Text Mining and Data Information Analysis for Network Public Opinion
DOI:
https://doi.org/10.5334/dsj-2019-007Keywords:
Network public opinion, text mining, emotional tendency, pattern matching, support vector machineAbstract
Network public opinion information is massive and complex, and it is difficult to make effective use of manual means. In this paper, a method based on pattern matching and machine learning (PMML) was proposed to analyze the emotional tendencies of network public opinion. Firstly, the key words in public opinion were extracted, then the patterns were extracted and matched, and the emotional tendencies of words were calculated to obtain the pattern sequence vectors. Support vector machine (SVM) classifier was used to classify emotional tendencies. The Internet reviews of Meituan hotel were taken as the experimental subject. PMML method was found to have a high classification performance, with a maximum accuracy of 86.75%. It suggested the effectiveness of the proposed method. Then PMML method was used to classify the emotional tendencies of the collected reviews, and the results showed that the negative emotional tendency was greater than the positive tendency, which showed the inadequacy of Meituan hotel. The experiments in this paper provide some basis for the application of PMML in sentiment analysis of Internet public opinion.
Published
Issue
Section
License
Copyright (c) 2019 The Author(s)

This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms. If a submission is rejected or withdrawn prior to publication, all rights return to the author(s):
-
Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
-
Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
-
Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work.
Submitting to the journal implicitly confirms that all named authors and rights holders have agreed to the above terms of publication. It is the submitting author's responsibility to ensure all authors and relevant institutional bodies have given their agreement at the point of submission.
Note: some institutions require authors to seek written approval in relation to the terms of publication. Should this be required, authors can request a separate licence agreement document from the editorial team (e.g. authors who are Crown employees).