Unpredictable Attributes in Market Comment Generation

Yumi Hamazono, Tatsuya Ishigaki, Yusuke Miyao, Hiroya Takamura, Ichiro Kobayashi
2021 Pacific Asia Conference on Language, Information and Computation  
There are two types of datasets for data-totext: one uses raw data obtained in the real world, and the other is constructed artificially for a controlled task. A straightforwardly output text is generated from its paired input data for a manually constructed dataset because the dataset is well constructed without any excess or deficiencies. However, it may not be possible to generate a correct output text from the input data for a dataset constructed with realworld data and text. In such cases,
more » ... we have to provide additional data, for example, data or text attribute labels, in order to generate the expected output text from the paired input. This paper discusses the importance of additional input labels in data-to-text for real-world data. The content and style of a market comment change depending on its medium, the market situation, and the time of day. However, as the stock price, which is the input data, does not contain any such aforementioned information, it cannot generate comments appropriately from the data alone. Therefore, we analyse the dataset and provide additional labels which are unpredictable with input data for the appropriate parts in the model. Thus, the accuracy of sentence generation is greatly improved compared to the case without the labels.The result suggests unpredictable attributes should be given as a part of the input in the training of the text generating model.
dblp:conf/paclic/HamazonoIMTK21 fatcat:olmxk4egbvhohflemr3j63tyqu