Personal/national mood tracker

From Computer Laboratory Group Design Projects
Revision as of 06:55, 10 September 2012 by afb21 (talk | contribs) (Created page with " I have received the following proposals from Tomas Cervenka at VisualDNA. Tomas is the contact person (tomas@visualdna.com) Predicting personality and detecting cheaters At V...")
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigationJump to search

I have received the following proposals from Tomas Cervenka at VisualDNA.

Tomas is the contact person (tomas@visualdna.com)

Predicting personality and detecting cheaters

At VisualDNA we focus on profiling online users using our patented visual quizzes. Users are presented with a set of questions and have to answer by clicking one or more images. You will be given an extensive dataset and asked to build a model and a visual tool to predict user's answers based on answers he has already given in the quiz. The ultimate goal is to algorithmically predict, whether user's answers are genuine or user was trying to 'cheat' the quiz. If you can beat our existing model, there's an extra prize involved!

Inferring personality from behaviour

"Tell me what you eat, and I will tell you what you are" is an old quote - at VisualDNA we have a different one: "Tell me what sites you visit and I will tell you what you are". You will be given an extensive anonymised dataset of user's browsing history (no dodgy sites included, trust us) and asked to build a model and a tool to predict, based on asking users for a set of sites / pages they have recently visited, their demographic, socioeconomic and personality traits. If you can beat our existing model, there's an extra prize involved!