| CPC H04L 51/216 (2022.05) [G06F 16/3329 (2019.01); G06F 16/3334 (2019.01); G06F 16/358 (2019.01)] | 22 Claims |
|
1. A system for creating an intelligent memory and providing contextual intelligent recommendations, the system comprising:
a memory storing program instructions; and
a processor executing the program instructions stored in the memory, wherein executing the program instructions causes the processor to execute;
an intelligent memory generation engine, the intelligent memory generation engine configured to:
extract electronic communications data corresponding to multiple active users queued for extracting by a scheduler queue unit, wherein the extraction is performed based on at least a time zone associated with the user, the time zone preferred by the user or time zone set by an administrator for the user;
perform a keyword tagging operation on the extracted electronic communications data, the keyword tagging operation comprising storing one or more keywords corresponding to one or more entities associated with the electronic communications data as a first tag in a record storage unit, and storing one or more keywords corresponding to conversation data associated with the extracted electronic communications data as a second tag in the record storage unit;
generate a multi-relational model representative of the conversation data associated with the electronic communications data and other data stored in a sub-record storage unit of the record storage unit in the form of graph nodes based on the first tag and the second tag, the multi-relational model comprising node embeddings which are computed for unseen nodes representing unseen relationships between the extracted electronic communications data and the other data, wherein the node embeddings are computed by determining new relationships and undefined behaviour patterns between the extracted electronic communications data and the other data by identifying functions based on properties associated with neighbouring nodes in the graph nodes, the neighbouring nodes including nodes associated with predetermined users and former users who have left an organization or have been reassigned within the organization and are not a sender or receiver of the conversation data; and
transmit one or more electronic Recommendation Action Communication (RAC) with embedded application program interface calls based on the multi-relational model, the embedded application program interface calls enabling actions to be taken on information units via a single click, wherein a user specific briefing message is generated by processing data stored in a reporting database for providing one or more intelligent recommendations, the briefing message is in the form of a Universal Resource Locator (URL) link, and wherein an application programming interface call is triggered, upon selection of the URL, which records a timestamp associated with the user specific briefing message and a subsequent application programming interface call is triggered to the information units by an employing authentication mechanism.
|