University of Cambridge > > NLIP Seminar Series > Using NLP and graph theory to capture speech abnormalities in psychosis

Using NLP and graph theory to capture speech abnormalities in psychosis

Add to your list(s) Download to your calendar using vCal

  • UserCaroline Nettekoven (University of Cambridge)
  • ClockFriday 04 March 2022, 12:00-13:00
  • HouseVirtual (Zoom).

If you have a question about this talk, please contact Georgi Karadzhov.

Mapping a psychosis patient’s speech as a network has proved to be a useful way of capturing early speech abnormalities in psychosis. However, to date, speech networks have ignored the semantic content of speech, which is altered in psychosis. We developed an NLP algorithm, “netts”, to map the semantic content of speech as a network, then applied netts to construct semantic speech networks to describe speech in the general population and in a clinical sample. We find that netts captures robust patient-control differences in speech and that these differences have not been described by established NLP measures. Netts could provide a fundamental step forward in deep phenotyping of psychosis and other mental conditions and to support this, we are releasing netts as a free and open-source Python package.

Speaker Bio:

Caroline Nettekoven is a neuroscientist studying how networks of cortical and subcortical brain regions gives rise to complex behaviours, such as language. Caroline received her PhD from the University of Oxford, where she investigated the physiological mechanisms of motor learning. Her thesis combined neuroimaging, brain stimulation and computational modelling techniques to show – for the first time in humans – plastic changes in cerebellar neurochemistry during movement. In her postdoctoral work at the University of Cambridge, Caroline combined NLP and graph theory to develop an algorithm that maps speech content as a network. These networks capture speech abnormalities in schizophrenia and could be used for deeper phenotyping of psychosis and other mental health conditions.

Georgi Karadzhov is inviting you to a scheduled Zoom meeting.

Topic: NLIP Seminar 04.03.2022 Time: Mar 4, 2022 12:00 PM London

Join Zoom Meeting

Meeting ID: 989 0045 5201 Passcode: 762506 One tap mobile +19292056099,,98900455201# US (New York) +12532158782,,98900455201# US (Tacoma)

Dial by your location +1 929 205 6099 US (New York) +1 253 215 8782 US (Tacoma) +1 301 715 8592 US (Washington DC) +1 312 626 6799 US (Chicago) +1 346 248 7799 US (Houston) +1 669 900 6833 US (San Jose) Meeting ID: 989 0045 5201 Find your local number:

This talk is part of the NLIP Seminar Series series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.


© 2006-2022, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity