Content quality apparatus, systems, and methods

Abstract

Embodiments herein receive a set of content quality threshold values, a search string, and a content data stream at a content quality metric (CQM) apparatus. Content segments associated with the content data stream are scored and/or graded according to a set of content relevance scales. The content data stream is then filtered to include only passing content segments and intermediate calculation values used to determine whether a content segment is passing. Other embodiments are described and claimed.

1 - 30 . (canceled) 
     
     
         31 . A method of defining a system for assessing a set of documents, the method comprising:
 providing a scoring module for generating a set of scores for each of a set of documents based on a set of scales that measure relevance of a document according to different criteria; and   providing a filter associated with the scoring module for identifying documents in the set of documents that have scores above a set of corresponding threshold levels for the set of scales.   
     
     
         32 . The method of  claim 31  further comprising providing a grading module for allocating a set of grades for each of the set of documents based on the set of scores. 
     
     
         33 . The method of  claim 31  further comprising providing an input module coupled to the filter for receiving, from a user, (i) at least one search string for identifying the set of documents to be scored by the scoring module and (ii) the set of corresponding threshold values. 
     
     
         34 . The method of  claim 31  further comprising providing a database that stores previously-received documents, wherein the scoring module accesses the database to compare a document with a previously-received document. 
     
     
         35 . The method of  claim 31  further comprising providing a document source lookup table that provides a score for a document source of a document. 
     
     
         36 . The method of  claim 31  further comprising providing a document type lookup table that provides a score for a document type of a document. 
     
     
         37 . The method of  claim 31  further comprising providing a connectedness lookup table that provides an obscurity score indicating the degree of connectedness of a document to a user. 
     
     
         38 . A method of computing content relevance for documents, the method comprising:
 receiving a set of documents;   scoring each of the set of documents based on a set of scales;   receiving, from a user, a set of threshold values for the set of scales; and   filtering the set of documents to obtain a subset of documents with scores that exceed the set of threshold values.   
     
     
         39 . The method of  claim 38 , wherein receiving the set of documents comprises receiving a search string from the user to identify the set of documents. 
     
     
         40 . The method of  claim 39  further comprising parsing the search string into a set of segments, wherein scoring a particular document comprises calculating a set of user query scores for the document, each user query score associated with a particular segment, wherein a user query score for a particular segment is based on at least one of (1) a frequency of occurrence of the segment in the document, (2) a prominence of location of the segment in the document, and (3) a prominence of textual attributes associated with the segment in the document. 
     
     
         41 . The method of  claim 38 , wherein scoring a document comprises:
 obtaining a source identifier associated with the document; and   retrieving a source score for the document from a source lookup table based on the source identifier.   
     
     
         42 . The method of  claim 38 , wherein scoring a document comprises:
 obtaining a type identifier associated with the document; and   retrieving a type score for the document from a type lookup table based on the type identifier.   
     
     
         43 . The method of  claim 38 , wherein scoring a particular document comprises:
 determining a set of previously-received documents stored in a document database;   concatenating each of the set of previously-received documents to create a concatenated document; and   performing a syntactic text comparison between the particular document and the concatenated document to (i) identify portions of the newly-received document that are different from the concatenated document and to (ii) calculate a syntactic difference score that compares the actual text of the concatenated document and the newly-received document.   
     
     
         44 . The method of  claim 38 , wherein scoring a particular document comprises:
 determining a set of previously-received documents stored in a document database;   concatenating each of the set of previously-received documents to create a concatenated document; and   performing a semantic text comparison between the particular document and the concatenated document to (i) identify portions of the newly-received document that are different from the concatenated document and (ii) calculate a semantic difference score that compares the actual meaning of the subject matter of the concatenated document and the newly-received document.   
     
     
         45 . The method of  claim 38 , wherein scoring a document comprises:
 retrieving a connectedness value from a connectedness lookup table for the document;   calculating an obscurity score associated with the document based on the connectedness value; and   weighing the obscurity score based on a user profile, wherein the user profile indicates a specialty topic associated with the user.   
     
     
         46 . The method of  claim 38 , wherein scoring a document comprises:
 identifying a keyword in the document;   retrieving an impact value associated with the keyword by using an impact lookup table; and   calculating an impact score for the document based on the impact value.   
     
     
         47 . A computer readable medium storing a computer program for computing content relevance for a set of documents associated with a stream of documents, the computer program comprising sets of instructions for:
 receiving a user query and a set of score thresholds for a set of relevance scales;   evaluating a set of documents according to the set of relevance scales, wherein at least one of the scales is affected by the user query; and   identifying documents in the set that score above the thresholds for the set of relevance scales.   
     
     
         48 . The computer readable medium of  claim 47 , wherein the scale affected by the user query evaluates the set of documents by measuring applicability to the user query based on a location of the user query in the document. 
     
     
         49 . The computer readable medium of  claim 47 , wherein the scale affected by the user query is a content source score that is affected by a presence of a particular identifier in the user query. 
     
     
         50 . The computer readable medium of  claim 49 , wherein the particular identifier identifies one of an entity and a market.