| 主权项 |
1. A computer-implemented method, comprising:
receiving, at a computing device including one or more processors, a composition input including one or more characters in a source language; determining, at the computing device, candidate selections for two or more target languages based on the composition input; determining, at the computing device, a language context value for each of the two or more target languages by evaluating the candidate selections against language models for the two or more target languages, wherein each language model includes a rule set for a corresponding target language; selecting, at the computing device, a set of the candidate selections based on the language context values, the set of candidate selections including at least one candidate selection in each of the two or more target languages; and outputting, from the computing device, the set of candidate selections in a single, interleaved list of candidate selections arranged based on a relative likelihood that a specific candidate selection was intended from the composition input. |
