Unpacking Public Perceptions of Qris with Twitter Data: A Vader And LDA Methodology
QRIS, a mobile payment transaction system standardized by Bank Indonesia, has become the subject of extensive public discourse on Twitter. Employing VADER for sentiment analysis and LDA for topic modeling, this study aims to capture the nuanced perspectives of the Indonesian public toward QRIS. Our...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
P3M Politeknik Negeri Banjarmasin
2023-12-01
|
| Series: | Jurnal ELTIKOM: Jurnal Teknik Elektro, Teknologi Informasi dan Komputer |
| Subjects: | |
| Online Access: | https://eltikom.poliban.ac.id/index.php/eltikom/article/view/742 |
| _version_ | 1797316763991408640 |
|---|---|
| author | Dzakiya Ishmatul Ulya Anang Kunaefi Dwi Rolliawati Bayu Adhi Nugroho |
| author_facet | Dzakiya Ishmatul Ulya Anang Kunaefi Dwi Rolliawati Bayu Adhi Nugroho |
| author_sort | Dzakiya Ishmatul Ulya |
| collection | DOAJ |
| description | QRIS, a mobile payment transaction system standardized by Bank Indonesia, has become the subject of extensive public discourse on Twitter. Employing VADER for sentiment analysis and LDA for topic modeling, this study aims to capture the nuanced perspectives of the Indonesian public toward QRIS. Our methodology includes real human validation for tweets that have been initially labeled by VADER. Our unique contributions lie in employing a mixed-methods approach for comprehensive sentiment and topic analysis, as well as making our dataset publicly available for future research. We achieve a sentiment labeling accuracy of 81.66%, uncovering that 67% of the sentiment towards QRIS is positive, 28.2% negative, and 4.17% neutral. Positive tweets mostly cover six dominant topics with a value of 0.488037, whereas negative sentiments are concentrated around three dominant topics with a value of 0.383938. These findings not only affirm the generally positive public response towards QRIS but also highlight areas requiring attention for its continued success. Our study translates these insights into actionable recommendations, aiming to provide a multidimensional understanding that stakeholders can leverage for system enhancement. This study serves as a foundation for future works in sentiment analysis and public opinion mining related to financial technologies, particularly in the Indonesian context. |
| first_indexed | 2024-03-08T03:24:04Z |
| format | Article |
| id | doaj.art-000cb7bf129f48a089ef09893b76e128 |
| institution | Directory Open Access Journal |
| issn | 2598-3245 2598-3288 |
| language | English |
| last_indexed | 2024-03-08T03:24:04Z |
| publishDate | 2023-12-01 |
| publisher | P3M Politeknik Negeri Banjarmasin |
| record_format | Article |
| series | Jurnal ELTIKOM: Jurnal Teknik Elektro, Teknologi Informasi dan Komputer |
| spelling | doaj.art-000cb7bf129f48a089ef09893b76e1282024-02-12T04:09:20ZengP3M Politeknik Negeri BanjarmasinJurnal ELTIKOM: Jurnal Teknik Elektro, Teknologi Informasi dan Komputer2598-32452598-32882023-12-017214515910.31961/eltikom.v7i2.742698Unpacking Public Perceptions of Qris with Twitter Data: A Vader And LDA MethodologyDzakiya Ishmatul Ulya0Anang Kunaefi1Dwi Rolliawati2Bayu Adhi Nugroho3UIN Sunan Ampel Surabaya, IndonesiaUIN Sunan Ampel Surabaya, IndonesiaUIN Sunan Ampel Surabaya, IndonesiaUIN Sunan Ampel Surabaya, IndonesiaQRIS, a mobile payment transaction system standardized by Bank Indonesia, has become the subject of extensive public discourse on Twitter. Employing VADER for sentiment analysis and LDA for topic modeling, this study aims to capture the nuanced perspectives of the Indonesian public toward QRIS. Our methodology includes real human validation for tweets that have been initially labeled by VADER. Our unique contributions lie in employing a mixed-methods approach for comprehensive sentiment and topic analysis, as well as making our dataset publicly available for future research. We achieve a sentiment labeling accuracy of 81.66%, uncovering that 67% of the sentiment towards QRIS is positive, 28.2% negative, and 4.17% neutral. Positive tweets mostly cover six dominant topics with a value of 0.488037, whereas negative sentiments are concentrated around three dominant topics with a value of 0.383938. These findings not only affirm the generally positive public response towards QRIS but also highlight areas requiring attention for its continued success. Our study translates these insights into actionable recommendations, aiming to provide a multidimensional understanding that stakeholders can leverage for system enhancement. This study serves as a foundation for future works in sentiment analysis and public opinion mining related to financial technologies, particularly in the Indonesian context.https://eltikom.poliban.ac.id/index.php/eltikom/article/view/742natural language processingsentiment analysisopinion miningtopic modelingmobile paymentqris |
| spellingShingle | Dzakiya Ishmatul Ulya Anang Kunaefi Dwi Rolliawati Bayu Adhi Nugroho Unpacking Public Perceptions of Qris with Twitter Data: A Vader And LDA Methodology Jurnal ELTIKOM: Jurnal Teknik Elektro, Teknologi Informasi dan Komputer natural language processing sentiment analysis opinion mining topic modeling mobile payment qris |
| title | Unpacking Public Perceptions of Qris with Twitter Data: A Vader And LDA Methodology |
| title_full | Unpacking Public Perceptions of Qris with Twitter Data: A Vader And LDA Methodology |
| title_fullStr | Unpacking Public Perceptions of Qris with Twitter Data: A Vader And LDA Methodology |
| title_full_unstemmed | Unpacking Public Perceptions of Qris with Twitter Data: A Vader And LDA Methodology |
| title_short | Unpacking Public Perceptions of Qris with Twitter Data: A Vader And LDA Methodology |
| title_sort | unpacking public perceptions of qris with twitter data a vader and lda methodology |
| topic | natural language processing sentiment analysis opinion mining topic modeling mobile payment qris |
| url | https://eltikom.poliban.ac.id/index.php/eltikom/article/view/742 |
| work_keys_str_mv | AT dzakiyaishmatululya unpackingpublicperceptionsofqriswithtwitterdataavaderandldamethodology AT anangkunaefi unpackingpublicperceptionsofqriswithtwitterdataavaderandldamethodology AT dwirolliawati unpackingpublicperceptionsofqriswithtwitterdataavaderandldamethodology AT bayuadhinugroho unpackingpublicperceptionsofqriswithtwitterdataavaderandldamethodology |