Systems and method for performing contextual classification using supervised and unsupervised training

Abstract

Computerized systems and methods are disclosed for performing contextual classification of objects using supervised and unsupervised training. In accordance with one implementation, content reviewers may review training objects and submit supervised training data for preprocessing and analysis. The supervised training data may be preprocessed to identify key terms and phrases, such as by stemming, tokenization, or n-gram analysis, and form vectorized objects. The vectorized objects may be used to train one or more models for subsequent classification of objects. In certain implementations, preprocessing or training, among other steps, may be performed in parallel over multiple machines to improve efficiency. The disclosed systems and methods may be used in a wide variety of applications, such as article classification and content moderation.

1 - 21 . (canceled) 
     
     
         22 . A computer-implemented method for performing comment classification of objects, the method comprising the following operations performed by one or more processors:
 receiving, from a first user, comment classifications related to one or more comments;   determining the first user is a trusted user based on the comment classifications;   receiving, from the trusted user, flagged comments;   processing the flagged comments to form at least one vectorized object based on the flagged comments;   training a plurality of models by applying a plurality of machine learning algorithms to the at least one vectorized object;   identifying an optimal model from the plurality of models; and   classifying other comments using the identified optimal model.   
     
     
         23 . The computer-implemented method of  claim 22 , wherein the first user is selected from the group consisting of an editor, an author, and a third-party user. 
     
     
         24 . The computer-implemented method of  claim 22 , wherein if the first user is a third-party reviewer, the third-party reviewer is trusted to classify objects. 
     
     
         25 . The computer-implemented method of  claim 22 , wherein processing the flagged comments comprises using training data to form at least one vectorized object by performing at least one operation selected from a group consisting of stemming, tokenization, and n-gram analysis. 
     
     
         26 . The computer-implemented method of  claim 22 , wherein the plurality of machine learning algorithms comprises a support vector machine algorithm. 
     
     
         27 . The computer-implemented method of  claim 22 , wherein processing the flagged comments comprises processing supervised training data to form at least one vectorized object by processing training data using a plurality of machines operating in parallel. 
     
     
         28 . The computer-implemented method of  claim 22 , wherein:
 receiving flagged comments comprises receiving at least one pre-defined tag selected by first user as being associated with at least one training article; and   classifying other comments using the identified optimal model comprises automatically classifying a candidate article.   
     
     
         29 . The computer-implemented method of  claim 22 , wherein:
 receiving flagged comments comprises receiving at least one training article and corresponding key terms selected by a trusted user specifying whether at least one comment associated with an article is abusive; and   classifying a candidate object using the identified optimal model comprises automatically filtering at least one abusive comment.   
     
     
         30 . A system for performing comment classification of objects, comprising:
 a memory configured to store training data collected from at least one content reviewer device;   at least one processor configured to:   receive, from a first user, comment classifications related to one or more comments;   determine the first user is a trusted user based on the comment classifications;   receive, from the trusted user, flagged comments;   process the flagged comments to form at least one vectorized object based on the flagged comments;   train a plurality of models by applying a plurality of machine learning algorithms to the at least one vectorized object;   identify an optimal model from the plurality of models; and   classify other comments using the identified optimal model.   
     
     
         31 . The system of  claim 30 , wherein the first user is selected from the group consisting of an editor, an author, and a third-party user. 
     
     
         32 . The system of  claim 30 , wherein processing the flagged comments comprises using training data to form at least one vectorized object by performing at least one operation selected from a group consisting of stemming, tokenization, and n-gram analysis. 
     
     
         33 . The system of  claim 30 , wherein the plurality of machine learning algorithms comprises a support vector machine algorithm. 
     
     
         34 . The system of  claim 30 , wherein processing the flagged comments comprises processing supervised training data to form at least one vectorized object by processing training data using a plurality of machines operating in parallel. 
     
     
         35 . The system of  claim 30 , wherein:
 receiving flagged comments comprises receiving at least one pre-defined tag selected by first user as being associated with at least one training article; and   classifying other comments using the identified optimal model comprises automatically classifying a candidate article.   
     
     
         36 . The system of  claim 30 , wherein:
 receiving flagged comments comprises receiving at least one training article and corresponding key terms selected by a trusted user specifying whether at least one comment associated with an article is abusive; and   classifying a candidate object using the identified optimal model comprises automatically filtering at least one abusive comment.   
     
     
         37 . A tangible computer-readable storage medium including instructions for performing comment classification of objects, which, when executed by at least one processor, cause the processor to perform operations comprising:
 receiving, from a first user, comment classifications related to one or more comments;   determining the first user is a trusted user based on the comment classifications;   receiving, from the trusted user, flagged comments;   processing the flagged comments to form at least one vectorized object based on the flagged comments;   training a plurality of models by applying a plurality of machine learning algorithms to the at least one vectorized object;   identifying an optimal model from the plurality of models; and   classifying other comments using the identified optimal model.   
     
     
         38 . The computer-readable storage medium of  claim 37 , wherein the first user is selected from the group consisting of an editor, an author, and a third-party user. 
     
     
         39 . The computer-readable storage medium of  claim 37 , wherein the operations performed by the at least one processor further comprise using training data to form at least one vectorized object by performing at least one operation selected from a group consisting of stemming, tokenization, and n-gram analysis. 
     
     
         40 . The computer-readable storage medium of  claim 37 , wherein the plurality of machine learning algorithms comprises a support vector machine algorithm. 
     
     
         41 . The computer-readable storage medium of  claim 37 , wherein processing the flagged comments comprises processing supervised training data to form at least one vectorized object by processing training data using a plurality of machines operating in parallel.