中文摘要 |
當訓練資料有限時,如何應用已標記的訓練資料,幫助目標任務的模型快速建構,是遷移式學習(Transfer Learning)的重要議題。在本論文中,以多任務學習(Multt-task Learning)的方式進行中文歌手命名實體辨識(Name Entity Recognition, NER)和基於面向的情感分析(Aspect-Based Sentiment Analysis,ABSA)的任務。我們應用參數生成網路(Jia et al., 2019)結合梯度反轉層(Gradient Adversarial Layer, GRL)(Ganin and Lempitsky, 2015)架構來建立模型,並且使用Tie/Break規則進行標記,動態調節權重的機制(Dynamic Weight Average, DWA)(Liu et al., 2019),依據每個任務的損失變化率來調整任務權重。實驗結果顯示,我們的擴展參數生成網路模型(Extended Parameter Generation Network, EPGN),在僅考慮NER任務時,F1可以達到90%,和IBHB效能86%相比,有所改善,加入ABSA任務後,平均F1能夠達到78%,和IBHB(Chiu, 2020)效能相差了22%,明顯的大幅成長。 |
英文摘要 |
When we are interested in a certain domain, we can collect and analyze data from the Internet. The newly collected data is not labeled, so the use of labeled data is hoped to be helpful to the new data. We perform name entity recognition (NER) and aspect-based sentiment analysis (ABSA) in multi-task learning, and combine parameter generation network and DANN architecture to build the model. In the NER task, the data is labeled with Tie, Break, and the task weight is adjusted according to the loss change rate of each task using Dynamic Weight Average (DWA). This study used two different source domain data sets. The experimental results show that Tie, Break can improve the results of the model; DWA can have better performance in the results; the combination of parameter generation network and gradient reversal layer can be used for every good learning in different domain. |