Methods, systems and techniques for blending online content from multiple disparate content sources including a personal content source or a semi-personal content source
The present teaching, which includes methods, systems and computer-readable media, relates to providing content from multiple disparate sources including a person's personal data sources and non-personal data sources. The disclosed techniques may include receiving a request for content from a person; obtaining first content from a first source private to the person based on the request; obtaining second content from at least one second source based on the request; blending the first content from the first source and the second content from the at least one second source to generate a blended content; and providing the blended content to the person in response to the request.
1. A method, implemented on at least one computing device each of which has at least one processor, storage, and a communication platform connected to a network for providing content, the method comprising:
receiving a request for content from a person;
constructing, via a person-centric system, a person centric space including a plurality of data sources, wherein information in each of the data sources is obtained by digesting and cross-linking information from a private space, a semi-private space, and a public space;
estimating an intent of the person based on previous interactions of the person with the person centric space;
obtaining first content from a first data source of the plurality of data sources based on the request;
identifying one or more cross-linking keys from the first content;
obtaining second content from a second data source of the plurality of data sources based on the identified cross-linking keys, the second data source being different than the first data source;
extracting a first attribute set from the first content via a first model, and a second attribute set from the second content via a second model, wherein the first model and the second model are selected based on the first data source and second data source, respectively;
determining, based on the first content and the second content, a feature space including a plurality of features;
generating a first normalized feature set by mapping the first attribute set to the feature space;
generating a second normalized feature set by mapping the second attribute set to feature space;
determining, based on the first normalized feature set and the second normalized feature set, a rank score for each of the first content and the second content, wherein the rank score represents relevance of a corresponding content to the request;
blending the first content and the second content to generate a blended content stream based on the estimated intent, wherein the first content and the second content are ranked in the blended content stream based on the rank scores; and
generating an answer card to be provided to the person in response to the request, wherein the answer card includes the blended content stream.
2. The method of claim 1 , wherein the first data source includes:
private electronic mails;
private calendar;
private contact list;
private messages;
private bookmarks;
private visual information;
private voice information; and
private log information.
3. The method of claim 1 , wherein the second source includes one of:
a semi-private data source; and
a public data source.
4. The method of claim 3 , wherein the semi-private data source includes:
a social media data source; and
a data sharing source.
5. The method of claim 1 , wherein the blended content stream comprises:
a first portion including the first content; and
a second portion including the second content.
6. The method of claim 1 , wherein the first and the second content are mixed in the blended content stream.
7. The method of claim 1 , wherein the first attribute set is different than the second attribute set.
8. The method of claim 1 , wherein the first attribute set is mapped to the feature space based on a set of rules.
9. The method of claim 1 , wherein the first content is in a first format and the second content is in a second format that is different than the first format.
10. The method of claim 1 , wherein a type of the answer card is determined based on the intent of the person.
11. The method of claim 1 , wherein the answer card includes the blended content stream presented in a format that is determined based on a type of the answer card.
12. The method of claim 1 , wherein the step of generating the answer card further comprises:
determining a type of answer card to be provided to the person, wherein the type of answer card is determined based on the intent.
13. The method of claim 1 , wherein
the rank scores are content relevancy scores for the first content and the second content.
14. A system having at least one processor, storage, and a communication platform for providing content, the system comprising:
an input request processing unit implemented on the at least one processor and configured to receive a request for content from a person;
a person-centric index system implemented on the at least one processor and configured to construct a person centric space including a plurality of data sources, wherein information in each of the data sources is obtained by digesting and cross-linking information from a private space, a semi-private space, and a public space;
an intent engine implemented on the at least one processor and configured to estimate an intent of the person based on previous interactions of the person with the person centric space;
a first content retriever implemented on the at least one processor and configured to obtain first content from a first data source of the plurality of data sources based on the request, and
identify one or more cross-linking keys from the first content;
a second content retriever implemented on the at least one processor and configured to obtain second content from a second data source of the plurality of data sources based on the identified cross-linking keys, the second data source being different than the first data source;
a content attribute extractor implemented on the at least one processor and configured to extract a first attribute set from the first content via a first model, and a second attribute set from the second content via a second model, wherein the first model and the second model are selected based on the first data source and second data source, respectively;
a feature space selector implemented on the at least one processor and configured to determine, based on the first content and the second content, a feature space including a plurality of features;
a feature generator implemented on the at least one processor and configured to
generate a first normalized feature set by mapping the first attribute set the feature space, and
generate a second normalized feature set by mapping the second attribute set to the feature space;
a content scoring engine implemented on the at least one processor and configured to determine, based on the first normalized feature set and the second normalized feature set, a rank score for each of the first content and the second content, wherein the rank score represents relevance of a corresponding content to the request; and
a blending engine implemented on the at least one processor and configured to:
blend the first content and the second content to generate a blended content stream based on the estimated intent, wherein the first content and the second content are ranked in the blended content stream based on the rank scores, and
generate an answer card to be provided to the person in response to the request,
wherein the answer card includes the blended content stream.
15. The system of claim 14 , wherein the first data source includes:
private electronic mails;
private calendar;
private contact list;
private messages;
private bookmarks;
private visual information;
private voice information; and
private log information.
16. The system of claim 14 , wherein the second source includes one of:
a semi-private data source; and
a public data source.
17. The system of claim 16 , wherein the semi-private data source includes:
a social media data source; and
a data sharing source.
18. The system of claim 14 , wherein the blended content stream comprises:
a first portion including the first content; and
a second portion including the second content.
19. The system of claim 14 , wherein the first and the second content are mixed in the blended content stream.
20. A non-transitory machine-readable medium having information recorded thereon for providing query suggestions, wherein the information, when read by the machine, causes the machine to perform operations comprising:
receiving a request for content from a person;
constructing, via a person-centric system, a person centric space including a plurality of data sources, wherein information in each of the data sources is obtained by digesting and cross-linking information from a private space, a semi-private space, and a public space;
estimating an intent of the person based on previous interactions of the person with the person centric space;
obtaining first content from a first data source of the plurality of data sources based on the request;
identifying one or more cross-linking keys from the first content;
obtaining second content from a second data source of the plurality of data sources based on the identified cross-linking keys, the second data source being different than the first data source;
extracting a first attribute set from the first content via a first model, and a second attribute set from the second content via a second model, wherein the first model and the second model are selected based on the first data source and second data source, respectively;
determining, based on the first content and the second content, a feature space including a plurality of features;
generating a first normalized feature set by mapping the first attribute set to the feature space;
generating a second normalized feature set by mapping the second attribute set to the feature space;
determining, based on the first normalized feature set and the second normalized feature set, a rank score for each of the first content and the second content, wherein the rank score represents relevance of a corresponding content to the request;
blending the first content and the second content to generate a blended content stream based on the estimated intent, wherein the first content and the second content are ranked in the blended content stream based on the rank scores; and
generating an answer card to be provided to the person in response to the request, wherein the answer card includes the blended content stream.
21. The medium of claim 20 , wherein the first data source includes:
private electronic mails;
private calendar;
private contact list;
private messages;
private bookmarks;
private visual information;
private voice information; and
private log information.
22. The medium of claim 20 , wherein the second source includes one of:
a semi-private data source; and
a public data source.
23. The medium of claim 22 , wherein the semi-private data source includes:
a social media data source; and
a data sharing source.
24. The medium of claim 20 , wherein the blended content stream comprises:
a first portion including the first content; and
a second portion including the second content.
25. The medium of claim 20 , wherein the first and the second content are mixed in the blended content stream.