• Research and creation of AI models and methods
• Acquisition, cleaning, and preprocessing of data for statistical analysis and machine learning
• Research of methods for data annotation and organization of annotation activities and for supporting fact-checkers
• Application of models on practical problem solving (based on actual projects)
• Experimental software development
• Design and execution of studies

• Master degree in any of (or equivalent):

1. Artificial Intelligence, Data Science
2. Computer Science
3. Software Engineering

• PhD or minimum 2 years of experience in research in AI and/or statistical analysis and/or automatic data collection
• Machine learning (and artificial intelligence in general)
• Data cleaning, pre-processing, and analysis
• Statistical analysis
• Preferred technologies: Python, Pandas, scikit-learn, PyTorch, Keras
• Skills that are nice to have:
• Automatic data collection from the web (scraping, APIs)
• Deep learning
• Software development and deployment (including machine-learned models)

We seek reinforcements for our team solving the international project CEDMO (Central European Digital Media Observatory). The goal of the project is to support the work of fact-checkers through research and development of AI-based tools. The interdisciplinary research cooperation (fact-checker, journalists, social and computer scientists) offers unique opportunities with real-life impacts. In this way, we aim at supporting the democracy processes and reducing one of the major threats of nowadays.

The tasks ahead us include:

• content collection (especially from social media),

• detection of potential disinformation,

• seeking of relevant reliable factual information sources

• seeking potential fact-checking duplicities

• detection of disinformation narratives and campaigns

• user (social medium actor) behavior analysis
  Within the broad field of AI, our research methods are usually close to at least some of the following:

• Limited (labeled) data learning (e.g., semi-supervised learning, transfer learning, weak supervision learning)

• Human-in-the-loop techniques

• Transparent, interpretable and explainable AI (and AI ethics in general)
  Beyond the CEDMO project, you will have the opportunity to participate in further projects of similar character. The topic of disinformation tackling is one of KInIT’s core application domains. Recently, KInIt has been very active in seeking further grant projects, especially within the Horizon Europe programme, where we submitted multiple further proposals as members of consortia with first-class institutions across Europe.