Methods, systems and techniques for ranking blended content retrieved from multiple disparate content sources

A non-transitory machine-readable medium having information recorded thereon for generating content, wherein the information, when read by the machine, causes the machine to perform operations comprising:

• receiving a query from a user for information from a dynamic person-centric space created by cross-linking relevant information of the user from public, semi-private, and private spaces and representing a unified and semantically meaningful space;
• estimating an intent of the user based on the query and information known about the user that includes the user action and domain;
• obtaining, based on the query and the intent, a plurality of content items from the dynamic person-centric space;
• extracting a set of attributes for each of the plurality of content items;
• determining a common feature space based on the sets of attributes;
• normalizing each set of attributes with respect to the common feature space;
• determining, for each of the plurality of content items, a rank score based on a corresponding normalized set of attributes in the common feature space;
• and
• generating a content stream by blending the plurality of content items in accordance with the rank scores, wherein the content stream is to be presented to the user via an answer card, which includes each of the plurality of content items in accordance with a format of the answer card.

A system having at least one processor, storage, and a communication platform for generating content, the system comprising:

• an input request processing unit implemented on the at least one processor and configured to
• receive a query from a user for information from a dynamic person-centric space created by cross-linking relevant information of the user from public, semi-private, and private spaces and representing a unified and semantically meaningful space,
• estimate an intent of the user based on the query and information known about the user that includes the user action and domain, and
• obtain, based on the query and the intent, a plurality of content items from the dynamic person-centric space;
• a content attributes extraction engine implemented on the at least one processor and configured to extract a set of attributes for each of the plurality of content items;
• a feature space selection unit implemented on the at least one processor and configured to determine a common feature space based on the sets of attributes;
• a feature normalization engine implemented on the at least one processor and configured to normalize each set of attributes with respect to the common feature space;
• a rank score determination unit implemented on the at least one processor and configured to determine, for each of the plurality of content items, a rank score based on a corresponding normalized set of attributes in the common feature space;
• and
• a content blending unit implemented on the at least one processor and configured to generate a content stream by blending the plurality of content items in accordance with the rank scores, wherein the content stream is to be presented to the user via an answer card, which includes each of the plurality of content items in accordance with a format of the answer card.