Word embeddings are used in various fields of natural language processing. The use of word embeddings and concept or word sense embeddings demonstrated effectiveness in many tasks, such as machine translation and text summarization. However, it is difficult to obtain a sufficiently large concept-tagged corpus, as the annotation of concept-tags is timeconsuming. Therefore, in this paper, we propose a method for generating concept embeddings of Word List by Semantic Principles, a Japanese

more »

... s, using both a corpus tagged by an all-words word sense disambiguation (WSD) system and a manually tagged corpus. We generated concept embeddings via fine-tuning using both an automatically tagged corpus and a small manually tagged corpus. In this paper, we propose a novel method of evaluating concept embeddings using the tree structure of Word List by Semantic Principles. Experiments revealed the effectiveness of fine-tuning. The best performance was achieved when the concept embeddings were initially trained with a corpus tagged by an all-words WSD system and retrained with a manually tagged corpus.