Media content reviews search

• US 7,630,966 B2
• Filed: 03/16/2006
• Issued: 12/08/2009
• Est. Priority Date: 03/16/2006
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Arora, Gagan
• Primary Examiner(s): Vital; Pierre M
• Assistant Examiner(s): Vo; Truong V
• Application Number:
  US11/276,847
• Publication Number:
  US 20070220056A1
• Time in Patent Office: 1,363 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06F 16/48 Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
• CPC Class Codes: G11B 27/28 by using information signals recorded by the same method as the main recording G11B27/22 takes precedence
• CPC Class Codes: H04N 21/4828 for searching program descriptors retrieval of video data G06F16/739
• CPC Class Codes: H04N 7/17318 Direct or substantially direct transmission and handling of requests
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching

Abstract

Media content reviews search is described. In an embodiment, an indexing service identifies keywords in a program review of a program where the program review includes one or more depictions of the program. The indexing service also identifies content identifiers in the program review of the program, and generates a program index to position the program into an index of programs that includes similar programs based on depictions of the programs.

Claims

1. A search system, comprising:

a television-based client device; and

a database coupled to the client device and searchable by the client device, the database to maintain;

identities of programs of various media content providers and associated reviews, wherein each review includes a subjective depiction of one of the programs, the subjective depiction including descriptive information that exceeds a generic plot description of the program; and

an index of the programs generated by an indexing service; and

the indexing service configured to;

identify keywords in the subjective depiction of a program;

identify content identifiers in the subjective depiction based on a list of known content identifiers;

filter the subjective depiction of the program to retain words that are the keywords and the content identifiers in the subjective depiction by object filtering to remove words that lack a relevance to the subjective depiction of the program;

weight the keywords with a naive Bayesian algorithm;

generate a hash table of the weighted keywords optimized to enable searching for other subjective depictions that share similar keywords;

generate the index of the programs based on a combination of;

the hash table of the weighted keywords;

the content identifiers; and

co-occurrence analysis; and

position each program in the index of the programs such that each program is associated with one or more similar programs, the association between programs based on the keywords, the content identifiers, and the co-occurrence analysis, wherein the indexing service is further configured to generate the program index based on the content identifiers which are descriptive terms in the subjective depiction and based on the keywords in the subjective depiction and wherein the indexing service is further configured to generate the program index as a graph in which the distance between the program and an additional program is an indication of whether the program and the additional program are similar.

2. A search system as recited in claim 1, wherein a subjective depiction of the program is at least one of an impression, a theme, or a feature of the program.

3. A search system as recited in claim 1, wherein the content identifiers are descriptive terms about the program in the subjective depiction.

4. A search system as recited in claim 1, further comprising a recommendation service to:

receive a search request for additional programs that are similar to the program;

communicate with the indexing service to identify the similar programs in the index of programs; and

provide a list of the similar programs as viewing recommendations.

5. A search system as recited in claim 4, wherein the recommendation service provides a list of the similar programs as viewing recommendations via an email message or a mobile text message.

7. A method as recited in claim 5, wherein the subjective depiction of the program is an impression, a theme, or a feature of the program.

8. A method as recited in claim 7, further comprising:

ranking the keywords and the key phrases identified in the subjective depiction;

determining whether the program is similar to an additional program based on a similarity rank that is established based on;

first, a similarity of the ranked keywords and key phrases in conjunction with the content identifiers between the program and the additional program;

second, a similarity of the ranked keywords and key phrases between the program and the additional program; and

third, a similarity of the content identifiers between the program and the additional program.

9. A method as recited in claim 5, wherein the content identifiers are identified in the subjective depiction based on a list of known content-based identifiers including a director'"'"'s name, a writer'"'"'s name, and actors'"'"' names.

10. A method as recited in claim 5, further comprising:

receiving a search request for additional programs that are similar to the program;

identifying the similar programs in the index of programs as being similar to the program; and

providing a list of the similar programs as viewing recommendations.

11. A method as recited in claim 5, wherein the subjective depictions describe with subjective relevance how the writers of the subjective depictions think about the program and wherein the writers are movie reviewers, critics, members of usenet groups, or bloggers.

6. A method implemented on a computing device by a processor configured to execute instructions that, when executed by the processor, direct the computing device to perform acts comprising:

identifying one or more reviews of a program, wherein each review includes a subjective depiction of the program, each subjective depiction including descriptive information that exceeds a generic plot description of the program;

identifying content identifiers and keywords in the subjective depiction, wherein the content identifiers are based on a list of known content-based identifiers and comprise descriptive terms about the program and the keywords represent the subjective depiction of the program;

filtering the subjective depiction of the program to eliminate words that are not the keywords or the content identifiers in the program subjective depiction;

generating the program index of the program includes weighting the keywords and generating the program index based on the weighted keywords and the content identifiers;

ranking the keywords and content identifiers with a naï

ve Bayesian algorithm;

generating a hash table of the ranked keywords and the content identifiers, the hash table optimized to enable searching for other subjective depictions that share similar keywords or content identifiers to the identified one or more subjective depictions of the program;

generating an index of the program based on the hash table such that the program is associated with similar programs in the index; and

storing the index in a database which is coupled to and searchable by a television-based client device, wherein generating the program index of the program includes generating a graph in which the distance between the program and an additional program is an indication of whether the program and the additional program are similar.

12. A method for identifying media programming and viewing recommendations that are subjectively similar to a program implemented on a computing device by a processor configured to execute instructions, that when executed by the processor direct the computing device to perform acts comprising:

receiving a search request from a television-based client device for additional programs that are similar to the program;

communicating the search request to an indexing service that is in communication with the television-based client device, that is searchable by the television-based client device, and that relates similar programs, wherein a similarity is based on subjective depictions of the programs by critics, opinion reviews, and additional information about the program beyond a plot description;

communicating a program subjective depiction of the program to the indexing service with the search request, the indexing service configured to;

identify content identifiers that are descriptive terms in the subjective depiction;

identify keywords in the subjective depiction of the program;

filter the subjective depiction of the program to retain words that are the keywords and the content identifiers in the subjective depiction by object filtering to remove words that lack a relevance to the subjective depiction of the program;

weight the keywords with a naï

ve Bayesian algorithm;

generate a hash table of the weighted keywords optimized to enable searching for other reviews that share similar keywords;

generate a program index of the program to determine which of the programs in the index of programs are similar, the program index being generated based on the content identifiers and the keywords in the subjective depiction; and

determining whether the program is similar to an additional program based on a similarity rank that is established based on;

a similarity of the ranked keywords and key phrases in conjunction with the content identifiers between the program and the additional program;

a similarity of the ranked keywords and key phrases between the program and the additional program; and

a similarity of the content identifiers between the program and the additional program;

determining a degree of the similarity by traversing an index graph, wherein a shorter distance between two programs on the index graph indicates a greater similarity;

receiving a list of the similar programs from the indexing service; and

communicating the list of similar programs to the television-based client device as viewing recommendations in a program guide selection displayed on a display device associated with the television-based client device, wherein the viewing recommendations are in a ranked order based on a distance offset in the index graph.

13. A method as recited in claim 12, wherein the list of similar programs includes programs having at least one of an impression, a theme, or a feature that is similar to the program of the search request.

14. A method as recited in claim 12, wherein the television-based client device comprises media content inputs configured to receive Internet Protocol (IP) input via an IP-based network.