Szymon Bobek

This page is read only. You can view the source, but not change it. Ask your administrator if you think this is wrong.
====== Research interests ======
My current reaserch interests are: machine learning, mobile context-aware systems and knowledge engineering.
===== Projects involvement =====
  * PEER -- [[https://peer-ai.eu/en/outcomes/|Redefining
Human-AI Collaboration
for Complex Decisions]]
  * XPM --[[https://www.chistera.eu/projects/xpm|Explainable Predictive Maintenance]]
  * PACMEL -- [[https://www.chistera.eu/projects/pacmel|Process-aware Analytics Support based on Conceptual Models for Event Logs]]
  * KnowMe -- [[http://glados.kis.agh.edu.pl|Uncertain and incomplete knowledge modelling for mediation in mobile context-aware systems]] (NCN)
  * Prosecco -- [[https://www.geist.re/pub:projects:prosecco:start|Processes Semantics Collaboration for Companies]] (NCBiR)
  * BIMLOQ -- [[https://www.geist.re/pub:projects:bimloq:start|Business Models Optimization for Quality ]]
  * REBIT -- [[http://kis.zarz.agh.edu.pl/rebit-flagowy-projekt-katedry/|System zarządzania regułami biznesowymi i technologicznymi]]
  * HeKatE -- [[https://www.geist.re/pub:projects:hekate:start|Hybrid Knowledge Engineering]]
===== Public Datasets Creation=====
  * [[https://doi.org/10.5281/zenodo.11448395|XAI-FUNGI]] Dataset from the user study on comprehensibility of XAI algorithms
  * [[https://doi.org/10.5281/zenodo.11469702| TCM]] Benchmark Datasets for Predictive Maintenance in Steel Manufacturing
===== Tools development =====
==== Explainable Artificial Intelligence (XAI) Tools ====
   * **[[https://github.com/sbobek/tsproto|TSProto]]** is post-host prototype-based explanations with rules for time-series classifiers. 
   * **[[https://github.com/sbobek/lux|LUX]]** is a Python library for generating uncertain local explanations in a form of human understandable  (and executable) rules.
   * **[[https://github.com/sbobek/knac|KnAC]]** is a Python library for Knowledge Augmented Clustering. It allows for human-in-the-loop conformance checking between domain expert and automatic clustering.
   * **[[https://github.com/sbobek/clamp|CLAMP]]** ia a Python library for Cluster analysis with multidimensional bounding boxes. It allows for in-depth cluster analysis with state-of-the-art XAI algorithms.
   * **[[https://github.com/sbobek/inxai|InXAI]]** is a Python library for Intelligible XAI that allows to evaluate, compare and analyse different explainable AI algorithms
==== Knowledge engineering Tools ====


{{ ::heart-logo-no-borders.png?150|}}
=== HeaRTDroid ===
HeaRTDroid is a rule-based inference engine for Android mobile devices. It is distributed under the GNU General Public License.
For more details see [[http://glados.kis.agh.edu.pl/doku.php?id=pub:software:heartdroid:start|HeaRTDroid]] webpage.

Key features are:
  * Inference support for **XTT2 rule representation** that is based on the **Attributive Logic with Set Values**
  * Pure Java implementation that allows for **integration with any Java code**, including Android applications
  * **Callbacks mechanism** based on Java reflection, that allows for easy integration with other systems
  * **Uncertainty** management mechanism based on modified certainty factors algebra and probabilistic approach
  * Capturing dynamics with **statistical functions** and **time parametrised operators**
  * **Different inference modes** and real time **reasoning triggers**
  * Under development: **Server mode** with TCP/IP protocol and REST API.
{{ ::hwed-logo.png?150|}}
=== HWEd ===


HWEd (HeaKatE Web Editor) is online web editor for creating XTT2 models, that are run by HeaRTDroid inference engine. For more information see the project [[http://glados.kis.agh.edu.pl/doku.php?id=pub:software:hwed:start|website]].

  * Free online access. You can use our online version of the editor that is available here: [[http://glados.kis.agh.edu.pl/hwed/| HWEd]]
  * For more information on how to set up your own HWEd see: [[https://bitbucket.org/sbobek/hwed| Project website]]
=== ContextViewer ===
Context Viewer is a tool that allows to visualise, and browse data generated by the AWARE framework with an Internet browser. 

Form more information see the project [[http://glados.kis.agh.edu.pl/doku.php?id=pub:software:contextviewer:start|website]].