US 12,141,827 B1
Global segmenting and sentiment analysis based on granular opinion detection
Sunil Mallya Kasaragod, San Francisco, CA (US); Abhinav Goyal, Snohomish, WA (US); Yahor Pushkin, Redmond, WA (US); Srikanth Doss Kadarundalagi Raghura, Seattle, WA (US); Rishita Rajal Anubhai, Seattle, WA (US); Kasturi Bhattacharjee, Sunnyvale, CA (US); Smaranda Muresan, New York, NY (US); Siddharth Chaitanyakumar Varia, Northville, MI (US); and Federico Torreti, Seattle, WA (US)
Assigned to Amazon Technologies, Inc., Seattle, WA (US)
Filed by Amazon Technologies, Inc., Seattle, WA (US)
Filed on Dec. 10, 2021, as Appl. No. 17/548,055.
Int. Cl. G06F 40/284 (2020.01); G06F 40/205 (2020.01); G06Q 10/0631 (2023.01); G06Q 30/0204 (2023.01)
CPC G06Q 30/0205 (2013.01) [G06F 40/205 (2020.01); G06F 40/284 (2020.01); G06Q 10/06315 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A system, comprising:
a global segmenting and analysis service hosted by a provider network and Internet-accessible by a plurality of clients, wherein the global segmenting and analysis service comprises one or more processors and one or more memories to store computer-executable instructions that, when executed, cause the one or more processors to, for a given client:
receive a plurality of documents comprising text, wherein for a given one of the documents, the global segmenting and analysis service is configured to identify one or more target entities respectively associated with an entity type of an entity taxonomy of the client, and wherein for a given target entity, the global segmenting and analysis service is configured to:
for different occurrences of the target entity within the document, determine a sentiment polarity for the target entity and at least one cause of the sentiment polarity;
store the sentiment polarity for the target entity and the at least one cause of the sentiment polarity for the different occurrences;
based on the sentiment polarity for the target entity for the different occurrences, determine a document-level sentiment polarity for the target entity for the document;
store the document-level sentiment polarity for the target entity for the document; and
determine an additional sentiment polarity for the target entity as a multi-document level sentiment polarity for the target entity based on the document-level sentiment polarity for the target entity previously determined for the document and at least another document-level sentiment polarity for the target entity previously determined for at least another document of the plurality of documents;
store the multi-document level sentiment polarity for the target entity;
receive, from the client, a query that indicates at least the target entity or the entity type associated with the target entity; and
in response to the query, provide to the client at least one or more of:
the multi-document level sentiment polarity for the target entity,
a multi-document level sentiment polarity for target entity and other target entities that are also associated with the entity type of the entity taxonomy, or
the document-level sentiment polarity for the target entity for the document and at least the other document-level sentiment polarity for the target entity for at least the other document.