Computational Characterization of Mental States: A Natural Language Processing Approach

Abstract

Psychiatry is an area of medicine that strongly bases its diagnoses on the psychiatrists subjective appreciation. The task of diagnosis loosely resembles the common pipelines used in supervised learning schema. Therefore, we propose to augment the psychiatrists diagnosis toolbox with an artificial intelligence system based on natural language processing and machine learning algorithms. This approach has been validated in many works in which the performance of the diagnosis has been increased with the use of automatic classification.

Differential 28-Days Cyclic Modulation of Affective Intensity in Female and Male Participants via Social Media

“…Specifically, we aimed to test the hypothesis that the cyclic regulation of the MC leaves an imprint in the linguistic production of females engaged in social media, strong enough to be discriminated from that of matching male participants. Previous studies have used language production to characterize changes in mental state elicited by psychoactive drug intake and psychosis, among others (Bedi et al, 2014, 2015; García et al, 2016; Mota et al, 2016; Carrillo, 2017; Carrillo et al, 2018; Corcoran et al, 2018). Massive textual content in social networks has been used to identify abrupt changes in semantic space of concepts caused by salient events (Carrillo et al, 2015), as a possible indicator of depression using subject's Facebook public information (De Choudhury et al, 2013), or more specifically to characterize and predict postpartum depression (De Choudhury et al, 2014).…”

A computational linguistic measure of affect modulation in the menstrual cycle (Preprint)

“…Specifically, we aim to study whether the affective content of language production varies with the phases of the MC cycle. Previous studies have used language production to characterize changes in mental state elicited by psychoactive drug intake and psychosis, among others [19,20,[39][40][41][42][43]. Massive textual content in social networks have been used to identify abrupt changes in semantic space of concepts caused by salient events [21], as a possible indicator of depression using subject's Facebook public information [22], or more specifically to characterize and predict postpartum depression [23].…”