[Corpora-List] 2nd Call for Participation: IALP 2016 Shared Task - Dimensional Sentiment Analysis for Chinese Words

Lung-Hao Lee lunghaolee at gmail.com
Mon Jun 20 04:03:20 CEST 2016

(With apologies for cross-posting) ---------------------------------------------------------------------------------------------------------- The 20th International Conference on Asian Language Processing (*IALP 2016*) November 21-23, 2016 at Tainan, Taiwan http://chinese.csie.ncku.edu.tw/IALP2016/ <https://webmail03.yzu.edu.tw/owa/redir.aspx?REF=0UokRqk2oZmJj-riRbvoyMo3JtsPkHzt6uJWkUk2fdBBWOBZGXTTCAFodHRwOi8vY2hpbmVzZS5jc2llLm5ja3UuZWR1LnR3L0lBTFAyMDE2Lw..> ----------------------------------------------------------------------------------------------------------

*Call for Participation*

*IALP 2016 Shared Task:* *Dimensional Sentiment Analysis for Chinese Words* http://nlp.innobic.yzu.edu.tw/tasks/dsa_w/index.html

Sentiment lexicons with valence-arousal ratings are useful resources for the development of dimensional sentiment applications. Due to the limited availability of such VA lexicons, especially for Chinese, the objective of the task is to automatically acquire the valence-arousal ratings of Chinese affective words.

Given a word, participants are asked to provide a real-valued score from 1 to 9 for both valence and arousaldimensions, indicating the degree from most negative to most positive for valence, and from most calm to most excited for arousal. The input format is “word_id, word”, and the output format is “word_id, vallence_rating, arousal_rating”. Below are the input/output formats of the example words 勝利(victory), 痛苦 (pain), 乏味 (tedious), and 放鬆 (relaxed).

- *Example 1*:

Input: 0001, 勝利

Output: 0001, 7.8, 7.2

- *Example 2*:

Input: 0002, 痛苦

Output: 0002, 2.4, 6.8

- *Example 3*:

Input: 0003, 乏味

Output: 0003, 3.4, 3.0

- *Example 4*:

Input: 0004, 放鬆

Output: 0004, 6.2, 2.0

*Data *

- Training: Chinese Valence-Arousal Words (CVAW) contains 1,653

affective words annotated with valence-arousal ratings.

- Test: At least 1,000 Chinese affective words will be manually

annotated with valence-arousal ratings as the gold-standard for testing

system performance. The policy of this shared task is an open test.

Participating systems are allowed to use other publicly available data for

this shared task, but the use of other data should be specified in the

final technical report.


The performance is evaluated by examining the difference between machine-predicted ratings and human-annotated ratings (valence and arousal are treated independently). The evaluation metrics include:

- Mean absolute error

- Pearson correlation coefficient


Participants need to register in order to obtain the training and test data. To register, please send the following information to Lung-Hao Lee ( lhlee at ntnu.edu.tw).

- Team Name

- Organization of your team

- Name and E-mail address of contact person for your team

*Important Dates*

- Registration open: May 5, 2016

- Release of training data: May 5, 2016

- Registration close: July 31, 2016

- Release of test data: August 29, 2016

- Testing results submission due: August 31, 2016

- Release of evaluation results: September 2, 2016

- Paper submission due: September 16, 2016

- Author notification: September 23, 2016

- Camera-ready due: September 30, 2016

- Conference dates: November 21-23, 2016


- Liang-Chih Yu (Yuan Ze University)

- Lung-Hao Lee (National Taiwan Normal University)

- Kam-Fei Wong (The Chinese University of Hong Kong)

-- Lung-Hao Lee (李龍豪), Ph.D. Postdoctoral Fellow Information Technology Center National Taiwan Normal University Email: lhlee at ntnu.edu.tw URL: http://web.ntnu.edu.tw/~lhlee/ -- -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: text/html Size: 7397 bytes Desc: not available URL: <https://mailman.uib.no/public/corpora/attachments/20160620/d19d62b9/attachment.txt>

More information about the Corpora mailing list