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.

Query rewriting in information retrieval: automatic context extraction from local user documents to improve query results

The central objective of research in Information Retrieval (IR) is to discover new techniques to retrieve relevant information in order to satisfy an Information Need. The Information Need is satisfied when relevant information can be provided to the user. In IR, relevance is a fundamental concept which has changed over time, from popular to personal, i.e., what was considered relevant before was information for the whole population, but what is considered relevant now is specific information for each user. Hence, there is a need to connect the behavior of the system to the condition of a particular person and his social context; thereby an interdisciplinary sector called Human-Centered Computing was born. For the modern search engine, the information extracted for the individual user is crucial. According to the Personalized Search (PS), two different techniques are necessary to personalize a search: contextualization (interconnected conditions that occur in an activity), and individualization (characteristics that distinguish an individual). This movement of focus to the individual's need undermines the rigid linearity of the classical model overtaken the ``berry picking'' model which explains that the terms change thanks to the informational feedback received from the search activity introducing the concept of evolution of search terms. The development of Information Foraging theory, which observed the correlations between animal foraging and human information foraging, also contributed to this transformation through attempts to optimize the cost-benefit ratio. This thesis arose from the need to satisfy human individuality when searching for information, and it develops a synergistic collaboration between the frontiers of technological innovation and the recent advances in IR. The search method developed exploits what is relevant for the user by changing radically the way in which an Information Need is expressed, because now it is expressed through the generation of the query and its own context. As a matter of fact the method was born under the pretense to improve the quality of search by rewriting the query based on the contexts automatically generated from a local knowledge base. Furthermore, the idea of optimizing each IR system has led to develop it as a middleware of interaction between the user and the IR system. Thereby the system has just two possible actions: rewriting the query, and reordering the result. Equivalent actions to the approach was described from the PS that generally exploits information derived from analysis of user behavior, while the proposed approach exploits knowledge provided by the user. The thesis went further to generate a novel method for an assessment procedure, according to the "Cranfield paradigm", in order to evaluate this type of IR systems. The results achieved are interesting considering both the effectiveness achieved and the innovative approach undertaken together with the several applications inspired using a local knowledge base.

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.

La digitalizzazione del processo costruttivo: la strategia Building Information Modeling and Management (BIMM) applicata ad un caso di studio di edilizia scolastica.

L’obiettivo della tesi è quello di fare una panoramica sulla strategia BIM e quindi sulla digitalizzazione del processo costruttivo. Grazie alla analisi di un caso di studio, altro obiettivo è quello di analizzare e valutare la metodologia BIM 4D/5D, ossia la gestione dei tempi e dei costi di realizzazione dell’opera. Nella prima fase si affronta il tema del BIM, con una analisi sull’evoluzione degli strumenti di elaborazione e rappresentazione digitale del progetto architettonico, su come questi strumenti si differenzino sia dal punto di vista operativo che concettuale rivoluzionando il flusso di lavoro odierno. Quindi, partendo da un’analisi che e ritrae l’estrema frammentazione del settore delle costruzioni, si va ad analizzare come il BIM aumenti e favorisca la collaborazione delle parti interessate, armonizzando l’intero processo costruttivo dell’opera. Si prosegue con l'esame della diffusione e del livello di maturità degli strumenti BIM, di come i privati e le amministrazioni pubbliche, a livello mondiale, stiano spingendo per favorire l’adozione della metodologia BIM. Inoltre si analizzano le dinamiche dell’interoperabilità, delle metodologie e protocolli di interscambio dati, che sono un elemento chiave per il successo del BIM per via dei numerosi strumenti, specializzati nelle varie discipline del settore edile. Nella terza parte, dedicata al Project Management di un caso di studio, si verifica la bontà delle metodologie teorizzate attraverso la realizzazione di un modello virtuale in Revit. Dal modello realizzato dal laureando sono estrapolate le informazioni necessarie alla gestione, e tramite il software STRVison CPM, si elaborano i principali documenti per la progettazione e gestione del cantiere: il CM, il CME, i tempi operativi, il cronoprogramma Gantt. Obbiettivo è constatare l’effettivo livello di maturità della strategia BIM 4D e 5D e la reale possibilità di un impiego capillare nel panorama italiano del settore delle costruzioni.

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.

Temporal PageRank

The our reality is characterized by a constant progress and, to follow that, people need to stay up to date on the events. In a world with a lot of existing news, search for the ideal ones may be difficult, because the obstacles that make it arduous will be expanded more and more over time, due to the enrichment of data. In response, a great help is given by Information Retrieval, an interdisciplinary branch of computer science that deals with the management and the retrieval of the information. An IR system is developed to search for contents, contained in a reference dataset, considered relevant with respect to the need expressed by an interrogative query. To satisfy these ambitions, we must consider that most of the developed IR systems rely solely on textual similarity to identify relevant information, defining them as such when they include one or more keywords expressed by the query. The idea studied here is that this is not always sufficient, especially when it's necessary to manage large databases, as is the web. The existing solutions may generate low quality responses not allowing, to the users, a valid navigation through them. The intuition, to overcome these limitations, has been to define a new concept of relevance, to differently rank the results. So, the light was given to Temporal PageRank, a new proposal for the Web Information Retrieval that relies on a combination of several factors to increase the quality of research on the web. Temporal PageRank incorporates the advantages of a ranking algorithm, to prefer the information reported by web pages considered important by the context itself in which they reside, and the potential of techniques belonging to the world of the Temporal Information Retrieval, exploiting the temporal aspects of data, describing their chronological contexts. In this thesis, the new proposal is discussed, comparing its results with those achieved by the best known solutions, analyzing its strengths and its weaknesses.

Graph-based Vision-Language Model for Multi-Modal Retrieval of Fashion Products

Artificial Intelligence is reshaping the field of fashion industry in different ways. E-commerce retailers exploit their data through AI to enhance their search engines, make outfit suggestions and forecast the success of a specific fashion product. However, it is a challenging endeavour as the data they possess is huge, complex and multi-modal. The most common way to search for fashion products online is by matching keywords with phrases in the product's description which are often cluttered, inadequate and differ across collections and sellers. A customer may also browse an online store's taxonomy, although this is time-consuming and doesn't guarantee relevant items. With the advent of Deep Learning architectures, particularly Vision-Language models, ad-hoc solutions have been proposed to model both the product image and description to solve this problems. However, the suggested solutions do not exploit effectively the semantic or syntactic information of these modalities, and the unique qualities and relations of clothing items. In this work of thesis, a novel approach is proposed to address this issues, which aims to model and process images and text descriptions as graphs in order to exploit the relations inside and between each modality and employs specific techniques to extract syntactic and semantic information. The results obtained show promising performances on different tasks when compared to the present state-of-the-art deep learning architectures.