Recent developments in information retrieval:preface

This paper summarizes the scientific work presented at the 32nd European Conference on Information Retrieval. It demonstrates that information retrieval (IR) as a research area continues to thrive with progress being made in three complementary sub-fields, namely IR theory and formal methods together with indexing and query representation issues, furthermore Web IR as a primary application area and finally research into evaluation methods and metrics. It is the combination of these areas that gives IR its solid scientific foundations. The paper also illustrates that significant progress has been made in other areas of IR. The keynote speakers addressed three such subject fields, social search engines using personalization and recommendation technologies, the renewed interest in applying natural language processing to IR, and multimedia IR as another fast-growing area.

Automatic indexing in information retrieval

DUE TO COPYRIGHT RESTRICTIONS ONLY AVAILABLE FOR CONSULTATION AT ASTON UNIVERSITY LIBRARY AND INFORMATION SERVICES WITH PRIOR ARRANGEMENT

Ontological Approach to Domain Knowledge Representation for Information Retrieval in Multiagent Systems

An ontological representation of buyer interests’ knowledge in process of e-commerce is proposed to use. It makes it more efficient to make a search of the most appropriate sellers via multiagent systems. An algorithm of a comparison of buyer ontology with one of e-shops (the taxonomies) and an e-commerce multiagent system are realised using ontology of information retrieval in distributed environment.

Survey of Existing Services in the Mathematical Digital Libraries and Repositories in the EuDML Project

This paper presents a survey of the existing services provided by the digital libraries and repositories on mathematics of the content provider partners in the EuDML project. The purpose is to support the development of the concepts, criteria and methods for the continuous evaluation of these and new relevant existing services. The work was concentrated on the classification of the relevant services in order to specify a common evaluating structure.

Digital Libraries as a Social Media

This paper presents an innovative approach for enhancing digital libraries functionalities. An innovative distributed architecture involving digital libraries for effective and efficient knowledge sharing was developed. In the frame of this architecture semantic services were implemented, offering multi language and multi culture support, adaptability and knowledge resources recommendation, based on the use of ontologies, metadata and user modeling. New methods for teacher education using digital libraries and knowledge sharing were developed. These new methods were successfully applied in more than 15 pilot experiments in seven European countries, with more than 3000 teachers trained.

Using domain-specific knowledge to improve information retrieval performance

Mémoire numérisé par la Direction des bibliothèques de l'Université de Montréal.

Using domain-specific knowledge to improve information retrieval performance

Mémoire numérisé par la Direction des bibliothèques de l'Université de Montréal.

Vocabulary and language model adaptation using information retrieval

International audience

Content and Connectivity in the Age of Digital Libraries

"Behind the world of virtual university is more than a method or system of work, you need to own, develop and master a connectivity structure of both technological and content development in multimedia digital text and then implement teaching methods line. "

Information Analysis and Databases

Means to automate the fact replace the man in their job functions for a man and machines automatic mechanism, ie documentary specialists in computer and computers are the cornerstone of any modern system of documentation and information. From this point of view immediately raises the problem of deciding what resources should be applied to solve the specific problem in each specific case. We will not let alone to propose quick fixes or recipes in order to decide what to do in any case. The solution depends on repeat for each particular problem. What we want is to move some points that can serve as a basis for reflection to help find the best solution possible, once the problem is defined correctly. The first thing to do before starting any automated system project is to define exactly the domain you want to cover and assess with greater precision possible importance.

Sperimentazione di Deep Metric Loss per Self-Supervised Information Retrieval Systems su CORD19.

Dopo lo sviluppo dei primi casi di Covid-19 in Cina nell’autunno del 2019, ad inizio 2020 l’intero pianeta è precipitato in una pandemia globale che ha stravolto le nostre vite con conseguenze che non si vivevano dall’influenza spagnola. La grandissima quantità di paper scientifici in continua pubblicazione sul coronavirus e virus ad esso affini ha portato alla creazione di un unico dataset dinamico chiamato CORD19 e distribuito gratuitamente. Poter reperire informazioni utili in questa mole di dati ha ulteriormente acceso i riflettori sugli information retrieval systems, capaci di recuperare in maniera rapida ed efficace informazioni preziose rispetto a una domanda dell'utente detta query. Di particolare rilievo è stata la TREC-COVID Challenge, competizione per lo sviluppo di un sistema di IR addestrato e testato sul dataset CORD19. Il problema principale è dato dal fatto che la grande mole di documenti è totalmente non etichettata e risulta dunque impossibile addestrare modelli di reti neurali direttamente su di essi. Per aggirare il problema abbiamo messo a punto nuove soluzioni self-supervised, a cui abbiamo applicato lo stato dell'arte del deep metric learning e dell'NLP. Il deep metric learning, che sta avendo un enorme successo soprattuto nella computer vision, addestra il modello ad "avvicinare" tra loro immagini simili e "allontanare" immagini differenti. Dato che sia le immagini che il testo vengono rappresentati attraverso vettori di numeri reali (embeddings) si possano utilizzare le stesse tecniche per "avvicinare" tra loro elementi testuali pertinenti (e.g. una query e un paragrafo) e "allontanare" elementi non pertinenti. Abbiamo dunque addestrato un modello SciBERT con varie loss, che ad oggi rappresentano lo stato dell'arte del deep metric learning, in maniera completamente self-supervised direttamente e unicamente sul dataset CORD19, valutandolo poi sul set formale TREC-COVID attraverso un sistema di IR e ottenendo risultati interessanti.

Self-supervised information retrieval: a novel approach based on Deep Metric Learning and Neural Language Models

Most of the existing open-source search engines, utilize keyword or tf-idf based techniques to find relevant documents and web pages relative to an input query. Although these methods, with the help of a page rank or knowledge graphs, proved to be effective in some cases, they often fail to retrieve relevant instances for more complicated queries that would require a semantic understanding to be exploited. In this Thesis, a self-supervised information retrieval system based on transformers is employed to build a semantic search engine over the library of Gruppo Maggioli company. Semantic search or search with meaning can refer to an understanding of the query, instead of simply finding words matches and, in general, it represents knowledge in a way suitable for retrieval. We chose to investigate a new self-supervised strategy to handle the training of unlabeled data based on the creation of pairs of ’artificial’ queries and the respective positive passages. We claim that by removing the reliance on labeled data, we may use the large volume of unlabeled material on the web without being limited to languages or domains where labeled data is abundant.

Self-supervised Information Retrieval basato su deep metric learning

La tesi ha lo scopo di ricercare, esaminare ed implementare un sistema di Machine Learning, un Recommendation Systems per precisione, che permetta la racommandazione di documenti di natura giuridica, i quali sono già stati analizzati e categorizzati appropriatamente, in maniera ottimale, il cui scopo sarebbe quello di accompagnare un sistema già implementato di Information Retrieval, istanziato sopra una web application, che permette di ricercare i documenti giuridici appena menzionati.

Avaluació dels tutorials de PubMed i Eric

La societat de la informació ofereix noves vies d'experimentació en el camp dels materials formatius. D'una banda, les tecnologies de la informació proporcionen maneres més interactives i adients de presentar el material i, de l'altra, apareixen nous conceptes en el camp de la formació i de l'educació, com ara la necessitat d'una formació continuada, la concepció de nous entorns educatius virtuals i, en definitiva, la necessitat d'elaborar nous materials didàctics que s'adaptin a aquestes situacions. El present estudi defineix una sèrie d'indicadors formals i d'aprenentatge que ajuden a avaluar i millorar els tutorials sobre la recuperació de la informació i els aplica, de manera comparativa, a dos tutorials de bases de dades bibliogràfiques: ERIC i PubMed. Els indicadors de contingut s'obtenen d'aspectes relacionats amb la didàctica i el model instructiu, mentre que els indicadors formals s'extreuen dels conceptes d'usabilitat i disseny de pàgines web. La finalitat d'aquest estudi és la sistematització d'uns criteris que permetin avaluar i millorar el disseny de tutorials sobre recuperació de la informació.