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.

L'approccio Multi-stakeholder per la progettazione di servizi abilitati da tecnologia in sanità: il caso dell'Ospedale Maggiore di Parma

Alcuni cambiamenti demografici, come l’aumento dell’età media della popolazione, stanno spingendo il mondo sanitario ad innovarsi. In questo contesto, emergono i servizi abilitati da tecnologia che assicurano alcuni vantaggi essenziali per raggiungere un’assistenza sanitaria centrata sulle persone. Per progettare soluzioni tecnologiche innovative che stiano al passo con i cambiamenti, viene proposto l’approccio del service design. In particolare si presenta la CeHRes Roadmap che, attraverso sei steps iterativi, guida il team di progettazione dall’analisi del contesto alla realizzazione di una soluzione che soddisfi i bisogni degli stakeholder. Le prime tre fasi di tale metodologia sono state applicate ad un caso studio reale, quello dell’Ospedale Maggiore di Parma. In particolare, si è progettata una soluzione che rende più efficiente la presa in carico dei pazienti che necessitano di una visita specialistica urgente. Lo strumento di analisi preponderante sono state le interviste agli stakeholder dalle quali sono stati raccolti dati qualitativi grezzi. Dall’analisi e rielaborazione di questi ultimi, sono stati estrapolati i bisogni degli attori chiave in ottica di una visione orientata alle persone. Successivamente si è proposto il pretotipo di una soluzione che, a seguito di test con gli utilizzatori, è stato modificato sulla base dei loro feedback. Il prodotto finale del progetto è un pretotipo finale che rispetta tutti i requisiti emersi nel corso dell’intero processo di progettazione. Tuttavia, la CeHRes Roadmap, si concentra solo sulle relazioni diadiche tra stakeholder. Per questo il lavoro è stato approfondito in ottica multi-stakeholder con l’applicazione del Service Design For Value Network, una metodologia che pone l’accento sulle relazioni molti-a-molti. L’elaborato si conclude con la definizione degli steps futuri da percorrere per la realizzazione del pretotipo proposto, portando a termine la CeHRes Roadmap.

Sviluppo di Metodi di Soft Labeling per la Multi-Document Summarization in Ambito Legale

In questo elaborato viene trattata l’analisi del problema di soft labeling applicato alla multi-document summarization, in particolare vengono testate varie tecniche per estrarre frasi rilevanti dai documenti presi in dettaglio, al fine di fornire al modello di summarization quelle di maggior rilievo e più informative per il riassunto da generare. Questo problema nasce per far fronte ai limiti che presentano i modelli di summarization attualmente a disposizione, che possono processare un numero limitato di frasi; sorge quindi la necessità di filtrare le informazioni più rilevanti quando il lavoro si applica a documenti lunghi. Al fine di scandire la metrica di importanza, vengono presi come riferimento metodi sintattici, semantici e basati su rappresentazione a grafi AMR. Il dataset preso come riferimento è Multi-LexSum, che include tre granularità di summarization di testi legali. L’analisi in questione si compone quindi della fase di estrazione delle frasi dai documenti, della misurazione delle metriche stabilite e del passaggio al modello stato dell’arte PRIMERA per l’elaborazione del riassunto. Il testo ottenuto viene poi confrontato con il riassunto target già fornito, considerato come ottimale; lavorando in queste condizioni l’obiettivo è di definire soglie ottimali di upper-bound per l’accuratezza delle metriche, che potrebbero ampliare il lavoro ad analisi più dettagliate qualora queste superino lo stato dell’arte attuale.

Neural Radiance Fields for Relighting and View Synthesis

Neural scene representation and neural rendering are new computer vision techniques that enable the reconstruction and implicit representation of real 3D scenes from a set of 2D captured images, by fitting a deep neural network. The trained network can then be used to render novel views of the scene. A recent work in this field, Neural Radiance Fields (NeRF), presented a state-of-the-art approach, which uses a simple Multilayer Perceptron (MLP) to generate photo-realistic RGB images of a scene from arbitrary viewpoints. However, NeRF does not model any light interaction with the fitted scene; therefore, despite producing compelling results for the view synthesis task, it does not provide a solution for relighting. In this work, we propose a new architecture to enable relighting capabilities in NeRF-based representations and we introduce a new real-world dataset to train and evaluate such a model. Our method demonstrates the ability to perform realistic rendering of novel views under arbitrary lighting conditions.

Hierarchical Multi-Label Text Classification in a Low-Resource Setting

In this thesis we address a multi-label hierarchical text classification problem in a low-resource setting and explore different approaches to identify the best one for our case. The goal is to train a model that classifies English school exercises according to a hierarchical taxonomy with few labeled data. The experiments made in this work employ different machine learning models and text representation techniques: CatBoost with tf-idf features, classifiers based on pre-trained models (mBERT, LASER), and SetFit, a framework for few-shot text classification. SetFit proved to be the most promising approach, achieving better performance when during training only a few labeled examples per class are available. However, this thesis does not consider all the hierarchical taxonomy, but only the first two levels: to address classification with the classes at the third level further experiments should be carried out, exploring methods for zero-shot text classification, data augmentation, and strategies to exploit the hierarchical structure of the taxonomy during training.

Doehlert Design-desirability Function Multi-criteria Optimal Separation Of 17 Phenolic Compounds From Extra-virgin Olive Oil By Capillary Zone Electrophoresis.

In Brazil, the consumption of extra-virgin olive oil (EVOO) is increasing annually, but there are no experimental studies concerning the phenolic compound contents of commercial EVOO. The aim of this work was to optimise the separation of 17 phenolic compounds already detected in EVOO. A Doehlert matrix experimental design was used, evaluating the effects of pH and electrolyte concentration. Resolution, runtime and migration time relative standard deviation values were evaluated. Derringer's desirability function was used to simultaneously optimise all 37 responses. The 17 peaks were separated in 19min using a fused-silica capillary (50μm internal diameter, 72cm of effective length) with an extended light path and 101.3mmolL(-1) of boric acid electrolyte (pH 9.15, 30kV). The method was validated and applied to 15 EVOO samples found in Brazilian supermarkets.

Behavior Of Listeria Monocytogenes In A Multi-species Biofilm With Enterococcus Faecalis And Enterococcus Faecium And Control Through Sanitation Procedures.

The formation of mono-species biofilm (Listeria monocytogenes) and multi-species biofilms (Enterococcus faecium, Enterococcus faecalis, and L. monocytogenes) was evaluated. In addition, the effectiveness of sanitation procedures for the control of the multi-species biofilm also was evaluated. The biofilms were grown on stainless steel coupons at various incubation temperatures (7, 25 and 39°C) and contact times (0, 1, 2, 4, 6 and 8days). In all tests, at 7°C, the microbial counts were below 0.4 log CFU/cm(2) and not characteristic of biofilms. In mono-species biofilm, the counts of L. monocytogenes after 8days of contact were 4.1 and 2.8 log CFU/cm(2) at 25 and 39°C, respectively. In the multi-species biofilms, Enterococcus spp. were present at counts of 8 log CFU/cm(2) at 25 and 39°C after 8days of contact. However, the L. monocytogenes in multi-species biofilms was significantly affected by the presence of Enterococcus spp. and by temperature. At 25°C, the growth of L. monocytogenes biofilms was favored in multi-species cultures, with counts above 6 log CFU/cm(2) after 8days of contact. In contrast, at 39°C, a negative effect was observed for L. monocytogenes biofilm growth in mixed cultures, with a significant reduction in counts over time and values below 0.4 log CFU/cm(2) starting at day 4. Anionic tensioactive cleaning complemented with another procedure (acid cleaning, disinfection or acid cleaning+disinfection) eliminated the multi-species biofilms under all conditions tested (counts of all micro-organisms<0.4 log CFU/cm(2)). Peracetic acid was the most effective disinfectant, eliminating the multi-species biofilms under all tested conditions (counts of the all microorganisms <0.4 log CFU/cm(2)). In contrast, biguanide was the least effective disinfectant, failing to eliminate biofilms under all the test conditions.

Isotopic view of vegetation and carbon and nitrogen cycles in a cerrado ecosystem, southeastern Brazil

Carbon and nitrogen biogeochemical cycles in savannas are strongly regulated by the seasonal distribution of precipitation and pulses of nutrients released during the wetting of the dry soil and are critical to the dynamics of microorganisms and vegetation. The objective of this study was to investigate the spatial and temporal variability of C and N isotope ratios as indicators of the cycling of these elements in a cerrado sensu stricto area, within a protected area in a State Park in the state of São Paulo, Brazil. The foliar δ13C and δ15N values varied from -33.6 to -24.4 ‰ and -2.5 to 4.5 ‰, respectively. The δ13C values showed a consistent relationship with canopy height, revealing the importance of structure of the canopy over the C isotopic signature of the vegetation. Carbon isotopic variations associated with the length of the dry season indicated the importance of recent fixed C to the integrated isotopic signature of the leaf organic C. The studied Cerrado species showed a depleted foliar δ15N, but a wide range of foliar Nitrogen with no difference among canopy heights. However, seasonal variability was observed, with foliar δ15N values being higher in the transition period between dry and rainy seasons. The variation of the foliar C and N isotope ratios presented here was consistent with highly diverse vegetation with high energy available but low availability of water and N.

A multi-screening approach for marine-derived fungal metabolites and the isolation of cyclodepsipeptides from Beauveria felina

Extracts obtained from 57 marine-derived fungal strains were analyzed by HPLC-PDA, TLC and ¹H NMR. The analyses showed that the growth conditions affected the chemical profile of crude extracts. Furthermore, the majority of fungal strains which produced either bioactive of chemically distinctive crude extracts have been isolated from sediments or marine algae. The chemical investigation of the antimycobacterial and cytotoxic crude extract obtained from two strains of the fungus Beauveria felina have yielded cyclodepsipeptides related to destruxins. The present approach constitutes a valuable tool for the selection of fungal strains that produce chemically interesting or biologically active secondary metabolites.

A hybrid heuristic for the multi-plant capacitated lot sizing problem with setup carry-over

This paper addresses the capacitated lot sizing problem (CLSP) with a single stage composed of multiple plants, items and periods with setup carry-over among the periods. The CLSP is well studied and many heuristics have been proposed to solve it. Nevertheless, few researches explored the multi-plant capacitated lot sizing problem (MPCLSP), which means that few solution methods were proposed to solve it. Furthermore, to our knowledge, no study of the MPCLSP with setup carry-over was found in the literature. This paper presents a mathematical model and a GRASP (Greedy Randomized Adaptive Search Procedure) with path relinking to the MPCLSP with setup carry-over. This solution method is an extension and adaptation of a previously adopted methodology without the setup carry-over. Computational tests showed that the improvement of the setup carry-over is significant in terms of the solution value with a low increase in computational time.

A escolha do cônjuge

Enfocando os tópicos namoro e noivado, idade ao casar e diferenças de idade entre os cônjuges, este artigo caminha por três grandes períodos históricos: o período colonial; o século XIX até finais do XX; e o início do século XXI. Trata-se de uma análise transdisciplinar dos pontos de vista histórico, social, demográfico, legal e jurídico, focalizando o desenrolar dessas situações seja por persistências e/ou mudanças; quem e quais são os atores envolvidos e sua importância nas escolhas dos futuros nubentes. No início, a seleção era fundada no parentesco e no território, privilegiando a grande rede familiar solidária no enfrentamento das dificuldades de sobrevivência. Com os avanços da industrialização, da urbanização, da tecnologia e de comunicação, além do crescimento da importância das pessoas, as escolhas foram se transformando, diminuindo a dependência da estrutura familiar e aumentando a escolha pessoal e afetiva, influenciadas pelas variadas formas de mudança social, demográfica, jurídica.

Dietary calcium intake and overweight: an epidemiologic view

OBJECTIVE: We evaluated the relation between overweight and calcium intake in adults living in the municipality of São Paulo, Brazil. METHODS: This was a cross-sectional population-based study on a sample of 1459 adults that was obtained by multistage cluster sampling. Dietary intake was measured by the 24-h recall method. Poisson's and linear regression analyses were performed to evaluate the relation between overweight and quartiles of calcium intake adjusted for energy. RESULTS: The prevalence of overweight was 43.1% and the average adjusted calcium intake was 448.6 mg. In the linear regression analyses, the regression coefficient for adjusted calcium was significant and negative (P = 0.019, beta(1) = -0.0001). Although evaluated by quartiles, the prevalence ratio for overweight in the first quartile of calcium intake was 1.24 (95% confidence interval 1.00-1.54) and that in the second quartile was 1.24 (95% confidence interval 1.03-1.49). CONCLUSION: In the present study, calcium intake showed a significant negative association with body mass index

Design methodology for multi-pumped discrete Raman amplifiers: case-study employing photonic crystal fibers

This paper proposes a new design methodology for discrete multi-pumped Raman amplifier. In a multi-objective optimization scenario, in a first step the whole solution-space is inspected by a CW analytical formulation. Then, the most promising solutions are fully investigated by a rigorous numerical treatment and the Raman amplification performance is thus determined by the combination of analytical and numerical approaches. As an application of our methodology we designed an photonic crystal fiber Raman amplifier configuration which provides low ripple, high gain, clear eye opening and a low power penalty. The amplifier configuration also enables to fully compensate the dispersion introduced by a 70-km singlemode fiber in a 10 Gbit/s system. We have successfully obtained a configuration with 8.5 dB average gain over the C-band and 0.71 dB ripple with almost zero eye-penalty using only two pump lasers with relatively low pump power. (C) 2009 Optical Society of America

The effectiveness of educational practice in diabetic foot: a view from Brazil

Background: The aim of the present study was to evaluate the prevention and self-inspection behavior of diabetic subjects with foot at ulcer risk, no previous episode, who participated in the routine visits and standardized education provided by the service and who received prescribed footwear. This evaluation was carried out using a questionnaire scoring from 0-10 (high scores reflect worse practice compliance). Results: 60 patients were studied (30 of each sex); mean age was 62 years, mean duration of the disease was 17 years. As for compliance, 90% showed a total score <= 5, only 8.7% regularly wore the footwear supplied; self foot inspection 65%, 28,3% with additional familiar inspection; creaming 77%; proper washing and drying 88%; proper cutting of toe nails 83%; no cuticle cutting 83%; routine shoe inspection 77%; no use of pumice stones or similar abrasive 70%; no barefoot walking 95%. Conclusion: the planned and multidisciplinary educational approach enabled high compliance of the ulcer prevention care needed in diabetic patients at risk for complications. In contrast, compliance observed for the use of footwear provided was extremely low, demonstrating that the issue of its acceptability should be further and carefully addressed. In countries of such vast dimensions as Brazil multidisciplinary educational approaches can and should be performed by the services providing care for patients with foot at risk for complications according to the reality of local scenarios. Furthermore, every educational program should assess the learning, results obtained and efficacy in the target population by use of an adequate evaluation system.

Multi-trait and random regression mature weight heritability and breeding value estimates in Nelore cattle

Mature weight breeding values were estimated using a multi-trait animal model (MM) and a random regression animal model (RRM). Data consisted of 82 064 weight records from 8 145 animals, recorded from birth to eight years of age. Weights at standard ages were considered in the MM. All models included contemporary groups as fixed effects, and age of dam (linear and quadratic effects) and animal age as covariates. In the RRM, mean trends were modelled through a cubic regression on orthogonal polynomials of animal age and genetic maternal and direct and maternal permanent environmental effects were also included as random. Legendre polynomials of orders 4, 3, 6 and 3 were used for animal and maternal genetic and permanent environmental effects, respectively, considering five classes of residual variances. Mature weight (five years) direct heritability estimates were 0.35 (MM) and 0.38 (RRM). Rank correlation between sires' breeding values estimated by MM and RRM was 0.82. However, selecting the top 2% (12) or 10% (62) of the young sires based on the MM predicted breeding values, respectively 71% and 80% of the same sires would be selected if RRM estimates were used instead. The RRM modelled the changes in the (co) variances with age adequately and larger breeding value accuracies can be expected using this model.