Text input method using continuous trace across two or more clusters of candidate words to select two or more words to form a sequence, wherein the candidate words are arranged based on selection probabilities

A computing device comprising:

• an input interface;
• a display;
• a processor, when operating a set of computer-readable instructions to perform a word prediction process, is configured to:
• select a first set of candidate words from a database;
• identify a second set of candidate words in the database after selecting said first set of candidate words;
• simultaneously display, via said display, a first word cluster and a second word cluster in advance of an input to select a plurality of candidate words from said first word cluster and said second word cluster, wherein said first word cluster comprises a plurality of words from said first set and said second word cluster comprises plurality of words from said second set;
• during simultaneous display of said first word cluster and said second word cluster, detect a continuous trace across said display and through said first word cluster and said second word cluster, said continuous trace generated by a user, and identifying identify said plurality of selected candidate words disposed along or proximal to said continuous trace, wherein one of said plurality of selected candidate words is from said first word cluster and another of said plurality of selected candidate words is from said second word cluster;
• and
• in response to detection of said continuous trace, display said plurality of selected candidate words in a text entry field of said display,
• wherein said candidate words in said first set are arranged within said first word cluster in dependence on a first statistical indicator of a likelihood of each candidate word of said first set being selected by said user;
• and said candidate words in said second set are arranged within said second word cluster in dependence on a second statistical indicator of a likelihood of each candidate word of said second set being selected by said user,
• wherein said first statistical indicator comprises a selection probability, said selection probability being a combined probability to provide an indication of a likelihood of two or more words being selected to form a sequence,
• wherein said processor is further configured to:
• identify one or more pairs of candidate words in said first and second sets having a relatively high combined selection probability;
• and
• display said one or more pairs of candidate words proximal to each other in said first and second word clusters,
• wherein, within each of said first word cluster and said second word cluster, said candidate words are arranged such that the candidate words having similar selection probabilities are spaced apart from each other.