I am currently working on these projects:
Digital Language Typology
Producing a platform to assess the structurally manifested family relationships within a set of languages with large digital textual and speech material. We focus in particular on low-resourced Uralic languages.
Building a demonstration of a future news ecosystem, using automatic data analysis and natural language generation.
Previous project: WHIM
In Computational Creativity research, we study how to engineer software which can take on some of the creative responsibility in arts and science projects. There has been much progress towards the creative generation of artefacts of cultural value such as poems, music and paintings. Often, when produced by people, such artefacts embed a fictional idea invented by the creator. There have only been a few small, ad-hoc studies of how to automate this process.
The time is therefore ripe to see whether we can derive, implement and test novel formalisms and processes which enable software to not only invent, but assess, explore and present such ideas.
We propose to investigate the following model for creative idea generation:
- Collect and analyse some information about a domain, to form a shallow world view of that domain;
- Form a set of what-if style ideas from the analysis using notions of surprise, semantic tension and incongruity;
- Assess, rank and select ideas based on the quality and quantity of narratives that can be generated using each idea;
- Use the world view, idea and narratives in linguistic renderings, taking into account notions of relevancy, expansion, obfuscation and affect.
We will also collect crowd-sourced data about how people value, appreciate and expand these ideas, and will machine learn predictors for how people will react to automatically generated ideas.
Extract from project objective.
For more about WHIM, see:
Computational Creativity is rather a young subfield of Artificial Intelligence. For an introduction to the field, see Tony Veale's free online books: Robot Comix.
Or, for an introduction with a slight leaning towards Cognitive Science, take a look at my talk from KVIT 2015.