Knowledge graph embedding aims at embedding entities and relations in a knowledge graph into a continuous, dense, low-dimensional and realvalued vector space. Among various embedding models appeared in recent years, t...Knowledge graph embedding aims at embedding entities and relations in a knowledge graph into a continuous, dense, low-dimensional and realvalued vector space. Among various embedding models appeared in recent years, translation-based models such as TransE, TransH and TransR achieve state-of-the-art performance. However, in these models, negative triples used for training phase are generated by replacing each positive entity in positive triples with negative entities from the entity set with the same probability;as a result, a large number of invalid negative triples will be generated and used in the training process. In this paper, a method named adaptive negative sampling (ANS) is proposed to generate valid negative triples. In this method, it first divided all the entities into a number of groups which consist of similar entities by some clustering algorithms such as K-Means. Then, corresponding to each positive triple, the head entity was replaced by a negative entity from the cluster in which the head entity was located and the tail entity was replaced in a similar approach. As a result, it generated a set of high-quality negative triples which benefit for improving the effectiveness of embedding models. The ANS method was combined with the TransE model and the resulted model was named as TransE-ANS. Experimental results show that TransE-ANS achieves significant improvement in the link prediction task.展开更多
This study is undertaken to find out writers' attention patterns in these two different writing methods (the direct method and the translation-based method) during the writing process through the think-aloud metho...This study is undertaken to find out writers' attention patterns in these two different writing methods (the direct method and the translation-based method) during the writing process through the think-aloud method and EFL learners' perception of the advantages and disadvantages of those two writing methods.The results show that the subjects' attention patterns in the direct method and in the translation-based method are quite different and the differences are mainly embodied in the linguistic level attention and personal comment.As far as the subjects are concerned,writing directly in English is easier and faster than writing through translation,and the direct writing method often helps them learn English language and forces them to focus on English expression.In contrast,for the translation-based writing method,the subjects as a whole relate that they have a wide range of vocabulary and expressions,have a greater number of ideas,and can think through ideas clearly.展开更多
This study attempts to explore the relationship between translation and English as a foreign language writing proficiency and the writing time's effect on this relationship through detailed analysis of a case stud...This study attempts to explore the relationship between translation and English as a foreign language writing proficiency and the writing time's effect on this relationship through detailed analysis of a case study of sixteen English majors who are required to write four English compositions, each time each group adopting one of the two different writing methods: the direct method and the translation-based method. The first two writing tasks have a time limit of 45 minutes while the latter two do not. The results show that the translation-based method has advantages over the direct writing method with regard to the quality of the compositions. Based on this finding, the author claims that the translation-based method is an effective way to train and enhance EFL learners' English writing ability.展开更多
基金the National Natural Science Foundation of China (Nos. U1501252, 61572146 and U1711263)the Natural Science Foundation of Guangxi Province (No. 2016GXNSFDA380006)+1 种基金the Guangxi Innovation-Driven Development Project (No. AA17202024)the Guangxi Universities Young and Middle-aged Teacher Basic Ability Enhancement Project (No. 2018KY0203).
文摘Knowledge graph embedding aims at embedding entities and relations in a knowledge graph into a continuous, dense, low-dimensional and realvalued vector space. Among various embedding models appeared in recent years, translation-based models such as TransE, TransH and TransR achieve state-of-the-art performance. However, in these models, negative triples used for training phase are generated by replacing each positive entity in positive triples with negative entities from the entity set with the same probability;as a result, a large number of invalid negative triples will be generated and used in the training process. In this paper, a method named adaptive negative sampling (ANS) is proposed to generate valid negative triples. In this method, it first divided all the entities into a number of groups which consist of similar entities by some clustering algorithms such as K-Means. Then, corresponding to each positive triple, the head entity was replaced by a negative entity from the cluster in which the head entity was located and the tail entity was replaced in a similar approach. As a result, it generated a set of high-quality negative triples which benefit for improving the effectiveness of embedding models. The ANS method was combined with the TransE model and the resulted model was named as TransE-ANS. Experimental results show that TransE-ANS achieves significant improvement in the link prediction task.
文摘This study is undertaken to find out writers' attention patterns in these two different writing methods (the direct method and the translation-based method) during the writing process through the think-aloud method and EFL learners' perception of the advantages and disadvantages of those two writing methods.The results show that the subjects' attention patterns in the direct method and in the translation-based method are quite different and the differences are mainly embodied in the linguistic level attention and personal comment.As far as the subjects are concerned,writing directly in English is easier and faster than writing through translation,and the direct writing method often helps them learn English language and forces them to focus on English expression.In contrast,for the translation-based writing method,the subjects as a whole relate that they have a wide range of vocabulary and expressions,have a greater number of ideas,and can think through ideas clearly.
文摘This study attempts to explore the relationship between translation and English as a foreign language writing proficiency and the writing time's effect on this relationship through detailed analysis of a case study of sixteen English majors who are required to write four English compositions, each time each group adopting one of the two different writing methods: the direct method and the translation-based method. The first two writing tasks have a time limit of 45 minutes while the latter two do not. The results show that the translation-based method has advantages over the direct writing method with regard to the quality of the compositions. Based on this finding, the author claims that the translation-based method is an effective way to train and enhance EFL learners' English writing ability.
