- Uncertainty of results of deep learning techniques
The thesis concerns the evaluation of graph-based representation of scientific data and the evaluation of uncertainty of results of deep learning techniques. Considered applications can be in one of the following areas: mathematical models for chemistry, tweet geolocation techniques, sensor data.
* Corsi di studio compatibili (BIO, CSE, TLC).
- Citizen science for monitoring climate impacts
Citizen Science for Monitoring Climate Impacts and Achieving Climate Resilience: the thesis will focus on the use of social media and other non-traditional data sources for more effective monitoring of Sustainable Development Goals (SDGs) by citizens. A specific focus will be on new types or unusual occurrence rates of natural events.
* Corsi di studio compatibili (CSE, TLC).
- Scientific models: big data validation, visualization, and cooperative processes. Several theses are available in the area of investigating how, given a large set of experimental data and a set of existing models which should approximate them, it is possible to automatically evaluate the performance of the computational model models. The thesis deals with combustion data, but no preliminary knowledge in chemistry is required. The thesis is in the fields of information systems, data science, collaborative work and machine learning. Paper to read if you are interested in the thesis. The following topics are currently proposed:
- collaboration processes in data sciences
- visualization of groups of experiments in a big data context
Contacts: firstname.lastname@example.org and email@example.com link to reference paper
- Quality analysis of social network information in emergency situations. The thesis will focus on quality aspects of the information gathered from geolocated tweets, and in particular images, in particular analyzing the precision of the image geolocation. Paper to read if you are interested in this topic: link
- thesis topic: improving geoparsing with machine learning techniques
- Benchmarking the benchmarks for big data systems. The thesis will analyze how companies use benchmarking tools, benchmark characteristics, and possible analysis dimensions. Paper to read if you are interested in this thesis: link
- thesis topic: analysis of the characteristics of bid data systems based on fs/QCA techniques and conceptual modeling
- Offerte di tesi di ricerca presso aziende con tirocinio link
If you are interested in some of the topics, look at the suggested reading and then write me for an appointment.
- Information Systems (in Italian) – see information and teaching material on Beep
- Digital Technology (in English) – see information and teaching material on Beep
- Progetto di Ingegneria Informatica – see web site (in AY 2018-19 the course is held by Prof. Zanero)
- Geoinformatics projects see the page on BeeP, with teaching material for registered students. List of available projects
- The role of social media in emergency rapid mapping (1 credit, extracurricular) link
all course information and material is made available on the Beep platform
For Polimi IT Master students: Honours programme “Scientific Research in Information Technology” (link)
For Polimi Master students: Honours programme “Engineering for Sustainable Development” at Politecnico di Milano (link)
IdeaLeague Doctoral School “Engineering Complex Systems with Big data and information Technology ECS-BIT’18” (link)
please send a message to barbara.pernici at polimi.it for an appointment