2026-10-13T11:26:05Z https://www.scidb.cn/oai

10.11922/sciencedb.j00104.00029 2020-10-16T13:34:39Z

2020-10-16 An Evaluation of Chinese Human-Computer Dialogue Technology doi:10.11922/sciencedb.j00104.00029 en Seven tables of this paper. Table 1 shows an example of air ticket booking. Table 2 presents some examples in training set of Task 1. Table 3 shows statistics of the intent data set in Task 1. Table 4 presents the top 8 teams of Task 1 ranked by F1 score. Table 5 shows the top 5 teams of Task 2. Note: C denotes task completion ratio, T denotes average dialogue turns, Sa denotes user satisfaction score, F denotes fluency degree of response, and G denotes uncovered data guidance capability. All these indicators are average scores of all test cases. Table A1 is the complete leaderboard of Task 1 ranking by F1 score. Table A2 shows the complete leaderboard of Task 2. Note: C denotes task completion ratio, T denotes average dialogue turns, Sa denotes user satisfaction score, F denotes fluency degree of response, and G denotes uncovered data guidance capability. All these indicators are average scores of all test cases. Chinese human-computer dialogue evaluation; Evaluation data; User intent classification; Taskoriented dialogue Zhengyu Zhao Weinan Zhang Wanxiang Che Zhigang Chen Yibo Zhang PUBLIC https://creativecommons.org/licenses/by/4.0/ dataset http://www.doi.org/10.1162/dint_a_00007 Science Data Bank