US 2026/0141190 A1Published application

Higher-order function with reducing function parameter to transliterate characters

AssigneeMastercard International Incorporated

InventorsKristóf GYIMESI, Péter ANDI, Amit ANAND

Life of this patent

Estimating the term…

Loading patent…

Abstract

Systems and methods of computational transliteration of an input sequence of characters in a source language such as Thai to a target language such as Latin. The output may include Romanization of Thai names. The output sequence may be used for machine transliteration understanding of words such as proper nouns. A system may execute a higher-order function that calls a reducing function that iterates through sliding windows of an input sequence and updates, with each iteration, an accumulator map that includes a vector of characters and an indication of a number of characters that can be skipped when processing the next window. Each window includes multiple characters in the input sequence for context-based transliteration using contextual transcription rules. Characters can be skipped when they have already been processed in a previous sliding window. Transliteration of certain source languages may include transposition, deletion, insertion, and transcription.

Claims20

Dependent claims

A system, comprising:

a processor programmed to:
access an input sequence having a plurality of characters in a source language to be transliterated into a target language;
execute a higher-order function that takes as input the input sequence, a window size parameter that defines a number of characters to be included in a window, and an identification of a reducing function;
generate, by the higher-order function, a plurality of sliding windows based on the input sequence and the window size parameter, wherein each sliding window comprises a number of characters from the input sequence that is based on the window size to capture context for one or more characters in the sliding window to be transliterated,
for each of at least some of the sliding windows from among the plurality of sliding windows:
execute, by the higher-order function, the reducing function to transliterate one or more characters in the sliding window based on the captured context from the sliding window;
transliterate, by the reducing function, one or more characters in the sliding window based on the captured context from the sliding window;
update, by the reducing function, an accumulator map comprising a vector of characters based on the transliteration and a number of characters to skip in a next iteration;
and
generate a sequence of characters in the target language based on the vector of characters.

7 more elements

A method, comprising:

accessing, by a processor, an input sequence having a plurality of characters in a source language to be transliterated into a target language;
executing, by the processor, a higher-order function that takes as input the input sequence, a window size parameter that defines a number of characters to be included in a window, and an identification of a reducing function;
generating, by the processor, by the higher-order function, a plurality of sliding windows based on the input sequence and the window size parameter, wherein each sliding window comprises a number of characters from the input sequence that is based on the window size to capture context for one or more characters in the sliding window to be transliterated,
for each of at least some of the sliding windows from among the plurality of sliding windows:
executing, by the processor, via the higher-order function, the reducing function to transliterate one or more characters in the sliding window based on the captured context from the sliding window;
transliterating, by the processor, via the reducing function, one or more characters in the sliding window based on the captured context from the sliding window;
updating, by the processor, via the reducing function, an accumulator map comprising a vector of characters based on the transliteration and a number of characters to skip in a next iteration;
and
generating, by the processor, a sequence of characters in the target language based on the vector of characters.

6 more elements

A non-transitory computer readable medium storing instructions that, when executed by a processor, programs the processor to:

access an input sequence having a plurality of characters in a source language to be transliterated into a target language;
execute a higher-order function that takes as input the input sequence, a window size parameter that defines a number of characters to be included in a window, and an identification of a reducing function;
generate, by the higher-order function, a plurality of sliding windows based on the input sequence and the window size parameter, wherein each sliding window comprises a number of characters from the input sequence that is based on the window size to capture context for one or more characters in the sliding window to be transliterated,
for each of at least some of the sliding windows from among the plurality of sliding windows:
execute, by the higher-order function, the reducing function to transliterate one or more characters in the sliding window based on the captured context from the sliding window;
transliterate, by the reducing function, one or more characters in the sliding window based on the captured context from the sliding window;
update, by the reducing function, an accumulator map comprising a vector of characters based on the transliteration and a number of characters to skip in a next iteration;
and
generate a sequence of characters in the target language based on the vector of characters.

6 more elements

Dependencies are inferred from the claim text.