Title: An efficient image-aware Kana-Kanji conversion algorithm for Twitter on mobile phones
Abstract: Mobile phones become widespread and their performance is improving. Applications like E-mail, Web search and so on are widely used on mobile phones now. Social networking service (SNS) such as Twitter is especially used since it has mobile apps for mobile phones like smart phones. Twitter enables us to send and read short messages called “tweets”. Therefore, opportunities and needs are increasing to input Japanese sentences for Twitter on mobile phones. A Twitter user is able to upload a photo and attach it to a tweet because most mobile phones are camera phones which are able to capture photographs. For example, a user sends a tweet with a photo about “TOKYO SKYTREE” when she/he visits there. We consider that the photo-image and the text of the tweet-message are strongly related in this tweet. Then, we propose an image-aware Kana-Kanji conversion algorithm for Twitter on mobile phones. A Kana-string inputted by a user is ambiguous because the Kana-string may corresponds to several Kanji-words in Japanese. For disambiguation, our proposed method uses the relation between the photo-image and the text of the tweet-message. We consider that the similarity of the text is high when the similarity of the images is high in tweets sent by users. The order of priority for the Kanji-word increases in the Kana-Kanji conversion algorithm when the similarity is high. Thus, our proposed method enables us to efficiently input Japanese text for Twitter on mobile phones. We show processes of our proposed method and also show the effectiveness of our proposed method by the result of the evaluation experiment.
Publication Year: 2015
Publication Date: 2015-09-01
Language: en
Type: article
Indexed In: ['crossref']
Access and Citation
AI Researcher Chatbot
Get quick answers to your questions about the article from our AI researcher chatbot