You are here

Opportunities For Students

 Master/Bachelor Stages and Thesis

We are currently offering several opportunities for undergrad and master thesis as well as stages. The topics are listed and briefly described below. The precise scope and workload of each assignment will be discussed with the students in order to match the effort required for the different types of  work (thesis vs stage, master vs undergrad, and so on).

An overview of topics and challenges addressed in previous internships may be found here.

==================================================================

TITLE: Javascript in the deep [Stage only]

TOPIC DESCRIPTION:
The student will work on the source code of the Signavio-core javascript library [1]: a library specifically created for modeling business processing directly on web pages. Recent updates of web browsers and of the policies of cross-domain interactions led to new versions of the most known javascript frameworks in order to make them compliant to such policies. Unfortunately, the Signavio-core library is not maintained since 2012 and some maintenance is required. During this stage the student will be responsible of: - fixing the library in order to make it compliant with recent web standards and policies; - add some functionalities useful for improving the effectiveness of the process modeling task; - analyze the structure of the library and produce documentation easing the integration of new functionalities.

PREREQUISITES:

  • Knowledge of the Javascript language.
  • Knowledge about how the Canvas object works.

COMPETENCIES TO BE ACQUIRED:

  • Improve javascript competencies.
  • Capacity of analyzing complex source code.
  • Produce effective documentation for supporting further development activities.


[1] https://github.com/dearshor/signavio-core-components

===================================================================

TITLE: Refactor of the MoKi Tool source code

TOPIC DESCRIPTION:
MoKi [1] is a process and knowledge modeling tool, currently maintained by the PDI research unit, as extension of the MediaWiki library [2].
The current integration between MoKi and Mediawiki is quite strong and this dependency limits, sometimes, the possibility of extending or optimizing MoKi functionalities.
The long term goal is to detach many MoKi components from Mediawiki in order to improve the overall efficiency of the tool.
During this stage, that can be extended also for a thesis, the student will be responsible of:

  • analyze the MoKi source code and draw the interaction between all components;
  • understand the points of interaction between MoKi and Mediawiki in order to develop a first abstraction layer;
  • design a proposal for a new architecture;
  • update MoKi components by adapting them to the new architecture;
  • refactor and clean the MoKi code in order to optimize it.

PREREQUISITES:

  • Knowledge of the object-oriented paradigm.
  • Knowledge of the PHP language.
  • Knowledge of the Javascript language.

COMPETENCIES TO BE ACQUIRED:

  • Improve PHP and Javascript competencies.
  • Increase the capability of analyzing complex source code architectures.
  • Improve object-oriented development competences.
  • Improve software design capabilities.


[1] https://moki.fbk.eu/website/index.php
[2] https://www.mediawiki.org/wiki/MediaWiki

===================================================================

TITLE: Deviance Mining [THESIS]

TOPIC DESCRIPTION: 
Deviance mining aims at identifying business process cases that deviate from the standard behavior (e.g., process executions that do not respect a constraint, process executions that are slower than expected, etc.) and at explaining the reasons why those cases present a deviance based on the control flow and the data flow characteristics of the process cases.

Purpose of the work is implementing in a usable tool and experimenting on existing techniques that aim at identifying and explaining deviant behaviors using declarative rules to describe control flow and the data flow characteristics of the process cases. Specifically, the declarative deviance mining techniques will be used for evaluating their capability of discovering declarative process models.

PREREQUISITES:

  • Programming skills (preferably Python)

COMPETENCIES TO BE ACQUIRED

  • Knowledge on deviance mining.
  • Knowledge on declarative process modelling languages (Declare).
  • Knowledge and analysis of ML algorithms.
  • Python state-of-the-art libraries.
  • Experimentation programming skills.
  • Possibly UI programming skills.

=====================================================================================

TITLE: Discovering hybrid process models [THESIS]

TOPIC DESCRIPTION:
Process discovery techniques return process models that are either formal (precisely describing the possible behaviours) or informal (merely a “picture” not allowing for any form of formal reasoning). Hybrid process elements that have formal and informal elements combine the best of the two worlds. Specifically, hybrid Petri nets allow for formal reasoning, but also reveal information that cannot be captured in mainstream formal models. In this thesis an approach for discovering hybrid Petri nets from event logs will be implemented leveraging the PM4Py Python library and an extensive evaluation on the performance of the approach will be carried out. 

PREREQUISITES:

  • Programming skills (preferably in Python)

COMPETENCIES TO BE ACQUIRED:

  • Knowledge on business process modelling languages (Petri Net).
  • Knowledge and analysis of process discovery algorithms.
  • Python state-of-the-art libraries.
  • Experimentation programming skills.

=====================================================================================

TITLE: Investigating variability and event log characteristics [THESIS]

TOPIC DESCRIPTION:
In the process mining field, several metrics have been proposed and investigated that put in relation event logs and process models (e.g., metrics for computing fitness, conformance checking). The metrics specifically focusing on log characteristics (e.g., the event log variability) are instead less consolidated and commonly agreed within the Process Mining (PM) community. Metrics measuring event log characteristics, however, can be very useful in order to detect in advance characteristics of the log for which specific discovery or predictive process monitoring techniques (e.g., declarative versus procedural process discovery, incremental versus non-incremental predictive monitoring techniques) perform well. In this thesis new metrics for measuring different characteristics of event logs will be investigated and possibly used for evaluating the suitability of different process mining techniques for event logs with different characteristics.

PREREQUISITES:

  • Programming skills (preferably in Python)

COMPETENCIES TO BE ACQUIRED:

  • Knowledge on business process modelling languages (Petri Net).
  • Knowledge and analysis of process discovery algorithms.
  • Python state-of-the-art libraries.