Mind-Reader (Psychometric Sentiment Analysis)
Cora & Cailean
The art of measuring personality is notoriously tricky. Development of accurate personality models is hindered by the quality of the questions in psychometric tests themselves. Although these questions aim to capture aspects of the test-taker's personality, test-takers often interpret the same questions differently depending on mood, context, or personal background. Moreover, multiple-choice questions do not reflect the full scale of human variability.
Rather than measuring personality with multiple-choice questions, we suggest measuring personality through semantic analysis. Our proposed psychometric system will consist of a personality modelling system for the Big 5 Personality types, each of which will include a set of words labelled with semantic values. We believe Haskell's dynamic type functionality will be especially useful in designing functions for each of the 5 personality dimensions.
We would like to adapt our psychometric semantic analyzer to user input. We could present the users with dichotomous sets of words encoded for the Big 5 personality dimensions and give them the option to choose which word they most identify with, then encode their choices for personality values. As our project grows, we would also add a feature to include text input from the user, allowing for sentences and even paragraphs to be encoded according to their Big 5 values. In the future, similar technology could be used to analyze personality from someone's Facebook or text message history.
What did we learn from doing this? (This should be written after you have done the work.) What is the bottom-line? Is logic programming suitable for (part-of) the task? Make sure you include the evidence for your claims.