Abstract
Psychologically motivated, lexicon-based text analysis methods such as LIWC (Pennebaker et al., 2015) have been criticized by compu- tational linguists for their lack of adaptabil- ity, but they have not often been systemati- cally compared with either human evaluations or machine learning approaches. The goal of the current study was to assess the effective- ness and predictive ability of LIWC on a rela- tionship goal classification task. In this paper, we compared the outcomes of (1) LIWC, (2) machine learning, and (3) a human baseline. A newly collected corpus of online dating pro- file texts (a genre not explored before in the ACL anthology) was used, accompanied by the profile writers’ self-selected relationship goal (long-term versus date). These three ap- proaches were tested by comparing their per- formance on identifying both the intended re- lationship goal and content-related text labels. Results show that LIWC and machine learning models both correlate with humans in terms of content-related label assignment. Further- more, LIWC’s content-related labels corre- sponded more strongly to humans than those of the machine learning model. Moreover, all approaches were similarly accurate in predict- ing the relationship goal.
Original language | English |
---|---|
Title of host publication | Proceedings of the 5th Workshop on Noisy User-generated Text (W-NUT 2019) |
Publisher | Association for Computational Linguistics |
Pages | 94-100 |
Number of pages | 7 |
DOIs | |
Publication status | Published - 2019 |
Externally published | Yes |
Event | 2019 The 5th Workshop on Noisy User-generated Text (W-NUT) - Hong Kong Duration: 4 Nov 2019 → … http://noisy-text.github.io/2019/ |
Workshop
Workshop | 2019 The 5th Workshop on Noisy User-generated Text (W-NUT) |
---|---|
Period | 4/11/19 → … |
Internet address |
Access to Document
Fingerprint
Dive into the research topics of 'Automatic identification of writers’ intentions: Comparing different methods for predicting relationship goals in online dating profile texts'. Together they form a unique fingerprint.
View full fingerprint
Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver
van der Lee, C., van der Zanden, T., Krahmer, E., Mos, M., & Schouten, A. (2019). Automatic identification of writers’ intentions: Comparing different methods for predicting relationship goals in online dating profile texts. In Proceedings of the 5th Workshop on Noisy User-generated Text (W-NUT 2019) (pp. 94-100). Association for Computational Linguistics. https://doi.org/10.18653/v1/D19-5512
van der Lee, Chris ; van der Zanden, Tess ; Krahmer, Emiel et al. / Automatic identification of writers’ intentions: Comparing different methods for predicting relationship goals in online dating profile texts. Proceedings of the 5th Workshop on Noisy User-generated Text (W-NUT 2019). Association for Computational Linguistics, 2019. pp. 94-100
@inproceedings{30d6b91563c845e5ae121a0c529dc0b2,
title = "Automatic identification of writers{\textquoteright} intentions: Comparing different methods for predicting relationship goals in online dating profile texts",
abstract = "Psychologically motivated, lexicon-based text analysis methods such as LIWC (Pennebaker et al., 2015) have been criticized by compu- tational linguists for their lack of adaptabil- ity, but they have not often been systemati- cally compared with either human evaluations or machine learning approaches. The goal of the current study was to assess the effective- ness and predictive ability of LIWC on a rela- tionship goal classification task. In this paper, we compared the outcomes of (1) LIWC, (2) machine learning, and (3) a human baseline. A newly collected corpus of online dating pro- file texts (a genre not explored before in the ACL anthology) was used, accompanied by the profile writers{\textquoteright} self-selected relationship goal (long-term versus date). These three ap- proaches were tested by comparing their per- formance on identifying both the intended re- lationship goal and content-related text labels. Results show that LIWC and machine learning models both correlate with humans in terms of content-related label assignment. Further- more, LIWC{\textquoteright}s content-related labels corre- sponded more strongly to humans than those of the machine learning model. Moreover, all approaches were similarly accurate in predict- ing the relationship goal.",
author = "{van der Lee}, Chris and {van der Zanden}, Tess and Emiel Krahmer and Maria Mos and Alexander Schouten",
year = "2019",
doi = "10.18653/v1/D19-5512",
language = "English",
pages = "94--100",
booktitle = "Proceedings of the 5th Workshop on Noisy User-generated Text (W-NUT 2019)",
publisher = "Association for Computational Linguistics",
note = "2019 The 5th Workshop on Noisy User-generated Text (W-NUT) ; Conference date: 04-11-2019",
url = "http://noisy-text.github.io/2019/",
}
van der Lee, C, van der Zanden, T, Krahmer, E, Mos, M & Schouten, A 2019, Automatic identification of writers’ intentions: Comparing different methods for predicting relationship goals in online dating profile texts. in Proceedings of the 5th Workshop on Noisy User-generated Text (W-NUT 2019). Association for Computational Linguistics, pp. 94-100, 2019 The 5th Workshop on Noisy User-generated Text (W-NUT), 4/11/19. https://doi.org/10.18653/v1/D19-5512
Automatic identification of writers’ intentions: Comparing different methods for predicting relationship goals in online dating profile texts. / van der Lee, Chris; van der Zanden, Tess; Krahmer, Emiel et al.
Proceedings of the 5th Workshop on Noisy User-generated Text (W-NUT 2019). Association for Computational Linguistics, 2019. p. 94-100.
Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review
TY - GEN
T1 - Automatic identification of writers’ intentions: Comparing different methods for predicting relationship goals in online dating profile texts
AU - van der Lee, Chris
AU - van der Zanden, Tess
AU - Krahmer, Emiel
AU - Mos, Maria
AU - Schouten, Alexander
PY - 2019
Y1 - 2019
N2 - Psychologically motivated, lexicon-based text analysis methods such as LIWC (Pennebaker et al., 2015) have been criticized by compu- tational linguists for their lack of adaptabil- ity, but they have not often been systemati- cally compared with either human evaluations or machine learning approaches. The goal of the current study was to assess the effective- ness and predictive ability of LIWC on a rela- tionship goal classification task. In this paper, we compared the outcomes of (1) LIWC, (2) machine learning, and (3) a human baseline. A newly collected corpus of online dating pro- file texts (a genre not explored before in the ACL anthology) was used, accompanied by the profile writers’ self-selected relationship goal (long-term versus date). These three ap- proaches were tested by comparing their per- formance on identifying both the intended re- lationship goal and content-related text labels. Results show that LIWC and machine learning models both correlate with humans in terms of content-related label assignment. Further- more, LIWC’s content-related labels corre- sponded more strongly to humans than those of the machine learning model. Moreover, all approaches were similarly accurate in predict- ing the relationship goal.
AB - Psychologically motivated, lexicon-based text analysis methods such as LIWC (Pennebaker et al., 2015) have been criticized by compu- tational linguists for their lack of adaptabil- ity, but they have not often been systemati- cally compared with either human evaluations or machine learning approaches. The goal of the current study was to assess the effective- ness and predictive ability of LIWC on a rela- tionship goal classification task. In this paper, we compared the outcomes of (1) LIWC, (2) machine learning, and (3) a human baseline. A newly collected corpus of online dating pro- file texts (a genre not explored before in the ACL anthology) was used, accompanied by the profile writers’ self-selected relationship goal (long-term versus date). These three ap- proaches were tested by comparing their per- formance on identifying both the intended re- lationship goal and content-related text labels. Results show that LIWC and machine learning models both correlate with humans in terms of content-related label assignment. Further- more, LIWC’s content-related labels corre- sponded more strongly to humans than those of the machine learning model. Moreover, all approaches were similarly accurate in predict- ing the relationship goal.
U2 - 10.18653/v1/D19-5512
DO - 10.18653/v1/D19-5512
M3 - Conference contribution
SP - 94
EP - 100
BT - Proceedings of the 5th Workshop on Noisy User-generated Text (W-NUT 2019)
PB - Association for Computational Linguistics
T2 - 2019 The 5th Workshop on Noisy User-generated Text (W-NUT)
Y2 - 4 November 2019
ER -
van der Lee C, van der Zanden T, Krahmer E, Mos M, Schouten A. Automatic identification of writers’ intentions: Comparing different methods for predicting relationship goals in online dating profile texts. In Proceedings of the 5th Workshop on Noisy User-generated Text (W-NUT 2019). Association for Computational Linguistics. 2019. p. 94-100 doi: 10.18653/v1/D19-5512