%0 Conference Proceedings
%T Between the Lines: Machine Learning for Prediction of Psychological Traits - A Survey
%+ Language Technology Group (LT Group)
%A Johannssen, Dirk
%A Biemann, Chris
%Z Part 2: MAKE-Text
%< avec comité de lecture
%( Lecture Notes in Computer Science
%B 2nd International Cross-Domain Conference for Machine Learning and Knowledge Extraction (CD-MAKE)
%C Hamburg, Germany
%Y Andreas Holzinger
%Y Peter Kieseberg
%Y A Min Tjoa
%Y Edgar Weippl
%I Springer International Publishing
%3 Machine Learning and Knowledge Extraction
%V LNCS-11015
%P 192-211
%8 2018-08-27
%D 2018
%R 10.1007/978-3-319-99740-7_13
%K Computational psychology
%K Machine learning
%K Survey
%K Natural language processing
%Z Computer Science [cs]
%Z Humanities and Social Sciences/Library and information sciencesConference papers
%X A connection between language and psychology of natural language processing for predicting psychological traits (NLPsych) is apparent and holds great potential for accessing the psyche, understand cognitive processes and detect mental health conditions. However, results of works in this field that we call NLPsych could be further improved and is sparse and fragmented, even though approaches and findings often are alike. This survey collects such research and summarizes approaches, data sources, utilized tools and methods, as well as findings. Approaches of included work can roughly be divided into two main strands: word-list-based inquiries and data-driven research. Some findings show that the change of language can indicate the course of mental health diseases, subsequent academic success can be predicted by the use of function words and dream narratives show highly complex cognitive processes – to name but a few. By surveying results of included work, we draw the ‘bigger picture’ that in order to grasp someone’s psyche, it is more important to research how people express themselves rather than what they say, which surfaces in function words. Furthermore, often research unawarely induce biases that worsen results, thus leading to the conclusion that future research should rather focus on data-driven approaches rather than hand-crafted attempts.
%G English
%Z TC 5
%Z TC 8
%Z TC 12
%Z WG 8.4
%Z WG 8.9
%Z WG 12.9
%2 https://inria.hal.science/hal-02060047/document
%2 https://inria.hal.science/hal-02060047/file/472936_1_En_13_Chapter.pdf
%L hal-02060047
%U https://inria.hal.science/hal-02060047
%~ SHS
%~ IFIP-LNCS
%~ IFIP
%~ IFIP-TC
%~ IFIP-TC5
%~ IFIP-WG
%~ IFIP-TC12
%~ IFIP-TC8
%~ IFIP-WG8-4
%~ IFIP-WG8-9
%~ IFIP-CD-MAKE
%~ IFIP-WG12-9
%~ IFIP-LNCS-11015