Cognitive Services for the automatic recognition and translation of the Italian signed language (LIS)Contact person
MARISTELLA MATERAEmail:
maristella.matera@polimi.itStudy course: Computer Science and Engineering
Other members of the research group:
Emanuele PucciWeb page:
https://www.linkedin.com/feed/update/urn:li:activity:6872189179258798081?updateEntityUrn=urn%3Ali%3Afs_feedUpdate%3A%28V2%2Curn%3Ali%3Aactivity%3A6872189179258798081%29Description
Description:
Within a project conducted in collaboration with Capgemini (Corporate Social Responsibility) and Informatici Senza Frontiere (https://www.informaticisenzafrontiere.org/), there are different thesis projects related to designing, training, evaluating AI models for the construction of cognitive services for the recognition and translation into text/voice of the Italian signed language.
Hide Description
Reconfigurable multi-carrier modulation schemes for 6G vehicular networksContact person
MAURIZIO MAGARINIEmail:
maurizio.magarini@polimi.itStudy course: Telecommunications Engineering
Other members of the research group:
Francesco LinsalataDescription
Description:
Connected mobility is a flagship element of smart cities and a mandatory step in the evolution towards automated driving in 6G networks. The goal of the thesis is to identify what are the most appropriate multi-carrier modulation scheme (eg. OFDM, GFDM, etc.) and the different settings of the communication to satisfy the quality of service requirements of different use cases in vehicular networks. The main activities will be:
1) development of a software in Matlab/Python (depending on the knowledge of the student) that is able to process channel data obtained through synthetic simulations (data set is already available) .
2) Application of machine learning principles for the optimization of the performance.
Hide Description
Transmitter localization in diffusion based molecular communications systemsContact person
MAURIZIO MAGARINIEmail:
maurizio.magarini@polimi.itStudy course: Automation Engineering, Computer Science and Engineering, Telecommunications Engineering
Description
Description:
Molecular communication is an emerging research area in the bio and nano science where transport of molecules allows nanomachines (e.g. nano-scale biological or artificially created devices) to communicate over a short distance. A key design challenge in molecular communication is controlling the propagation direction of information-carrying molecules towards their destination. The focus of the thesis is about the estimation of position of a point wise transmitter in a three-dimensional molecular communication system based on a diffusive channel. Possible applications are in targeted drug delivery where nano-robots must move toward the targeted site that, in most cases, is a tumor.
Hide Description
Reti neurali ‘spiking’ per l’analisi di segnali elettroneurografici provenienti dal nervo sciaticoContact person
MAURIZIO MAGARINIEmail:
maurizio.magarini@polimi.itStudy course: Biomedical Engineering, Computer Science and Engineering, Telecommunications Engineering
Other members of the research group:
Antonio CovielloDescription
Description:
La proposta di tesi riguarda l’applicazione di tecniche basate sull'impiego di reti neurali ‘spiking’ per l’analisi di segnali biomedicali elettroneurografici (ENG) acquisiti dal nervo sciatico. L’obiettivo del lavoro e’ quello di classificare con la maggior accuratezza possibile differenti tipi di segnali sensoriali (propriocettivo, nocicezione, tocco, ecc. …). I segnali ENG da analizzare sono disponibili da misure effettuate presso l’Universita’ di Toronto e di Newcastle.
Hide Description
Progettazione di un sistema per il trasferimento dati e l’alimentazione di un circuito di registrazione e stimolazione neurale completamente impiantabileContact person
MAURIZIO MAGARINIEmail:
maurizio.magarini@polimi.itStudy course: Biomedical Engineering, Electrical Engineering, Electronics Engineering, Telecommunications Engineering
Other members of the research group:
Antonio CovielloDescription
Description:
La proposta di tesi riguarda lo sviluppo di un sistema di comunicazione bidirezionale con un’interfaccia impiantabile e la sua alimentazione elettrica. Lo studio prenderà in considerazione sia l’approccio radio che quello cablato e prevede una parte sperimentale. L’obiettivo finale è quello di realizzare un bypass neuronale per ristabilire una funzione motoria parzialmente persa. Il lavoro si inserisce nell’ambito di una collaborazione in corso con l’Imperial College di Londra e il Politecnico di Torino.
Hide Description
Conversational AI for accessing the WebContact person
MARISTELLA MATERAEmail:
maristella.matera@polimi.itStudy course: Computer Science and Engineering
Description
Description:
Voice is the most natural user interface: ConWeb wants to transform the interaction between user and website by restructuring the available information thanks to a framework for conversational AI. The goal of the thesis is ti refine the NLP pipeline of the framework to better process the user requests and the Website content. The project is developed in collaboration with associations of visually impaired and low-vision people.
Hide Description
Metaverse and Conversational AIContact person
MARISTELLA MATERAEmail:
maristella.matera@polimi.itStudy course: Computer Science and Engineering
Description
Description:
The metaverse is reality: as researchers, we must investigate this new approach to HCI. The goal of the thesis is to analyze the current approaches used to create a digital twin and ideate a framework to formalize the process and create common standards. The activity will be conducted with reference to ConWeb, a framework for accessing the Web through Conversational AI.
Hide Description
Explainable AIContact person
MARISTELLA MATERAEmail:
maristella.matera@polimi.itStudy course: Computer Science and Engineering
Other members of the research group:
Emanuele PucciDescription
Description:
Today the ability to transparently explain Artificial Intelligence algorithms is paramount: we want to investigate how argumentation theory, combined with adequate interaction mechanisms, can help us improve explainability and build trust between the user and the AI systems. In particular, the goal of the thesis is the development of an ontology and a related reasoning framework, which can constitute a reference for the argumentation algorithms, by providing the domain knowledge needed to achieve meaningful and more contextualized explanations.
Hide Description
misura di livello con sistemi a radiofrequenzaContact person
MICHELE NORGIAEmail:
michele.norgia@polimi.itStudy course: Electrical Engineering, Electronics Engineering, Telecommunications Engineering
Description
Description:
Si intende realizzare un sensore di livello per lquidi, basato sulla misura di sfasamento di segnali a radiofrequenza propaganti in un semplice doppino.
L'applicazione trainante è la misura di livello di carburante per aerei ed elicotteri.
Hide Description
Sistema optoelettronico di misura di vibrazioni e distanzaContact person
MICHELE NORGIAEmail:
michele.norgia@polimi.itStudy course: Automation Engineering, Electronics Engineering, Telecommunications Engineering
Description
Description:
Si intende investigare i limiti di misura di nuove strutture ottiche per la misura senza contatto, basate su interferometria a retroiniezione.
Recentemente è stata sviluppata una tecnica ottica completamente nuova, che misura la modulazione in frequenza che un diodo laser subisce quando risulta sottoposto a una debole retroiniezione ottica.
Lo scopo è realizzare strumenti ad altissime prestazioni per applicazioni industriali di monitoraggio.
Hide Description
Enforcing data sovereignty with blockchainContact person
PIERLUIGI PLEBANIEmail:
pierluigi.plebani@polimi.itStudy course: Computer Science and Engineering
Other members of the research group:
prof. Giovanni MeroniWeb page:
https://plebani.faculty.polimi.itDescription
Description:
Even in a trustless ecosystem organizations must share data but there is no way to be sure that the data are managed properly when they are transmitted to other parties, thus to ensure the data sovereignty.
The goal of the thesis is to consider the adoption of Ethereum to increase the trust among the organizations and to evaluate the performances and the costs related to the proposed solution.
More information on data sovereignty can be found at (https://www.itu.int/en/journal/002/Documents/ITU2018-11.pdf)
Hide Description
Data sovereignty in computing continuumContact person
PIERLUIGI PLEBANIEmail:
pierluigi.plebani@polimi.itStudy course: Computer Science and Engineering
Web page:
https://plebani.faculty.polimi.itDescription
Description:
Data sovereignty aims to find a good balance between the need for protecting one’s data and the need for sharing one’s data with others. This problem is particularly relevant when considering the cloud continuum, i.e., an environment encompassing resources from the edge to the cloud. In fact, while the edge is usually managed by the owner of the resources (thus the owner of the data) while moving to the cloud, data are managed by external actors. Goal of the thesis is to investigate how the data sovereignty can be assured as a combination of container based approaches and blockchain platforms.
Hide Description
Data catalog for trustworthy federated data lakesContact person
PIERLUIGI PLEBANIEmail:
pierluigi.plebani@polimi.itStudy course: Computer Science and Engineering
Other members of the research group:
prof. Cinzia CappielloWeb page:
https://plebani.faculty.polimi.itDescription
Description:
Trustworthy data lakes, especially when considering a federated setting, call for a shared data governance model among the involved organizations and a proper application – by tuning the underlying IT system – able to properly enact the corresponding policies. A central element for a correct data governance is the data catalog where data sets managed by a data lake are indexed and described by means of meta-data.
The goal of the thesis is to propose an innovative data catalog model which is open to cooperate with other data catalogs in a federation to improve data sharing among organisations. Case studies in manufacturing and healthcare domains are available.
For the relevance of trustworthy data sharing have a look to the EU data strategy document (https://eur-lex.europa.eu/legal-content/EN/TXT/?qid=1593073685620&uri=CELEX%3A52020DC0066)
Hide Description
Reliable process mining with low-quality event logs Contact person
PIERLUIGI PLEBANIEmail:
pierluigi.plebani@polimi.itStudy course: Computer Science and Engineering
Other members of the research group:
prof. Cinzia CappielloWeb page:
https://plebani.faculty.polimi.itDescription
Description:
Process mining (read the manifesto) is defined as data science in action. The aim is to extract insights and suggest actions from event logs produced by the systems in charge of executing the activities composing a business process. In this context, it is crucial the quality of the event logs: the higher the quality the better the results of the process mining. Unfortunately, most of the time, event logs are not complete, data not accurate or produced with delays.
Goal of the thesis is to investigate the relationship between quality of event logs and reliability of process mining to define the relationship between process models, event logs, and data quality.
For an introduction on this topic please have a look to this paper (https://www.sciencedirect.com/science/article/abs/pii/S0306437921000995?via%3Dihub).
Hide Description
Data-driven prediction of paper break events in paper manufacturing processContact person
LUIGI PIRODDIEmail:
luigi.piroddi@polimi.itStudy course: Automation Engineering
Other members of the research group:
Dr. Federico BianchiDescription
Description:
During the paper manufacturing process, some paper break events can occur. If a break happens, the entire process is stopped, any found problem is fixed and the production is resumed. Resuming the process can take more than an hour, which represents a severe cost for the factory. Even a 5% reduction in the break events will yield a significant cost saving.
The objective of this thesis is the development of suitable techniques to predict such breaks in advance (early prediction) and identify the potential cause(s) to prevent the break. To build such a prediction model, the applicant will use data collected from a network of sensors deployed to monitor a real manufacturing plant.
Hide Description
Prediction of stuck-pipe events in oil & gas drilling operationsContact person
LUIGI PIRODDIEmail:
luigi.piroddi@polimi.itStudy course: Automation Engineering, Computer Science and Engineering
Other members of the research group:
Prof. Matteo Matteucci, Dr. Federico Bianchi.Description
Description:
Drilling is a complex process that consists in boring a hole (that can be several thousands of meters long) in order to create a well for oil (or gas) production. A sticking occurs when the drillstring cannot be neither rotated nor moved along the axis of the wellbore. A pipe is considered stuck if it cannot be freed from the hole without damaging the pipe, and without exceeding the drilling rig's maximum allowed hook load. This phenomenon can be due to pressure-related issues (if the pressure in the annulus exceeds that in the formation being drilled, the drillstring is pulled against the wall of the borehole and held against it) or to mechanical blocking (e.g., due to incomplete removal of the cuttings, or partial collapse of the borehole). Stuck-pipe phenomena can have disastrous effects on drilling performance, with outcomes that may range from time delays (note that a day of drilling can cost millions of euros) to loss of expensive machinery. This thesis aims at the development (using machine learning techniques) of a data-based prediction model that can be used to anticipate the occurrence of the mentioned events. A large dataset of oil&gas wells can be used to train and validate the model.
This work is part of an ongoing collaboration with Eni's Drilling & Completion Risk Analysis Department.
Hide Description
Control of timed discrete event systems with timed specificationsContact person
LUIGI PIRODDIEmail:
luigi.piroddi@polimi.itStudy course: Automation Engineering
Other members of the research group:
Francesco Basile (Univ. Salerno)Description
Description:
Discrete event systems (DES) are characterized by a discrete set of states, and state changes are driven by the occurrence of (asynchronous) discrete events. This is a very general class of models, which is often encountered in automated manufacturing, communication and transportation networks, control applications, etc.
In timed DES the transitions between activities are further endowed with time bounds in order to represent the duration of activities. A timed control task involves reaching specific states (while avoiding others) within prescribed time limits.
The design of a controller for timed DES is a complex and challenging task that requires the development of suitable tools for reachability analysis, path optimization, and control policy design. This thesis aims at the development of a Matlab library of such tools for time Petri nets and its testing on some timed DES control benchmarks.
Hide Description
Endurance testing of contact materials for low voltage circuit breakersContact person
ENRICO RAGAINIEmail:
enrico.ragaini@polimi.itStudy course: Electrical Engineering
Web page:
new.abb.comDescription
Description:
The student shall complete the design and installation of a lab setup for endurance testing of contact materials.
Then, he/she shall perform testing of different contact materials and compare results under different operating conditions.
Thesis with internship at ABB SpA, located in Bergamo.
Hide Description
Hardware trusted environments for the creation of secure architecturesContact person
CHRISTIAN PILATOEmail:
christian.pilato@polimi.itStudy course: Electronics Engineering, Computer Science and Engineering
Web page:
https://pilato.faculty.polimi.itDescription
Description:
The key idea of this thesis is to build secure hardware architectures to protect data at rest and in use. In this thesis, the student has to develop a trusted environment including hardware sandboxes and encrypted memories to protect the execution of applications. The solution will be prototyped on FPGA devices, while the student will need to evaluate also ASIC implementations.
Hide Description
Memory architectures for big data analyticsContact person
CHRISTIAN PILATOEmail:
christian.pilato@polimi.itStudy course: Electronics Engineering, Computer Science and Engineering
Web page:
https://pilato.faculty.polimi.itDescription
Description:
This thesis focuses on the design of domain-specific architectures to optimize the data transfers around computational kernels. In this thesis, the student has to develop a methodology to automatically create a systolic array-like architecture with an adaptive memory system. Such a memory system will feature several components (reconfigurable caches, adaptive PLMs, overlays) to stream the data inside and outside the accelerators. To validate the generated architectures, the student will use FPGA prototyping.
Hide Description
Integration of custom accelerators with RISC-V processorsContact person
CHRISTIAN PILATOEmail:
christian.pilato@polimi.itStudy course: Electronics Engineering, Computer Science and Engineering
Web page:
https://pilato.faculty.polimi.itDescription
Description:
This thesis aims at creating a design methodology and proper hardware components to support the integration of HLS-generated hardware accelerators with a complete RISC-V CPU. The student must understand the details of the CPU and how the code is executed to interface with device components. Then, he/she must design proper hardware components to interface components generated through HLS with such infrastructure in a semi-automatic way. Execution of the complete system on FPGA is required to complete the evaluation.
Hide Description
Electromagnetic modeling of submarine power cablesContact person
LUCA DI RIENZOEmail:
luca.dirienzo@polimi.itStudy course: Electrical Engineering
Description
Conversational Web BrowsingContact person
MARISTELLA MATERAEmail:
maristella.matera@polimi.itStudy course: Computer Science and Engineering
Other members of the research group:
Cinzia Cappiello, Marcos Baez, Emanuele PucciWeb page:
https://link.springer.com/chapter/10.1007%2F978-3-030-62522-1_17Description
Description:
The thesis will focus on generating chatbots out of a website equipped with simple, bot-specific HTML annotations. The approach is part of what we call conversational web browsing, i.e., a dialog-based, natural language interaction with websites. The goal is to enable users to use content and functionality accessible through visual UIs by “talking to websites” instead of by operating the graphical UI using keyboard and mouse. The project is held in collaboration with associations for visually impaired and low-vision people.
Hide Description
End-User Development of chatbots for data explorationContact person
MARISTELLA MATERAEmail:
maristella.matera@polimi.itStudy course: Computer Science and Engineering
Web page:
http://matera.faculty.polimi.it/wp-content/uploads/chatbot.pdfDescription
Description:
Chatbots for data exploration support conversations that let the user progressively discover and interactively retrieve data from known data sources. Given the multiple advantages that the literature recognizes to the conversational paradigm for data access, and its current diffusion through different applications, there is now much emphasis on methodologies for chatbot design. A diffused opinion is that domain experts, with their special knowledge and understanding of a specific field, are critical in the development of chatbots; but it also emerges that engaging them in chatbot development is difficult due to their lack of technical competencies. Starting from a conceptual model for the rapid generation of chatbots, the thesis project will focus on the definition of interactive, visual paradigms for the End-User Development (EUD) of chatbots for data exploration.
Hide Description
Rapid Prototyping of Chatbots for Data ExplorationContact person
MARISTELLA MATERAEmail:
maristella.matera@polimi.itStudy course: Computer Science and Engineering
Web page:
http://matera.faculty.polimi.it/wp-content/uploads/chatbot.pdfDescription
Description:
This project aims to extend the current capabilities of frameworks for chatbot development in the domain of data exploration. Nowadays, chatbot applications are growing in a larger number of sectors and activities. However, it is still unclear how to capitalize on technologies for data access and exploration. The creation of the knowledge base with which the conversational agent will interact and its integration with the dialog system are still considered obstacles. This project aims to understand whether it is possible to extend the capabilities of current technologies to simplify the management of the chatbot knowledge base and fasten the development of chatbots for data exploration through model-based techniques for the automatic generation of the dialog system.
Hide Description
Adaptive Pipelines for Crowd-Enhanced ProcessesContact person
MONICA VITALIEmail:
monica.vitali@polimi.itStudy course: Computer Science and Engineering
Web page:
https://vitali.faculty.polimi.it/Description
Description:
Social media are a relevant source of information for decision-makers but their extraction and processing can be challenging. In fact, although data is extracted according to specific keywords through a crawler, a huge amount of irrelevant information is present in the resulting data set. These data have to be manually analyzed by human operators, for instance, Crowd workers, for annotating them with needed metadata. In this scenario, filtering out all irrelevant elements is necessary to make the approach feasible. This thesis aims at providing the user a methodology for supporting the definition of a pipeline able to eliminate irrelevant information according to the goal for the analysis to be performed on the data. Given a set of available components for the pipeline, each one is evaluated in terms of its efficiency (precision and recall, execution time, execution cost), and the best combination of components is selected according to the user’s quality requirements.
KEYWORDS: Business Process, Adaptive Processes, Crowd Computing, Social-media analysis, Big Data
Hide Description
Monitoring the Sustainability of Multi-Component ApplicationsContact person
MONICA VITALIEmail:
monica.vitali@polimi.itStudy course: Computer Science and Engineering
Web page:
https://vitali.faculty.polimi.it/Description
Description:
Modern applications are composed of several components (microservices, functions, monoliths) that cooperate together for reaching the overall application goal. They can be hosted in different locations in a cloud environment or in the owner facilities, as well as in edge or IoT devices. Most of the existing infrastructures don’t provide information about the energy consumption at the application component level, and even less information about the emissions generated by the application execution. Thus, some models to derive this information are needed if we want to envision a real sustainable application future. The aim of this thesis is to explore the current state of the art and the existing tools for deriving energy consumption of single-component applications and their emissions and to propose a method to extend these tools by enabling the derivation of the overall energy consumption of an application and sustainability impact.
KEYWORDS: Green IT, Microservices, Energy Monitoring, Sustainability-Awareness
Hide Description
Sustainable Applications Deployment in a Fog EnvironmentContact person
MONICA VITALIEmail:
monica.vitali@polimi.itStudy course: Computer Science and Engineering
Web page:
https://vitali.faculty.polimi.it/Description
Description:
Sustainability and Energy Efficiency are a driver for the improvement of our society and a necessity for a better future. While Data Centers are getting greener, this is not the case for other computing locations like fog and edge computing, composed of small and inefficient data centers. Considering applications as composed of several microservices interconnected in a workflow and exchanging data, the research question is how it is possible to adaptively deploy each component in a different location by considering the trade-off between QoS and Energy Efficiency. This thesis aims at providing an algorithm for deploying multi-component applications in a heterogeneous dynamic fog environment while minimizing emissions and respecting QoS constraints.
KEYWORDS: Green IT, Microservices, Fog Computing, Machine Learning
Hide Description
From Sustainable Applications Models to Executable Sustainable WorkflowsContact person
MONICA VITALIEmail:
monica.vitali@polimi.itStudy course: Computer Science and Engineering
Web page:
https://vitali.faculty.polimi.it/Description
Description:
Sustainability and Energy Efficiency are a driver for the improvement of our society and a necessity for a better future. While Data Centers are getting greener, applications hosted by them are keeping requiring more and more resources. This thesis aims at providing a methodology to design sustainable applications using existing modeling tools and enriching them with additional features and to implement the translation from a sustainable workflow model into an executable workflow using microservice-based applications and an orchestrator. The goal is to provide applications sensitive to their current energy consumption and emissions able to adapt by enacting modifications in their workflow by: i) defining optional and mandatory tasks, ii) defining alternative execution modalities (sustainable or performance-oriented) for the same task, iii) defining rules to activate different versions of the workflow based on the application monitoring data.
KEYWORDS: Green IT, BPM, Microservices
Hide Description
Adaptive Monitoring in Heterogeneous Distributed EnvironmentsContact person
MONICA VITALIEmail:
monica.vitali@polimi.itStudy course: Computer Science and Engineering
Web page:
https://vitali.faculty.polimi.it/Description
Description:
A Fog Environment sees computational resources spread in different locations with heterogeneous capabilities and characteristics. Data is most of the time collected at the edge of the network and moved to the cloud to be stored and analysed. With additional computing power at the edge, computation is moving more and more nearer the data, but some constraints have to be considered. This thesis will investigate the synergy between application deployment and data availability by providing a methodology for adaptively deploying applications in a fog environment while considering resource availability, data location, data quality requirements, and QoS requirements. The methodology aims at reducing the amount of data traveling from the edge to the cloud by proposing an adaptive monitoring system (AMS). The AMS employs Machine Learning techniques to detect correlations between monitoring data and build models to reconstruct missing data directly in the cloud. This thesis will explore different deployments of the AMS components to identify their outcome on the QoS of the monitoring data collection and on the Data Quality of the applications using the data. The goal is to provide a methodology to automatically tuning a monitoring data collection system according to the context of usage of the data.
KEYWORDS: Fog Computing, Adaptive Monitoring, IoT, Machine Learning
Hide Description
