Metadata tag description generation

Abstract

One or more techniques and/or systems are provided for metadata tag evaluation. For example, a metadata tag, associated with content, may be identified (e.g., a hashtag #ML may be used to tag a social network post). A set of characters, within the content, may be evaluated utilizing a probability matrix and the content to identify an expanded metadata tag (e.g., an expanded hashtag “machine learning”). Descriptive content, such as websites, articles, social network posts, and/or other content associated with the expanded metadata tag, may be retrieved. A description for the metadata tag may be generated based upon the descriptive content (e.g., a definition for machine learning). In this way, the description, related metadata tags, and/or supplemental content may be provided to users having an interest in learning about the metadata tag.

1 . (canceled) 
     
     
         2 . The method of  claim 21 , further comprising:
 receiving a query comprising the first metadata tag;   identifying the expanded first metadata tag as an expansion of the first metadata tag; and   providing the first description for the first metadata tag as a query result for the query.   
     
     
         3 . The method of  claim 21 , further comprising:
 retrieving a second content comprising the first metadata tag and a second metadata tag; and   identifying, based on the second content, the second metadata tag as being related to the first metadata tag.   
     
     
         4 . The method of  claim 3 , further comprising:
 receiving a query comprising the first metadata tag;   identifying the expanded first metadata tag as an expansion of the first metadata tag; and   providing the first description for the first metadata tag and a metadata tag suggestion, comprising the second metadata tag, as a query result for the query.   
     
     
         5 - 7 . (canceled) 
     
     
         8 . The method of  claim 21 , the determining that words or phrases correspond to the sets of one or more characters comprising:
 identifying a first word or phrase for a first set of one or more characters based upon the first word or phrase having a first frequency probability, within the probability matrix, above a threshold;   identifying a second word or phrase for a second set of one or more characters based upon the second word or phrase having a second frequency probability, within the probability matrix, above the threshold; and   identifying the first word or phrase and the second word or phrase for inclusion within the expanded first metadata tag based upon a longer phrase, comprising the first word or phrase and the second word or phrase, having a phrase frequency probability, within the probability matrix, above a phrase threshold.   
     
     
         9 . The method of  claim 21 , the generating the first description comprising:
 extracting a first descriptive phrase from the retrieved descriptive content; and   responsive to the first descriptive phrase occurring within the retrieved descriptive content above a threshold, including the first descriptive phrase within the first description.   
     
     
         10 . The method of  claim 21 , the generating the first description comprising:
 generating a first description candidate and a second description candidate for the first metadata tag based upon the retrieved descriptive content; and   receiving a user selection of either the first description candidate or the second description candidate for utilization in generating the first description for the metadata tag.   
     
     
         11 . The method of  claim 21 , further comprising:
 responsive to determining that the expanded first metadata tag corresponds to at least one of spam or explicit content, marking at least one of the first metadata tag or the first description as at least one of spam or explicit content.   
     
     
         12 . The method of  claim 21 , further comprising:
 associating a first temporal identifier with the first description, the first temporal identifier corresponding to a first time range during which the first description is descriptive of the first metadata tag;   storing the first description within a data structure based upon the first temporal identifier;   generating a second description for the first metadata tag based upon a second expanded version of the first metadata tag and a second descriptive content associated with a second time range;   associating a second temporal identifier with the second description, the second temporal identifier corresponding to the second time range during which the second description is descriptive of the first metadata tag; and   storing the second description within the data structure based upon the second temporal identifier.   
     
     
         13 - 14 . (canceled) 
     
     
         15 . The one or more computing devices of  claim 23 , wherein the computer-readable memory comprises further computer-executable instructions, which, when executed by the one or more processing units, cause the one or more computing devices to:
 receive a query comprising the first metadata tag;   identify the expanded first metadata tag as an expansion of the first metadata tag; and   provide the first description for the first metadata tag as a query result for the query.   
     
     
         16 . The one or more computing devices of  claim 23 , wherein the computer-readable memory comprises further computer-executable instructions, which, when executed by the one or more processing units, cause the one or more computing devices to:
 retrieve a second content comprising the first metadata tag and a second metadata tag;   identify, based on the second content, the second metadata tag as being related to the first metadata tag;   receive a query comprising the first metadata tag;   identify the expanded first metadata tag as an expansion of the first metadata tag; and   provide the first description for the first metadata tag and a metadata tag suggestion, comprising the second metadata tag, as a query result for the query.   
     
     
         17 . (canceled) 
     
     
         18 . The one or more computing devices of  claim 23 , wherein the computer-readable memory comprises further computer-executable instructions, which, when executed by the one or more processing units, cause the one or more computing devices to:
 identify a first word or phrase for a first set of one or more characters based upon the first word or phrase having a first frequency probability, within the probability matrix, above a threshold;   identify a second word or phrase for a second set of one or more characters based upon the second word or phrase having a second frequency probability, within the probability matrix, above the threshold; and   identify the first word or phrase and the second word or phrase for inclusion within the expanded first metadata tag based upon a longer phrase, comprising the first word or phrase and the second word or phrase, having a phrase frequency probability, within the probability matrix, above a phrase threshold.   
     
     
         19 . The one or more computing devices of  claim 23 , wherein the computer-readable memory comprises further computer-executable instructions, which, when executed by the one or more processing units, cause the one or more computing devices to:
 extract a first descriptive phrase from the retrieved descriptive content; and   responsive to the first descriptive phrase occurring within the retrieved descriptive content above a threshold, include the first descriptive phrase within the first description.   
     
     
         20 . (canceled) 
     
     
         21 . A method of improving computer search results by extending a computer search index with generated metadata tag descriptions, the method comprising:
 obtaining, at a computing device, a first content, the first content comprising a first metadata tag, the first metadata tag comprising multiple textual characters;   determining, with the computing device, that words or phrases correspond to sets of one or more characters of the first metadata tag based on a first probability matrix;   generating, with the computing device, an expanded version of the first metadata tag by replacing the sets of the one or more characters of the first metadata tag with the words or phrases determined to correspond to the sets of the one or more characters;   retrieving, at the computing device, descriptive content providing further information about the expanded first metadata tag, the descriptive content differing from the first content; and   generating, at the computing device, a first description for the first metadata tag based upon the retrieved descriptive content;   wherein the individual sets of the one or more characters are each less than all of the multiple textual characters of which the first metadata tag is comprised;   wherein the probability matrix comprises, for each of the sets of the one or more characters of the first metadata tag, probabilities that each of multiple different words or phrases correspond to the each of the sets of the one or more characters; and   wherein the probabilities contained in the first probability matrix are dependent, at least in part, on the first content.   
     
     
         22 . The method of  claim 21 , further comprising:
 identifying social media content comprising both the first metadata tag and a reference to supplemental content that is external to the social media content;   receiving a search query comprising the first metadata tag; and   providing both the first description for the first metadata tag and the supplemental content as search results for the query.   
     
     
         23 . One or more computing devices for improving computer search results by extending a computer search index with generated metadata tag descriptions, the one or more computing devices, in aggregate, comprising:
 one or more processing units; and   computer-readable memory comprising computer-executable instructions, which, when executed by the one or more processing units, cause the one or more computing devices to:
 obtain a first content, the first content comprising a first metadata tag, the first metadata tag comprising multiple textual characters; 
 determine that words or phrases correspond to sets of one or more characters of the first metadata tag based on a first probability matrix; 
 generate an expanded version of the first metadata tag by replacing the sets of the one or more characters of the first metadata tag with the words or phrases determined to correspond to the sets of the one or more characters; 
 retrieve descriptive content providing further information about the expanded first metadata tag, the descriptive content differing from the first content; and 
 generate a first description for the first metadata tag based upon the retrieved descriptive content; 
   wherein the individual sets of the one or more characters are each less than all of the multiple textual characters of which the first metadata tag is comprised;   wherein the probability matrix comprises, for each of the sets of the one or more characters of the first metadata tag, probabilities that each of multiple different words or phrases correspond to the each of the sets of the one or more characters; and   wherein the probabilities contained in the first probability matrix are dependent, at least in part, on the first content.   
     
     
         24 . The one or more computing devices of  claim 23 , wherein the computer-readable memory comprises further computer-executable instructions, which, when executed by the one or more processing units, cause the one or more computing devices to:
 associate a first temporal identifier with the first description, the first temporal identifier corresponding to a first time range during which the first description is descriptive of the first metadata tag;   generate a second description for the first metadata tag associated with a second time range, the second description differing from the first description, and based on a second descriptive content that differs from the first descriptive content; and   associate a second temporal identifier with the second description, the second temporal identifier corresponding to the second time range during which the second description is descriptive of the first metadata tag.   
     
     
         25 . The one or more computing devices of  claim 23 , wherein the computer-readable memory comprises further computer-executable instructions, which, when executed by the one or more processing units, cause the one or more computing devices to:
 identify social media content comprising both the first metadata tag and a reference to supplemental content that is external to the social media content;   receive a search query comprising the first metadata tag; and   provide both the first description for the first metadata tag and the supplemental content as search results for the query.   
     
     
         26 . A computer readable storage medium comprising computer-executable instructions, which, when executed, cause a computing device to improve computer search results by extending a computer search index with generated metadata tag descriptions by performing steps comprising:
 obtaining a first content, the first content comprising a first metadata tag, the first metadata tag comprising multiple textual characters;   determining that words or phrases correspond to sets of one or more characters of the first metadata tag based on a first probability matrix;   generating an expanded version of the first metadata tag by replacing the sets of the one or more characters of the first metadata tag with the words or phrases determined to correspond to the sets of the one or more characters;   retrieving descriptive content providing further information about the expanded first metadata tag, the descriptive content differing from the first content; and   generating a first description for the first metadata tag based upon the retrieved descriptive content;   wherein the individual sets of the one or more characters are each less than all of the multiple textual characters of which the first metadata tag is comprised;   wherein the probability matrix comprises, for each of the sets of the one or more characters of the first metadata tag, probabilities that each of multiple different words or phrases correspond to the each of the sets of the one or more characters; and   wherein the probabilities contained in the first probability matrix are dependent, at least in part, on the first content.   
     
     
         27 . The computer readable medium of  claim 26 , comprising further computer-executable instructions, which, when executed, further cause the computing device to perform steps comprising:
 retrieving a second content comprising the first metadata tag and a second metadata tag;   identifying, based on the second content, the second metadata tag as being related to the first metadata tag;   receiving a search query comprising the first metadata tag; and   providing both the first description for the first metadata tag and, as a suggestion, the second metadata tag as a result for the search.   
     
     
         28 . The computer readable medium of  claim 26 , comprising further computer-executable instructions, which, when executed, further cause the computing device to perform steps comprising:
 associating a first temporal identifier with the first description, the first temporal identifier corresponding to a first time range during which the first description is descriptive of the first metadata tag;   generating a second description for the first metadata tag associated with a second time range, the second description differing from the first description and based on a second descriptive content that differs from the first descriptive content; and   associating a second temporal identifier with the second description, the second temporal identifier corresponding to the second time range during which the second description is descriptive of the first metadata tag.