GFGiovanni Felici

Giovanni Felici

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Giovanni Felici has a total of 188 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Optimizing accuracy and diversity: a multi-task approach to forecast combinations

    by Giovanni Felici, Antonio M. Sudoso

    We present a multi-task optimization approach based on a deep learning architecture for time series forecasting. We leverage large collections of time series to identify the weights of forecasting models that can be combined to produce forecasts for each series. This method jointly addresses two tasks: the selection of different forecasting models, and their effective combination. In doing so, it keeps into account, in an original way, both the accuracy and diversity of the forecasting methods. For a given time series, the model combination module extracts features and uses them to optimize the weights of the forecasting methods. Simultaneously, the model selection module extracts other features to identify the subset of methods to be used for the prediction. This selection process is framed as a classification problem, with the labels representing the set of models to be used for a series. These labels are determined by solving an auxiliary optimization problem that identifies accurate and diverse methods for each time series. The outputs of the two modules are then combined and the entire neural network is jointly trained by minimizing a custom loss function via gradient descent optimization. Experimental results on a large set of series from the M4 competition dataset and from real road traffic data show that our proposal enhances point forecast accuracy compared to state-of-the-art methods.

  • Open Access

    Meta16S: large-scale discovery and taxonomic assignment of unknown microbes from 16S amplicon sequencing samples

    by Fabio Cumbo, Giovanni Felici, Daniel Blankenberg, Federica Valeriani, Vincenzo Romano Spica, Daniele Santoni

    Background: The exponential growth of public metagenomic datasets offers an unprecedented opportunity to explore microbial diversity. However, analyzing this vast amount of data presents significant computational challenges. While shotgun metagenomics provides deep functional and taxonomic resolution, its high cost still limits its application. On the other hand, 16S rRNA gene sequencing remains a cost-effective and widely used alternative, but tools are needed to maximize its discovery potential. Traditional clustering is not scalable, obstructing the creation of a comprehensive and continuously updated catalog of microbial life from 16S data. Methods: We developed a reproducible and scalable Snakemake pipeline for the incremental clustering of 16S rRNA amplicons. The workflow begins by constructing a reference database from bacterial and archaeal genomes. It then processes 16S rRNA samples sequentially. For each new sample, sequences are first mapped against the existing cluster centroids. Sequences that match known centroids are assigned accordingly, while unmapped sequences are clustered independently to form novel operational taxonomic units (OTUs). These new centroids are then merged with the existing database, allowing it to grow dynamically without the need for computationally prohibitive all-at-once re-clustering. Results: Our pipeline enables the efficient and continuous expansion of a 16S rRNA cluster database. By processing a large corpus of public 16S rRNA samples, we generated a comprehensive atlas of tens of thousands of OTUs. A significant fraction of these clusters, particularly at the genus and family levels, were classified as unknown. Conclusions: This work provides a powerful, open-source tool for large-scale analysis of 16S rRNA samples. The incremental clustering strategy overcomes the scalability limitations of traditional methods, allowing researchers to leverage public data and discover novel microbes in their own microbiome samples.

  • Open Access

    Multi-Head Networks for Missing Modalities: A Study on Robustness in Multimodal Learning

    by Giovanni Felici, Diego Maria Pinto, Antonio Vicinanza

    Multimodal learning has emerged as a critical area of research in artificial intelligence, enabling models to integrate and process diverse data sources such as images, text, audio, and sensor readings. However, real-world multimodal systems are often challenged by missing modalities due to sensor failures, occlusions, or data transmission issues, leading to degraded performance. To address this, we investigate the robustness of Multi-Head Networks (MHNs) trained with an orthogonality component in the loss function, which enforces decorrelation among latent representations to enhance feature diversity and resilience to missing data. We develop and evaluate three MHN architecture variants, each leveraging different feature extraction strategies. Our experiments, conducted on a recently published multimodal dataset for behavioral biometrics, include an ablation study in which we systematically remove the most informative features and assess the impact of the orthogonality constraint. Results show that CNN-based architectures benefit the most from orthogonality loss, FNNs show moderate gains, while LSTM networks exhibit a variable response, suggesting that the effectiveness of this constraint is model-dependent. These findings also highlight the need for architecture-specific robustness strategies.

  • Open Access

    Digital twins for intelligent cities: the case study of Matera

    by Riccardo De Benedictis, Amedeo Cesta, Riccardo Pellegrini, Matteo Diez, Diego Maria Pinto, Paolo Ventura, Giuseppe Stecca, Giovanni Felici, Andreas Scalas, Michela Mortara, et al.

    The digital twin (DT) paradigm provides a purpose-built digital representation of a physical system. DTs are often composed by several interconnected models, which need to be specifically tailored to fit the DT purposes. The objective of this work is the development of an urban digital twin (UDT) for the city of Matera, following urban intelligence (UI) paradigm. The latter leverages the multi-disciplinary integration and optimization of the city systems and subsystems to develop purpose-driven DTs and support the decision-making process. The UDT is intended to support governance, stakeholders, and citizens, integrating morphological data for the city representation, reliable simulation tools and data-driven methods for the city state prediction (e.g., traffic, solar irradiation), optimization algorithms for planning and emergency response, sensors for vehicle and pedestrian traffic volumes and environmental monitoring (e.g., pollutant distributions), and a participatory data collection process from the citizens. A Data Lake is used to make available to the UDT modules the data produced by the morphological representation, simulations/data-driven methods, sensors, and the participatory data. The Data Lake provides a standardized approach for the aggregation and extraction of information. Finally, an Urban Sensing Engine supports the decision-making process with artificial intelligence forms of reasoning, to effectively combine the information produced by the UDT components.


  • Open Access

    Predicting Transcription Factor Binding Sites with Deep Learning

    by Nimisha Ghosh, Daniele Santoni, Indrajit Saha, Giovanni Felici

    Prediction of binding sites for transcription factors is important to understand how the latter regulate gene expression and how this regulation can be modulated for therapeutic purposes. A consistent number of references address this issue with different approaches, Machine Learning being one of the most successful. Nevertheless, we note that many such approaches fail to propose a robust and meaningful method to embed the genetic data under analysis. We try to overcome this problem by proposing a bidirectional transformer-based encoder, empowered by bidirectional long-short term memory layers and with a capsule layer responsible for the final prediction. To evaluate the efficiency of the proposed approach, we use benchmark ChIP-seq datasets of five cell lines available in the ENCODE repository (A549, GM12878, Hep-G2, H1-hESC, and Hela). The results show that the proposed method can predict TFBS within the five different cell lines very well; moreover, cross-cell predictions provide satisfactory results as well. Experiments conducted across cell lines are reinforced by the analysis of five additional lines used only to test the model trained using the others. The results confirm that prediction across cell lines remains very high, allowing an extensive cross-transcription factor analysis to be performed from which several indications of interest for molecular biology may be drawn.


  • Open Access

    Optimizing accuracy and diversity: a multi-task approach to forecast combinations

    by Giovanni Felici, Antonio M. Sudoso

    Forecast combination involves using multiple forecasts to create a single, more accurate prediction. Recently, feature-based forecasting has been employed to either select the most appropriate forecasting models or to optimize the weights of their combination. In this paper, we present a multi-task optimization paradigm that focuses on solving both problems simultaneously and enriches current operational research approaches to forecasting. In essence, it incorporates an additional learning and optimization task into the standard feature-based forecasting approach, focusing on the identification of an optimal set of forecasting methods. During the training phase, an optimization model with linear constraints and quadratic objective function is employed to identify accurate and diverse methods for each time series. Moreover, within the training phase, a neural network is used to learn the behavior of that optimization model. Once training is completed the candidate set of methods is identified using the network. The proposed approach elicits the essential role of diversity in feature-based forecasting and highlights the interplay between model combination and model selection when optimizing forecasting ensembles. Experimental results on a large set of series from the M4 competition dataset show that our proposal enhances point forecast accuracy compared to state-of-the-art methods.


  • Open Access

    Simultaneous feature selection and outlier detection with optimality guarantees

    by Luca Insolia, Ana Kenney, Francesca Chiaromonte, Giovanni Felici

    Biomedical research is increasingly data rich, with studies comprising ever growing numbers of features. The larger a study, the higher the likelihood that a substantial portion of the features may be redundant and/or contain contamination (outlying values). This poses serious challenges, which are exacerbated in cases where the sample sizes are relatively small. Effective and efficient approaches to perform sparse estimation in the presence of outliers are critical for these studies, and have received considerable attention in the last decade. We contribute to this area considering high‐dimensional regressions contaminated by multiplemean‐shift outliersaffecting both the response and the design matrix. We develop a general framework and usemixed‐integer programmingto simultaneously perform feature selection and outlier detection with provably optimal guarantees. We prove theoretical properties for our approach, that is, a necessary and sufficient condition for therobustly strong oracle property, where the number of features can increase exponentially with the sample size; the optimal estimation of parameters; and the breakdown point of the resulting estimates. Moreover, we provide computationally efficient procedures to tune integer constraints and warm‐start the algorithm. We show the superior performance of our proposal compared to existing heuristic methods through simulations and use it to study the relationships between childhood obesity and the human microbiome.


  • Open Access

    Informed peer review for publication assessments: Are improved impact measures worth the hassle?

    by Giovanni Abramo, Ciriaco Andrea D’angelo, Giovanni Felici

    In this work we ask whether and to what extent applying a predictor of publications’ impact better than early citations, has an effect on the assessment of research performance of individual scient...


  • Open Access

    CamurWeb: a classification software and a large knowledge base for gene expression data of cancer

    by Emanuel Weitschek, Silvia Di Lauro, Eleonora Cappelli, Paola Bertolazzi, Giovanni Felici

    The high growth of Next Generation Sequencing data currently demands new knowledge extraction methods. In particular, the RNA sequencing gene expression experimental technique stands out for case-control studies on cancer, which can be addressed with supervised machine learning techniques able to extract human interpretable models composed of genes, and their relation to the investigated disease. State of the art rule-based classifiers are designed to extract a single classification model, possibly composed of few relevant genes. Conversely, we aim to create a large knowledge base composed of many rule-based models, and thus determine which genes could be potentially involved in the analyzed tumor. This comprehensive and open access knowledge base is required to disseminate novel insights about cancer. We propose CamurWeb, a new method and web-based software that is able to extract multiple and equivalent classification models in form of logic formulas (“if then” rules) and to create a knowledge base of these rules that can be queried and analyzed. The method is based on an iterative classification procedure and an adaptive feature elimination technique that enables the computation of many rule-based models related to the cancer under study. Additionally, CamurWeb includes a user friendly interface for running the software, querying the results, and managing the performed experiments. The user can create her profile, upload her gene expression data, run the classification analyses, and interpret the results with predefined queries. In order to validate the software we apply it to all public available RNA sequencing datasets from The Cancer Genome Atlas database obtaining a large open access knowledge base about cancer. CamurWeb is available at http://bioinformatics.iasi.cnr.it/camurweb. The experiments prove the validity of CamurWeb, obtaining many classification models and thus several genes that are associated to 21 different cancer types. Finally, the comprehensive knowledge base about cancer and the software tool are released online; interested researchers have free access to them for further studies and to design biological experiments in cancer research.


  • Exact approaches for the orderly colored longest path problem: Performance comparison

    by Francesco Carrabs, Raffaele Cerulli, Giovanni Felici, Gaurav Singh