Systems and methods for building a scripting knowledge base of a platform
AssigneeKinaxis Inc.
InventorsMarin Creanga, Arslan Shahid, Jian Wu
Life of this patent
Estimating the term…
Abstract
Systems and methods for building a scripting knowledge base of a platform, which includes: extracting text from scripting documentation related to the platform, to produce a text document; chunking the text document based on one or more delimiters to produce a plurality of chunks; and sequentially processing each chunk of the plurality of chunks separately by: creating a vector associated with a respective chunk; and persisting the vector and the respective chunk.
Claims12
A computing apparatus for building a scripting knowledge base of a platform, the computing apparatus comprising:
- a processor;
- and
- a memory storing instructions that, when executed by the processor, configure the apparatus to:
- extract text from scripting documentation related to the platform, to produce a text document;
- chunk the text document based on one or more delimiters to produce a plurality of chunks;
- and
- sequentially process each chunk of the plurality of chunks separately;
- create a vector associated with a respective chunk;
- and
- persist the vector and the respective chunk.
A non-transitory computer-readable storage medium for building a scripting knowledge base of a platform, the computer-readable storage medium including instructions that when executed by a computer, cause the computer to:
- extract text from scripting documentation related to the platform, to produce a text document;
- chunk the text document based on one or more delimiters to produce a plurality of chunks;
- and
- sequentially process each chunk of the plurality of chunks separately;
- create a vector associated with a respective chunk;
- and
- persist the vector and the respective chunk.
A computer-implemented method for building a scripting knowledge base of a platform, the method comprising:
- extracting, by a processor, text from scripting documentation related to the platform, to produce a text document;
- chunking, by the processor, the text document based on one or more delimiters to produce a plurality of chunks;
- and
- sequentially processing, by the processor, each chunk of the plurality of chunks separately by:
- creating, by the processor, a vector associated with a respective chunk;
- and
- persisting, by the processor, the vector and the respective chunk.
Dependencies are inferred from the claim text.