Mining Facebook Data for Predictive Personality Modeling

the authors highlight difficulties in traditional explanatory research in the field of psychology and argue in favour of novel data-driven science. By applying machine-learning methods to large data sets, predictive power has been shown to increase significantly. Geological engineers are responsible for a wide range of applications, including the design of tunnels, dams, foundations, and mines. While the field of geological engineering stands on solid mechanistic grounds, we argue that its predictive aspect aligns more closely with psychology than other mechanistic sciences. We therefore propose a paradigm shift in geological engineering research towards a prediction-centric approach. Potentially, this could enhance cost-effectiveness in structural design and lead to substantial societal savings.

People use different words when expressing their opinions. Sentiment analysis as a way to automatically detect and categorize people's opinions in text, needs to reflect this diversity and individuality. One possible approach to analyze such traits is to take a person's past opinions into consideration. In practice, such a model can suffer from the data sparsity issue, thus it is difficult to develop. In this article, we take texts from social platforms and propose a preliminary model for evaluating the effectiveness of including user information from the past, and offer a solution for the data sparsity. Furthermore, we present a finer-designed, enhanced model that focuses on frequent users and offers to capture the decay of past opinions using various gaps between the creation time of the text. An attention-based Hawkes process on top of a recurrent neural network is applied for this purpose, and the performance of the model is evaluated with Twitter data. With the proposed framework, positive results are shown which opens up new perspectives for future research.

Personality extraction through social networks is a field that only recently started to capture the attention of researchers. The task consists in, starting with a corpus of user profiles on a particular social network, classifying their personalities correctly, according to a specific personality model as described in psychology. In this master thesis, three innovations to the domain are presented. Firstly, the collection of a corpus of LinkedIn users. Secondly, the extraction of the personality according to two personality models, DiSC and MBTI, the extraction with DiSC having never been done before. Lastly, the idea of going from one personality model to the other is explored, thus creating the possibility of having the results on two personality models with only one personality test.