Contributor
Vasilis Mageirakos

Support for Rucio Users with Natural Language Processing


Mentors
Martin Barisits, Mario Lassnig, Tomas Javurek
Organization
CERN-HSF

The aim of the project is to use Natural Language Processing (NLP) to develop an intelligent bot able to provide satisfying answers to users and handle support requests up to a certain level of complexity, forwarding only the remaining ones to the experts.

For this purpose, the support requests sent as emails to the support team need to be downloaded through IMAP client and then parsed and transformed into a text corpus for further analysis.

Addional data sources besides support emails will be used as input such as Rucio documentation pages and GitHub issues.

Python 3.x will be the programming language used, preferably with NLTK library for the implementation of text mining algorithms.

The key deliverables in the final codebase include fetcing and parsing of the input data, data storage, as well as machine learning and information retrieval and extraction methods. All applied to the task of supporting Rucio User requests.