|
webpage: March 21, 2024
Publications
Interpreting Black-Box Classifiers Using Instance-Level Visual Explanations
Paolo Tamagnini, Josua Krause, Aritra Dasgupta and Enrico Bertini
In Workshop on
Human-In-the-Loop Data Analytics (HILDA), 2017
KNIME Projects
► ...
→ Articles..
→ Applications..
→ Blog Posts..
→ White Papers..
► ...
eXplainable AI: Interpreting Machine Learning with XAI (KNIME Press
Guided Labeling: Human-in-the-Loop Label Generation (KNIME Press
→ Webinars..
→ Talks..
→ Workshops..
► ...
AI Learnathon // Topics: LLMs, Retrieval Augmented Generation (RAG), chatbots // Forum // Press Release
Python Integration // Topics: develop low-code/no-code tools in Python
Pydata, (Berlin - EN) (2022-11-16 ) [Meetup ]
Data Science Dojo, (Online - EN) (2022-10-19 ) [Meetup ]
PyData, (London - EN) (2022-10-03 ) [Meetup ]
Data Science Learnathon // Topics: data prep, modelling, deployment
KNIME Pros Learnathon // Topics: sharing, open source, community
Guided Analytics Learnathon // Topics: autoML, data visualization, visual analytics.
Media
Paolo Tamagnini
Data scientist specialized in guided analytics applications
paolotamag [at] gmail [dot] com
Data scientist with research experience, fond of data visualization and machine learning, currently working for KNIME in Berlin.
► ...
OR..
Data scientist designing tools where the alternation of automated data analysis and interactive on-demand visualization help the user in solving complex problems and detecting hidden patterns/anomalies in the data.
Keywords : guided analytics ; visual analytics ; machine learning ; data science ; big data ; Python ; JavaScript ; d3.js ; KNIME; AWS; ...
In 2018 I moved to Berlin to work as a data scientist in the evangelism team of KNIME . My work at KNIME is variegated, but it initially focused on guided analytics , automated machine learning and eXplainable AI (XAI) . Since 2021 I have been coordinating the Verified Components team of data scientists, releasing data science libraries, functions and tools via reusable and reliable KNIME components. On top of that since 2023 I have been working in the company Educators Alliance to support academics to teach and research with KNIME.
► ...
While working for KNIME I have been travelling to teach the company free and open-source tool for Meetups and business conferences (e.g. ODSC ). Besides that, I have been producing webinars, videos, blog posts and blueprint applications available on KNIME Hub . To see the details of my work inspect the section "KNIME Projects " on the left.
Between November 2017 and February 2018 I worked on an open-source python library called partial_dependence in collaboration with Josua Krause and Enrico Bertini .
► ...
The purpose of this project is to develop another viz tool using a black box approach to recognize sets of instances where the model makes similar decisions.
In October 2017 I graduated at the Master (MSc) of Data Science from Sapienza University of Rome and my thesis is available here .
► ...
I was part of the first graduation in Italy for a Data Science degree. The ceremony had many guests from companies like Google, Unicredit and Eurostat (more info here ). To celebrate this day I wrote with the other graduating students a journal with articles talking about our experiences in the last two years and about the development of our theses.
In August 2016 I wrote my thesis in the research team of Enrico Bertini , associate professor at New York University, Tandon School of Engineering .
► ...
My project was involving the development of a visual analytics tool able to display interactively machine learning explanations.
I published a paper about this project in the workshop HILDA 2017 , which took place in Chicago at SIGMOD 2017 . After graduation I went back to NY to present my thesis at Google (more info here ).
Projects
Rivelo - visual analytics tool for machine learning explanations.partial_dependence - python package for machine learning visualization.
Certifications
|
|
