Penggunaan Metode Fuzzy Logic untuk Pemantauan Sentimen Brand pada Media Sosial

Beki Subaeki, Fatkhan Gunawan, Aldy Rialdy Atmadja

Abstract


The purpose of this research is to monitor the sentiments of a brand and classify it into positive,  negative or neutral sentiments. The steps of research have started from collecting data, indexing, searching and weighting process. Data are collected by crawling data from social media, such as Facebook and Twitter. After collecting data, then weighting process is done with a fuzzy logic method, where the fuzzy set is determined based on the highest number of positive and negative words in a sentence. Weighting process is calculated from TF (Term Frequency) which is the number of words that sought in the document. From the results, TF can be used to find the fuzzy set value and the number of positive or negative sentiments in a document. Mamdani method used to calculate the value of the final sentiment. From the calculation results, it can be shown that the average of sentiment analysis is 63.15%.

 Keywords:  Information, Sentiment analysis, brand, fuzzy logic, social media.

 


Full Text:

PDF

Refbacks

  • There are currently no refbacks.




Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.

 Indexed By:

                        

                  

                        

                     

        

                     

                         

         

             

                                                                    

                                      

 

 

 

 

        

                 

                                     

 

     

  

Free counters!

  Web Analytics Made Easy - StatCounter

 

 

View My Stats

 

Creative Commons License

Query is licensed under a Creative Commons Attribution 4.0 International License.