US 12,461,715 B2
Methods and apparatus for facilitating usage of development environments for computer programming with conversational artificial intelligence
Radhika Grover, San Jose, CA (US)
Assigned to Radhika S. Grover, San Jose, CA (US)
Filed by Radhika Grover, San Jose, CA (US)
Filed on Mar. 3, 2023, as Appl. No. 18/117,382.
Prior Publication US 2024/0296016 A1, Sep. 5, 2024
Int. Cl. G06F 8/33 (2018.01); G06F 3/01 (2006.01); G06F 8/30 (2018.01); G06F 8/73 (2018.01); G06F 8/75 (2018.01); G06F 40/289 (2020.01); G10L 15/06 (2013.01); G10L 15/18 (2013.01); G10L 15/22 (2006.01); G10L 15/08 (2006.01)
CPC G06F 8/30 (2013.01) [G06F 3/017 (2013.01); G06F 40/289 (2020.01); G10L 15/063 (2013.01); G10L 15/18 (2013.01); G10L 15/22 (2013.01); G06F 8/33 (2013.01); G06F 8/75 (2013.01); G10L 2015/0635 (2013.01); G10L 2015/088 (2013.01); G10L 2015/223 (2013.01)] 24 Claims
OG exemplary drawing
 
1. A computer-implemented method for processing user-issued commands expressed in a grammatically structured human language such as English and received via a development environment used for programming, the method comprising:
loading into memory at least one reference document comprising phrases, weights, and actions associated with programming tasks;
splitting each phrase in the reference document into its keywords;
creating a reference fuzzy hash map from the at least one reference document for the development environment for computer programming, wherein the reference fuzzy hash map stores, for each keyword in the reference document, a (key, value) pair comprising of the keyword as the key and a list of corresponding (action, weight) pairs of that keyword as the value;
obtaining the user-issued command through at least one of a keyboard module, speech recognition module and a gesture recognition module;
extracting from the user-issued command, a plurality of unique keywords;
identifying for each keyword of the plurality of keywords, a plurality of corresponding candidate actions using the fuzzy hash map;
determining information metrics and their weights for each candidate action of the plurality of candidate actions;
ranking each of the candidate actions using the information metrics and their weights;
selecting one or more of the candidate actions based on the ranking and selection criteria;
storing the one or more selected candidate actions; and
executing the one or more selected candidate actions in the development environment.