Abstract:

Aiming at the characteristics of the multiaspect performance appeared in the automotive product reviews,this paper proposed a novel method for recognizing the multiple aspects of performance about car comment text based on multilabel learning.Firstly,appropriate words were selected as features by multilabel text feature selection method combined with the text mining technology,and then,the unstructured document corpus are transformed into structured multilabel dataset.After that,we finished marking one or more aspect tags for the unrecognized comment text with four multilabel classification methods.Finally,the recognition accuracy of multiple aspects was assessed by eight multilabel evaluation metrics.On the Sina car review corpus,experimental results indicate the subset accuracy reaches up to 95%.Hence,our method was feasible for recognizing the multiple aspects of automobile reviews.

Key words: multilabel learning, text processing, car reviews, multiaspect recognition