US 12,332,857 B2
Automated garbage collection based on contextual analysis driven by NLP audio
Prem Pradeep Motgi, Austin, TX (US); Dharmesh M. Patel, Round Rock, TX (US); and Manpreet Singh Sokhi, Santa Clara, CA (US)
Assigned to Dell Products L.P., Round Rock, TX (US)
Filed by Dell Products L.P., Round Rock, TX (US)
Filed on Aug. 30, 2023, as Appl. No. 18/458,373.
Prior Publication US 2025/0077488 A1, Mar. 6, 2025
Int. Cl. G06F 7/00 (2006.01); G06F 16/215 (2019.01); G06F 16/23 (2019.01); G06F 16/906 (2019.01)
CPC G06F 16/215 (2019.01) [G06F 16/2365 (2019.01); G06F 16/906 (2019.01)] 20 Claims
OG exemplary drawing
 
1. A method for managing limited storage space in a data management system that stores data for a user, the method comprising:
performing a data assessment process based at least on topics in the data to proactively deny or allow storage of the data into the limited storage space as the data is obtained by the data management to ensure that the limited storage space is used only for data containing topics associated with subsequent downstream processes that use the data to provide one or more services for the user by at least:
obtaining the data for storage;
classifying the data as being related to at least one topic, the at least one topic being one of the topics and classifying the data comprises at least:
generating a first set of topics for the data using an inference model hosted by the data management system;
identifying spoken topics in at least one conversation between at least two people to generate a second set of topics for the data, the data being associated with the at least one conversation; and
filtering out from the first set of topics any topics not included in the second set of topics to obtain the at least one topic for the data, the at least one topic being one of one or more topic classifications for the data, the one or more topic classifications comprising only non-filtered out topics from the first set of topics;
identifying quality requirements based on the at least one topic; and
making a determination, based on the quality requirements, regarding whether the data meets the quality requirements where:
in a first instance of the determination where the quality requirements are not met by the data, the method further comprises denying storage of the data at least temporarily; and
in a second instance of the determination where the quality requirements are met by the data, the method further comprises storing the data.