Evaluation of Number-Kanji Translation Method of Non-Segmented Japanese Sentences Using Inductive Learning with Degenerated Input

Applied Computational Intelligence and Soft Computing

Suzuki²

2012

Self Cite

Opportunities and needs are increasing to input Japanese sentences on mobile phones since performance of mobile phones is improving. Applications like E-mail, Web search, and so on are widely used on mobile phones now. We need to input Japanese sentences using only 12 keys on mobile phones. We have proposed a method to input Japanese sentences on mobile phones quickly and easily. We call this method number-Kanji translation method. The number string inputted by a user is translated into KanjiKana mixed sentence in our proposed method. Number string to Kana string is a one-to-many mapping. Therefore, it is difficult to translate a number string into the correct sentence intended by the user. The proposed context-aware mapping method is able to disambiguate a number string by artificial neural network (ANN). The system is able to translate number segments into the intended words because the system becomes aware of the correspondence of number segments with Japanese words through learning by ANN. The system does not need a dictionary. We also show the effectiveness of our proposed method for practical use by the result of the evaluation experiment in Twitter data.

Section: Results and Considerationsmentioning

confidence: 76%

“…However, the system is for English mainly. Some input methods have been proposed for Japanese [3][4][5]. The methods enable us to input one Kana character per key press.…”

Section: Introductionmentioning

confidence: 99%

Effectiveness of Context-Aware Character Input Method for Mobile Phone Based on Artificial Neural Network

Applied Computational Intelligence and Soft Computing

Suzuki²

2012

Self Cite

“…and the number of times the user has to activate an input key to select one syllable is large. In order to achieve efficient input we use the Number-Kanji Translation method [12], [13]. In this method the user selects a syllable set with a single key press (for example, the syllable set corresponding to "key 1") and disambiguation is left for later.…”

Section: Allocation Of Japanese Syllablesmentioning

confidence: 99%

A Japanese Input Method for Mobile Terminals Using Surface EMG Signals

Hatano

Araki

New Frontiers in Artificial Intelligence

2009

Self Cite

Abstract. The common use of mobile terminals is for text input. However, mobile terminals cannot be equipped with sufficient amount of keys because of the physical restrictions. To solve this problem we developed an input method using surface electromyogram (sEMG), treating arm muscle movements as input signals. This method involves no physical keys and can be used to input Japanese texts. In our experiments, the system was capable of inputting Japanese characters with a finger motion recognition rate of approximately 80%.

“…Although the cyclic character specification method has been adopted as a general input method, this input method requires multiple key-presses to enter one character. Therefore, to enable rapid input in the proposed technique, we decided to adopt the degenerated character information method [8] so that one character is entered according to one key-press [9,10]. To enter yakyuu, for example, 1 + 2 + 5 + 3 = 11 keystrokes are required with the cyclic character specification method, while input is completed in 1 + 1 + 1 + 1 = 4 keystrokes with the degenerated character information method.…”

Section: Introductionmentioning

confidence: 99%

“…To translate this number string into an English sentence, we also considered a method of performing Japanese-to-English machine translation after using a number-kanji conversion technique [9,10] to temporarily convert the number string to a Japanese sentence. However, proofreading by the user is indispensable for converting the input number string to a Japanese sentence by using a number-kanji conversion technique.…”

Section: Introductionmentioning

confidence: 99%

Machine translation method using inductive learning for mobile terminal

Electron Comm Jpn Pt III

Araki

Tochinai

2004

Self Cite

SUMMARYThe authors propose a machine translation technique for a compact terminal such as a mobile telephone. With a compact terminal that emphasizes portability such as a mobile phone, the number of keys that the terminal can be equipped with is necessarily restricted by the limited size of the terminal itself. Therefore, to enable input by using only this small number of keys, the proposed technique assigns multiple characters to a single input key. In addition, to enable rapid input, a single character is entered by a single key-press. As a result, the number sequence that is the input string to the proposed technique is associated with multiple character strings, and thus there also exist multiple translation candidates for this number sequence, thereby increasing ambiguity. Nevertheless, by using inductive learning, the proposed technique can automatically acquire translation rules that depend on the subject and use these rules for translation. This high level of adaptability enables the proposed technique to eliminate ambiguity as much as possible during the translation process to obtain a correct translation. The authors created a system based on the proposed technique and assigned a small number of word translation rules in advance to perform Japanese-to-English machine translation experiments. The results verified the effectiveness of the proposed technique by showing that the final translation precision of approximately 70% and the translation efficiency were high.