TaughtNet: Learning Multi-Task Biomedical Named Entity Recognition From Single-Task Teachers.

In Biomedical Named Entity Recognition (BioNER), the use of current cutting-edge deep learning-based methods, such as deep bidirectional transformers (e.g. BERT, GPT-3), can be substantially hampered by the absence of publicly accessible annotated datasets. When the BioNER system is required to annotate multiple entity types, various challenges arise because the majority of current publicly available datasets contain annotations for just one entity type: for example, mentions of disease entities may not be annotated in a dataset specialized in the recognition of drugs, resulting in a poor ground truth when using the two datasets to train a single multi-task model. In this work, we propose TaughtNet, a knowledge distillation-based framework allowing us to fine-tune a single multi-task student model by leveraging both the ground truth and the knowledge of single-task teachers. Our experiments on the recognition of mentions of diseases, chemical compounds and genes show the appropriateness and relevance of our approach w.r.t. strong state-of-the-art baselines in terms of precision, recall and F1 scores. Moreover, TaughtNet allows us to train smaller and lighter student models, which may be easier to be used in real-world scenarios, where they have to be deployed on limited-memory hardware devices and guarantee fast inferences, and shows a high potential to provide explainability. We publicly release both our code on github1 and our multi-task model on the huggingface repository.2.

Music Recommendation via Hypergraph Embedding.

In recent years, we have witnessed an ever wider spread of multimedia streaming platforms (e.g., Netflix, Spotify, and Amazon). Hence, it has become more and more essential to provide such systems with advanced recommendation facilities, in order to support users in browsing these massive collections of multimedia data according to their preferences and needs. In this context, the modeling of entities and their complex relationships (e.g., users listening to topic-based songs or authors creating different releases of their lyrics) represents the key challenge to improve the recommendation and maximize the users' satisfaction. To this end, this is the first study to leverage the high representative power of hypergraph data structures in combination with modern graph machine learning techniques in the context of music recommendation. Specifically, we propose hypergraph embeddings for music recommendation (HEMR), a novel framework for song recommendation based on hypergraph embedding. The hypergraph data model allows us to represent seamlessly all the possible and complex interactions between users and songs with the related characteristics; meanwhile, embedding techniques provide a powerful way to infer the user-song similarities by vector mapping. We have experimented the effectiveness and efficiency of our approach with respect to the state-of-the-art most recent music recommender systems, exploiting the Million Song dataset. The results show that HEMR significantly outperforms other state-of-the-art techniques, especially in scenarios where the cold-start problem arises, thus making our system a suitable solution to embed within a music streaming platform.

Predictive Analysis of Healthcare-Associated Blood Stream Infections in the Neonatal Intensive Care Unit Using Artificial Intelligence: A Single Center Study.

BACKGROUND: Neonatal infections represent one of the six main types of healthcare-associated infections and have resulted in increasing mortality rates in recent years due to preterm births or problems arising from childbirth. Although advances in obstetrics and technologies have minimized the number of deaths related to birth, different challenges have emerged in identifying the main factors affecting mortality and morbidity. Dataset characterization: We investigated healthcare-associated infections in a cohort of 1203 patients at the level III Neonatal Intensive Care Unit (ICU) of the "Federico II" University Hospital in Naples from 2016 to 2020 (60 months). METHODS: The present paper used statistical analyses and logistic regression to identify an association between healthcare-associated blood stream infection (HABSIs) and the available risk factors in neonates and prevent their spread. We designed a supervised approach to predict whether a patient suffered from HABSI using seven different artificial intelligence models. RESULTS: We analyzed a cohort of 1203 patients and found that birthweight and central line catheterization days were the most important predictors of suffering from HABSI. CONCLUSIONS: Our statistical analyses showed that birthweight and central line catheterization days were significant predictors of suffering from HABSI. Patients suffering from HABSI had lower gestational age and birthweight, which led to longer hospitalization and umbilical and central line catheterization days than non-HABSI neonates. The predictive analysis achieved the highest Area Under Curve (AUC), accuracy and F1-macro score in the prediction of HABSIs using Logistic Regression (LR) and Multi-layer Perceptron (MLP) models, which better resolved the imbalanced dataset (65 infected and 1038 healthy).

A comprehensive Benchmark for fake news detection.

Nowadays, really huge volumes of fake news are continuously posted by malicious users with fraudulent goals thus leading to very negative social effects on individuals and society and causing continuous threats to democracy, justice, and public trust. This is particularly relevant in social media platforms (e.g., Facebook, Twitter, Snapchat), due to their intrinsic uncontrolled publishing mechanisms. This problem has significantly driven the effort of both academia and industries for developing more accurate fake news detection strategies: early detection of fake news is crucial. Unfortunately, the availability of information about news propagation is limited. In this paper, we provided a benchmark framework in order to analyze and discuss the most widely used and promising machine/deep learning techniques for fake news detection, also exploiting different features combinations w.r.t. the ones proposed in the literature. Experiments conducted on well-known and widely used real-world datasets show advantages and drawbacks in terms of accuracy and efficiency for the considered approaches, even in the case of limited content information.

An Epidemiological Neural Network Exploiting Dynamic Graph Structured Data Applied to the COVID-19 Outbreak.

With the recent COVID-19 outbreak, we have assisted to the development of new epidemic models or the application of existing methodologies to predict the virus spread and to analyze how the different lock-down strategies can effectively influence the epidemic diffusion. In this paper, we propose a novel machine learning based framework able to estimate the parameters of any epidemiological model, such as contact rates and recovery rates, based on static and dynamic features of places. In particular, we model mobility data through a graph series whose spatial and temporal features are investigated by combining Graph Convolutional Neural Networks (GCNs) and Long short-term memories (LSTMs) in order to infer the parameters of SIR and SIRD models. We evaluate the proposed approach using data related to the COVID-19 dynamics in Italy and we compare the forecasts of the trained model with available data about the epidemic spread.

A survey of Big Data dimensions vs Social Networks analysis.

The pervasive diffusion of Social Networks (SN) produced an unprecedented amount of heterogeneous data. Thus, traditional approaches quickly became unpractical for real life applications due their intrinsic properties: large amount of user-generated data (text, video, image and audio), data heterogeneity and high speed generation rate. More in detail, the analysis of user generated data by popular social networks (i.e Facebook (https://www.facebook.com/), Twitter (https://www.twitter.com/), Instagram (https://www.instagram.com/), LinkedIn (https://www.linkedin.com/)) poses quite intriguing challenges for both research and industry communities in the task of analyzing user behavior, user interactions, link evolution, opinion spreading and several other important aspects. This survey will focus on the analyses performed in last two decades on these kind of data w.r.t. the dimensions defined for Big Data paradigm (the so called Big Data 6 V's).