Predicting psychological health from childhood essays with convolutional neural networks for the CLPsych 2018 shared task (Team UKNLP)

Anthony Rios; Tung Tran; Ramakanth Kavuluru

Predicting psychological health from childhood essays with convolutional neural networks for the CLPsych 2018 shared task (Team UKNLP)

Anthony Rios, Tung Tran, Ramakanth Kavuluru

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

This paper describes the systems we developed for tasks A and B of the 2018 CLPsych shared task. The first task (task A) focuses on predicting behavioral health scores at age 11 using childhood essays. The second task (task B) asks participants to predict future psychological distress at ages 23, 33, 42, and 50 using the age 11 essays. We propose two convolutional neural network based methods that map each task to a regression problem. Among seven teams we ranked third on task A with disattenuated Pearson correlation (DPC) score of 0.5587. Likewise, we ranked third on task B with an average DPC score of 0.3062.

Original language	English
Title of host publication	Proceedings of the 5th Workshop on Computational Linguistics and Clinical Psychology
Subtitle of host publication	From Keyboard to Clinic, CLPsych 2018 at the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HTL 2018
Editors	Kate Loveys, Kate Niederhoffer, Emily Prud�hommeaux, Rebecca Resnik, Philip Resnik
Pages	107-112
Number of pages	6
ISBN (Electronic)	9781948087124
State	Published - 2018
Event	5th Workshop on Computational Linguistics and Clinical Psychology: From Keyboard to Clinic, CLPsych 2018 at the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HTL 2018 - New Orleans, United States Duration: Jun 5 2018 → …

Publication series

Name	Proceedings of the 5th Workshop on Computational Linguistics and Clinical Psychology: From Keyboard to Clinic, CLPsych 2018 at the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HTL 2018

Conference

Conference	5th Workshop on Computational Linguistics and Clinical Psychology: From Keyboard to Clinic, CLPsych 2018 at the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HTL 2018
Country/Territory	United States
City	New Orleans
Period	6/5/18 → …

Bibliographical note

Publisher Copyright:
© 2018 Association for Computational Linguistics

Funding

We thank anonymous reviewers for their constructive comments despite the short turnaround. This research is supported by the U.S. National Library of Medicine through grant R21LM012274. We also gratefully acknowledge the support of the NVIDIA Corporation for its donation of the Titan X Pascal GPU used for this research.

Funders	Funder number
U.S. National Library of Medicine	R21LM012274
Nvidia

ASJC Scopus subject areas

Language and Linguistics
Linguistics and Language
Computer Science Applications
Artificial Intelligence

Cite this

Rios, A., Tran, T., & Kavuluru, R. (2018). Predicting psychological health from childhood essays with convolutional neural networks for the CLPsych 2018 shared task (Team UKNLP). In K. Loveys, K. Niederhoffer, E. Prud�hommeaux, R. Resnik, & P. Resnik (Eds.), Proceedings of the 5th Workshop on Computational Linguistics and Clinical Psychology: From Keyboard to Clinic, CLPsych 2018 at the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HTL 2018 (pp. 107-112). (Proceedings of the 5th Workshop on Computational Linguistics and Clinical Psychology: From Keyboard to Clinic, CLPsych 2018 at the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HTL 2018).

Rios, Anthony ; Tran, Tung ; Kavuluru, Ramakanth. / Predicting psychological health from childhood essays with convolutional neural networks for the CLPsych 2018 shared task (Team UKNLP). Proceedings of the 5th Workshop on Computational Linguistics and Clinical Psychology: From Keyboard to Clinic, CLPsych 2018 at the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HTL 2018. editor / Kate Loveys ; Kate Niederhoffer ; Emily Prud�hommeaux ; Rebecca Resnik ; Philip Resnik. 2018. pp. 107-112 (Proceedings of the 5th Workshop on Computational Linguistics and Clinical Psychology: From Keyboard to Clinic, CLPsych 2018 at the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HTL 2018).

@inproceedings{bf46fdf188f44d0d89f8989012ed0704,

title = "Predicting psychological health from childhood essays with convolutional neural networks for the CLPsych 2018 shared task (Team UKNLP)",

abstract = "This paper describes the systems we developed for tasks A and B of the 2018 CLPsych shared task. The first task (task A) focuses on predicting behavioral health scores at age 11 using childhood essays. The second task (task B) asks participants to predict future psychological distress at ages 23, 33, 42, and 50 using the age 11 essays. We propose two convolutional neural network based methods that map each task to a regression problem. Among seven teams we ranked third on task A with disattenuated Pearson correlation (DPC) score of 0.5587. Likewise, we ranked third on task B with an average DPC score of 0.3062.",

author = "Anthony Rios and Tung Tran and Ramakanth Kavuluru",

note = "Publisher Copyright: {\textcopyright} 2018 Association for Computational Linguistics; 5th Workshop on Computational Linguistics and Clinical Psychology: From Keyboard to Clinic, CLPsych 2018 at the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HTL 2018 ; Conference date: 05-06-2018",

year = "2018",

language = "English",

series = "Proceedings of the 5th Workshop on Computational Linguistics and Clinical Psychology: From Keyboard to Clinic, CLPsych 2018 at the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HTL 2018",

pages = "107--112",

editor = "Kate Loveys and Kate Niederhoffer and Emily Prud�hommeaux and Rebecca Resnik and Philip Resnik",

booktitle = "Proceedings of the 5th Workshop on Computational Linguistics and Clinical Psychology",

}

Rios, A, Tran, T & Kavuluru, R 2018, Predicting psychological health from childhood essays with convolutional neural networks for the CLPsych 2018 shared task (Team UKNLP). in K Loveys, K Niederhoffer, E Prud�hommeaux, R Resnik & P Resnik (eds), Proceedings of the 5th Workshop on Computational Linguistics and Clinical Psychology: From Keyboard to Clinic, CLPsych 2018 at the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HTL 2018. Proceedings of the 5th Workshop on Computational Linguistics and Clinical Psychology: From Keyboard to Clinic, CLPsych 2018 at the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HTL 2018, pp. 107-112, 5th Workshop on Computational Linguistics and Clinical Psychology: From Keyboard to Clinic, CLPsych 2018 at the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HTL 2018, New Orleans, United States, 6/5/18.

Predicting psychological health from childhood essays with convolutional neural networks for the CLPsych 2018 shared task (Team UKNLP). / Rios, Anthony; Tran, Tung; Kavuluru, Ramakanth.
Proceedings of the 5th Workshop on Computational Linguistics and Clinical Psychology: From Keyboard to Clinic, CLPsych 2018 at the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HTL 2018. ed. / Kate Loveys; Kate Niederhoffer; Emily Prud�hommeaux; Rebecca Resnik; Philip Resnik. 2018. p. 107-112 (Proceedings of the 5th Workshop on Computational Linguistics and Clinical Psychology: From Keyboard to Clinic, CLPsych 2018 at the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HTL 2018).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Predicting psychological health from childhood essays with convolutional neural networks for the CLPsych 2018 shared task (Team UKNLP)

AU - Rios, Anthony

AU - Tran, Tung

AU - Kavuluru, Ramakanth

PY - 2018

Y1 - 2018

N2 - This paper describes the systems we developed for tasks A and B of the 2018 CLPsych shared task. The first task (task A) focuses on predicting behavioral health scores at age 11 using childhood essays. The second task (task B) asks participants to predict future psychological distress at ages 23, 33, 42, and 50 using the age 11 essays. We propose two convolutional neural network based methods that map each task to a regression problem. Among seven teams we ranked third on task A with disattenuated Pearson correlation (DPC) score of 0.5587. Likewise, we ranked third on task B with an average DPC score of 0.3062.

AB - This paper describes the systems we developed for tasks A and B of the 2018 CLPsych shared task. The first task (task A) focuses on predicting behavioral health scores at age 11 using childhood essays. The second task (task B) asks participants to predict future psychological distress at ages 23, 33, 42, and 50 using the age 11 essays. We propose two convolutional neural network based methods that map each task to a regression problem. Among seven teams we ranked third on task A with disattenuated Pearson correlation (DPC) score of 0.5587. Likewise, we ranked third on task B with an average DPC score of 0.3062.

UR - http://www.scopus.com/inward/record.url?scp=85119989461&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85119989461&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:85119989461

T3 - Proceedings of the 5th Workshop on Computational Linguistics and Clinical Psychology: From Keyboard to Clinic, CLPsych 2018 at the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HTL 2018

SP - 107

EP - 112

BT - Proceedings of the 5th Workshop on Computational Linguistics and Clinical Psychology

A2 - Loveys, Kate

A2 - Niederhoffer, Kate

A2 - Prud�hommeaux, Emily

A2 - Resnik, Rebecca

A2 - Resnik, Philip

T2 - 5th Workshop on Computational Linguistics and Clinical Psychology: From Keyboard to Clinic, CLPsych 2018 at the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HTL 2018

Y2 - 5 June 2018

ER -

Rios A, Tran T, Kavuluru R. Predicting psychological health from childhood essays with convolutional neural networks for the CLPsych 2018 shared task (Team UKNLP). In Loveys K, Niederhoffer K, Prud�hommeaux E, Resnik R, Resnik P, editors, Proceedings of the 5th Workshop on Computational Linguistics and Clinical Psychology: From Keyboard to Clinic, CLPsych 2018 at the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HTL 2018. 2018. p. 107-112. (Proceedings of the 5th Workshop on Computational Linguistics and Clinical Psychology: From Keyboard to Clinic, CLPsych 2018 at the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HTL 2018).

Predicting psychological health from childhood essays with convolutional neural networks for the CLPsych 2018 shared task (Team UKNLP)

Abstract

Publication series

Conference

Bibliographical note

Funding

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this