traces”, Adrián Domínguez presented “Clustering of transactional data with unsupervised learning techniques and large language models”, and Jan Carlos Franco presented “Improvements in entity retrieval from knowledge graphs using embeddings and natural language processing“. The three of these works are going to be continued to prepare a paper version for different international venues.

Last but not least, Gonzalo Valero presented his work done during an internship about an intelligent agent architecture to help data scientists in their daily life titled “Intelligent Agent System for Data Science Workflow automation”, which I externally supervised.

All of the thesis defended in this batch are really interesting works with a lot of potential applications (and interesting future work … ). A great pleasure to have worked with all these students! Great job!

Information-Aware Entity Indexing in Knowledge Graphs to Enable Semantic Search accepted at ESWC’25 (presented by Samuel García)

admin — Mon, 16 Jun 2025 15:46:46 +0000

In the context of Semantic Search and Entity Retrieval, traditional approaches prepare an adhoc schema to index the contents of a Knowledge Graph and/or learn ranking functions out from annotated query sets (difficult and expensive to obtain, completely attached to a particular KG). To get rid of these limitations, we have devised a non-supervised approach based on information quantity measures which builds a template out from the KG to direclty use a regular indexing engine (e.g., ElasticSearch) and get equivalent results to supervised approaches regardless the domain. The results and the developed project were accepted at ESWC’25, held this year in Portoroz (Slovenia). #ESWC25

Congratulations to Samuel García! The paper can be found here.

Yet another two new engineers released in the wild!

admin — Fri, 14 Mar 2025 15:42:56 +0000

During the last exam period, both Clara Cerdán and Daniel Carrizo defended their BSc thesis. Clara defended her work done during an internship within Hiberus, titled “Integration of CCSV libraries for document optimization and security management in AMD projects Aragón in an enterprise”, which I externally supervised. On the other hand, Daniel presented his thesis titled “Design and development of an ERP for non-profit associations″, where he addressed the needs he detected thanks to his voluteering at a non-profit association (the results are public and available in GitHub, https://github.com/DaniUnizar02/TFG ).

Great work done by both students !! Godspeed!

Another BSc Thesis successfully defended

admin — Thu, 03 Oct 2024 10:01:24 +0000

A week ago, Carlota Quintana defended her BSc thesis titled “Semantic visualization and exploration of ontologies: OntView 2.0″ (co-supervised by Eduardo Mena), getting an A mark (9.2) in her presentation. Great work which will be continued and released as part … Continue reading →

Ontology-Driven Automated Reasoning About Property Crimes accepted at BISE (journal)

admin — Mon, 02 Sep 2024 08:51:52 +0000

Again, within the field of Ángel’s ongoing PhD. Thesis, we have been collaborating with Francisco Navarrete, Antonio Vallecillo and Manuel Atencia (University of Málaga) to build a system that handles automatically police reports about property crimes assessing possible offenses even in the absence of important information. Built on NLP, ontologies, DL reasoners and KGs, we build a subjective assessment which has been evaluated with the collaboration of several judges. The results have been accepted and published at Business & Information Systems Engineering (BISE), a very strong journal on Information Systems. It can be found at (open access) https://doi.org/10.1007/s12599-024-00886-3

Great work!

Two new engineers released in the wild!

admin — Fri, 12 Jul 2024 10:00:41 +0000

Yesterday, two students whose BSc thesis I was supervising presented their work and have officially taken their degrees. On the one hand, Paula Ezpeleta presented her work titled “Text analysis for citation and reference validation” (co-supervised by Lucía Pitarch), and on the other one, Álvaro Pérez presented his work titled “HLD: Exploring different AI strategies for card games”. Both of them did it successfully, of course!

Congratulations to both, and godspeed!

Language-Model Based Informed Partition of Databases to Speed Up Pattern Mining accepted (and already presented) at SIGMOD’24

admin — Fri, 12 Jul 2024 09:53:58 +0000

As a continuation of our previous work on mining graph structures via propositionalization, we needed to explore new methods to speed up pattern mining. Given the similarities between transactions in the context of frequent itemset mining and sentences (both can be seen as sequences of tokens), we explored the possibility to partition the mined databases in an informative way by exploiting the distributional semantics that is captured by language models. The results supported our hypothesis and were accepted at SIGMOD’24, held this year in Santiago de Chile. #SIGMOD24

Congratulations to my co-authors, Jordi Bernad and Pierre Maillot!

Building MUSCLE, a Dataset for MUltilingual Semantic Classification of Links between Entities accepted at LREC-COLING 2024

admin — Tue, 27 Feb 2024 10:00:07 +0000

Exploiting Wikidata as source of structured knowledge, we have done a paradigm shift regarding building linguistic datasets building on semantic maps theory. In this case, we propose a fresh knowledge-based multilingual dataset for classification of lexico-semantical relationships (LRC) between words. Apart from providing the dataset, we thoroughly test it against different possible cases of memorization, and provide a baseline using the SoA for LRC. Our results are included in “Building MUSCLE, a Dataset for MUltilingual Semantic Classification of Links between Entities” which has been accepted at LREC-COLING 2024. We will soon release the dataset and the camera ready #COLING24

Lucía Pitarch, David Avián, Jorge Gracia, and Jordi Bernad … Great Work!!

SJORS: A Semantic Recommender System for Journalists accepted at BISE (journal)

admin — Tue, 26 Dec 2023 10:02:48 +0000

As part of Ángel’s ongoing PhD. Thesis, Ángel Garrido, Sole Pera and myself have been working on a different approach to provide journalists with timely news recommendations to help them writing new articles, instead of focusing on news readers as the main approaches in the field do. The results have been accepted and published at Business & Information Systems Engineering (BISE), a very strong journal on Information Systems. It can be found at (open access) https://doi.org/10.1007/s12599-023-00843-6

Great work!