Diyas Puspandari
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Diyas Puspandari has a total of 14 co-authors. Here are the top co-authors by number of shared publications:
- Yuliant Sibaroni
Telkom University
3 shared publications - Anisa Herdiani
Telkom University, Bandung
1 shared publication - Ariqoh Novitasari
Telkom University
1 shared publication - Fahrul Raykhan Ridho
Telkom University
1 shared publication I Lukmana
School of Computing, Telkom University, Bandung, 40257, Indonesia
1 shared publication- Ida Widia
Universitas Pendidikan Indonesia
1 shared publication
Latest Publications
- Open Access
- 7 September 2024
- Published by Forum Kerjasama Pendidikan Tinggi (FKPT) in Building of Informatics, Technology and Science (BITS)
- Vol. 6 (2) , 1034-1044
Multi-Aspect Sentiment Analysis Using Elman Recurrent Neural Network (ERNN) Method for TripAdvisor App User Reviews
TripAdvisor is the world's largest travel platform that assists 463 million travelers each month in making their trips the best they can be. Users of TripAdvisor can provide reviews, comments, and ratings of travel destinations. However, reviews on TripAdvisor are considered insufficient in helping prospective travelers understand the strengths and weaknesses of a hotel. Therefore, a multiaspect sentiment analysis of TripAdvisor reviews on hotels was conducted to identify commonly discussed rating aspects among visitors and to determine specific evaluations. In this study, the Elman Recurrent Neural Network (ERNN) method was employed to build a classification system for multiaspect sentiment analysis of user reviews on the TripAdvisor application. The aspects examined in this research include Service, Cleanliness, Location, Value, Rooms, and Overall Experience, aiming to provide insights into the hotels under consideration. The results indicate that the ERNN method can deliver superior outcomes in multiaspect sentiment analysis of TripAdvisor hotel reviews. The ERNN model's performance in multiaspect sentiment analysis shows optimal accuracies: 81.35% for Service, 98.71% for Cleanliness, 74.87% for Location, 93.84% for Value and 71.52% for Rooms. These findings can assist travelers in better understanding the strengths and weaknesses of accommodations.
- 7 September 2024
- Published by Forum Kerjasama Pendidikan Tinggi (FKPT) in Building of Informatics, Technology and Science (BITS)
- Vol. 6 (2) , 1034-1044
Open AccessMulti-Aspect Sentiment Analysis Using Elman Recurrent Neural Network (ERNN) Method for TripAdvisor App User Reviews
TripAdvisor is the world's largest travel platform that assists 463 million travelers each month in making their trips the best they can be. Users of TripAdvisor can provide reviews, comments, and ratings of travel destinations. However, reviews on TripAdvisor are considered insufficient in helping prospective travelers understand the strengths and weaknesses of a hotel. Therefore, a multiaspect sentiment analysis of TripAdvisor reviews on hotels was conducted to identify commonly discussed rating aspects among visitors and to determine specific evaluations. In this study, the Elman Recurrent Neural Network (ERNN) method was employed to build a classification system for multiaspect sentiment analysis of user reviews on the TripAdvisor application. The aspects examined in this research include Service, Cleanliness, Location, Value, Rooms, and Overall Experience, aiming to provide insights into the hotels under consideration. The results indicate that the ERNN method can deliver superior outcomes in multiaspect sentiment analysis of TripAdvisor hotel reviews. The ERNN model's performance in multiaspect sentiment analysis shows optimal accuracies: 81.35% for Service, 98.71% for Cleanliness, 74.87% for Location, 93.84% for Value and 71.52% for Rooms. These findings can assist travelers in better understanding the strengths and weaknesses of accommodations.
- Open Access
- 7 September 2024
- Published by Forum Kerjasama Pendidikan Tinggi (FKPT) in Building of Informatics, Technology and Science (BITS)
- Vol. 6 (2) , 825-834
Multi-aspect Sentiment Analysis of Shopee Application Reviews using RNN Method and Query Expansion Ranking
Online shopping using e-commerce is a common activity society does in this digital era. Shopee is one of the well-known e-commerce in Indonesia. There are a lot of e-commerce platforms that can easily be accessed through mobile applications like Google Play Store. Users are allowed to review and rate the application they have downloaded. The reviews from the users become an opportunity for e-commerce companies to advance their performances and services. To enhance the understandability of user reviews, a system that can efficiently analyze the sentiment is needed. This study aims to design and establish a system that can perform sentiment analysis on the selected aspects. Sentiment classification is implemented by using the Recurrent Neural Network (RNN) algorithm and Query Expansion Ranking feature selection to classify Shopee application reviews into two classes, which are positive and negative. Feature selection is used to reduce less useful features so that the classification model conducts the classification process optimally and more efficiently. In conclusion, the evaluation results based on an 80:20 data split ratio indicate that the RNN achieves the highest accuracy of 95% in the delivery cost aspect, 93% in the delivery speed aspect, and 86% in the application access aspect.
- 7 September 2024
- Published by Forum Kerjasama Pendidikan Tinggi (FKPT) in Building of Informatics, Technology and Science (BITS)
- Vol. 6 (2) , 825-834
Open AccessMulti-aspect Sentiment Analysis of Shopee Application Reviews using RNN Method and Query Expansion Ranking
Online shopping using e-commerce is a common activity society does in this digital era. Shopee is one of the well-known e-commerce in Indonesia. There are a lot of e-commerce platforms that can easily be accessed through mobile applications like Google Play Store. Users are allowed to review and rate the application they have downloaded. The reviews from the users become an opportunity for e-commerce companies to advance their performances and services. To enhance the understandability of user reviews, a system that can efficiently analyze the sentiment is needed. This study aims to design and establish a system that can perform sentiment analysis on the selected aspects. Sentiment classification is implemented by using the Recurrent Neural Network (RNN) algorithm and Query Expansion Ranking feature selection to classify Shopee application reviews into two classes, which are positive and negative. Feature selection is used to reduce less useful features so that the classification model conducts the classification process optimally and more efficiently. In conclusion, the evaluation results based on an 80:20 data split ratio indicate that the RNN achieves the highest accuracy of 95% in the delivery cost aspect, 93% in the delivery speed aspect, and 86% in the application access aspect.
- Open Access
- 2 February 2024
- Published by Brazilian Journals in Journal of Law and Sustainable Development
- Vol. 12 (2) , e3224
PLAIN LANGUAGE INTERVENTION TO IMPROVE PUBLIC UNDERSTANDING OF LEGAL LANGUAGE: DESCRIPTIVE ANALYSIS OF CONTROVERSIAL ARTICLES IN THE JOB CREATION LAW
Objective: This article discusses the plain language intervention to enhance public understanding of legal language, particularly regarding the controversial provisions of the Job Creation Law. Methodology: The research was conducted through a survey distributed to individuals from various educational and professional backgrounds. Theoretical framework: This research uses several simple language approaches such as, Bivins (2008); Mukherjee, et al. (2017); Wojcik (2013); and Wydick & Sloan (2019). This framework is used to intervene in legal language to make it simpler and easier to understand. Results and Conclusion: The questionnaire results indicated that 82.6% of respondents had a better understanding of the simplified (plain language) version of legal text, while only 17.4% understood the original legal language version. This correlated with a decrease in the lexical density index by 9.07 (25.11%) after the implementation of plain language. These findings demonstrate that the general public faces difficulties in comprehending legal language and expresses the desire for improvements in the use of simpler and more understandable legal language. Legal documents tend to be challenging to understand due to the extensive use of technical terms, long and complex sentences, and ambiguity in expressing legal meanings. Originality/Value: In the context of the Job Creation Law, the reduction in lexical density through the implementation of plain language has helped improve public understanding of legal language. However, there is a need for awareness and improvement in the use of legal language by policymakers and authorities to ensure that legal language is accessible to the general public.
- 2 February 2024
- Published by Brazilian Journals in Journal of Law and Sustainable Development
- Vol. 12 (2) , e3224
Open AccessPLAIN LANGUAGE INTERVENTION TO IMPROVE PUBLIC UNDERSTANDING OF LEGAL LANGUAGE: DESCRIPTIVE ANALYSIS OF CONTROVERSIAL ARTICLES IN THE JOB CREATION LAW
Objective: This article discusses the plain language intervention to enhance public understanding of legal language, particularly regarding the controversial provisions of the Job Creation Law. Methodology: The research was conducted through a survey distributed to individuals from various educational and professional backgrounds. Theoretical framework: This research uses several simple language approaches such as, Bivins (2008); Mukherjee, et al. (2017); Wojcik (2013); and Wydick & Sloan (2019). This framework is used to intervene in legal language to make it simpler and easier to understand. Results and Conclusion: The questionnaire results indicated that 82.6% of respondents had a better understanding of the simplified (plain language) version of legal text, while only 17.4% understood the original legal language version. This correlated with a decrease in the lexical density index by 9.07 (25.11%) after the implementation of plain language. These findings demonstrate that the general public faces difficulties in comprehending legal language and expresses the desire for improvements in the use of simpler and more understandable legal language. Legal documents tend to be challenging to understand due to the extensive use of technical terms, long and complex sentences, and ambiguity in expressing legal meanings. Originality/Value: In the context of the Job Creation Law, the reduction in lexical density through the implementation of plain language has helped improve public understanding of legal language. However, there is a need for awareness and improvement in the use of legal language by policymakers and authorities to ensure that legal language is accessible to the general public.
- Open Access
- 28 November 2023
- Published by IntSys Research in Techno.Com
- Vol. 22 (4) , 833-842
Penerapan Metode Long Short-Term Memory dan Word2Vec dalam Analisis Sentimen Ulasan pada Aplikasi Ferizy
Tranportasi merupakan hal yang penting bagi masyarakat dalam mobilitas sehari-hari. Karena memiliki peranan penting dan dapat memudahkan kehidupan masyarakat, pemerintah mulai mengoptimalkan pembangunan sarana transportasi dan memulai inovasi digital, salah satunya pada moda transportasi laut. Perseroan Terbatas Angkutan Sungai, Danau, dan Penyeberangan Indonesia (PT ASDP) meluncurkan aplikasi Ferizy pada Google Playstore. Dalam inovasi ini, sentimen masyarakat dapat membantu untuk mengetahui kepuasan, kekurangan, saran, dan kritik. Terkait hal tersebut maka diperlukan analisis sentimen untuk memahami maksud ulasan. Analisis ini mengekstrak data ulasan lalu mengolah data tekstual secara otomatis untuk mendapatkan makna sentimen yang terkandung dalam ulasan. Penelitian ini mengimplementasikan klasifikasi Long Short-Term Memory (LSTM) dan ekstraksi fitur Word2Vec variasi skip-gram serta CBOW pada dataset ulasan aplikasi Ferizy. Hasil pengujian dari model menghasilkan nilai akurasi sebesar 88,20% untuk variasi skip-gram dan 74,20% untuk variasi CBOW.
- 28 November 2023
- Published by IntSys Research in Techno.Com
- Vol. 22 (4) , 833-842
Open AccessPenerapan Metode Long Short-Term Memory dan Word2Vec dalam Analisis Sentimen Ulasan pada Aplikasi Ferizy
Tranportasi merupakan hal yang penting bagi masyarakat dalam mobilitas sehari-hari. Karena memiliki peranan penting dan dapat memudahkan kehidupan masyarakat, pemerintah mulai mengoptimalkan pembangunan sarana transportasi dan memulai inovasi digital, salah satunya pada moda transportasi laut. Perseroan Terbatas Angkutan Sungai, Danau, dan Penyeberangan Indonesia (PT ASDP) meluncurkan aplikasi Ferizy pada Google Playstore. Dalam inovasi ini, sentimen masyarakat dapat membantu untuk mengetahui kepuasan, kekurangan, saran, dan kritik. Terkait hal tersebut maka diperlukan analisis sentimen untuk memahami maksud ulasan. Analisis ini mengekstrak data ulasan lalu mengolah data tekstual secara otomatis untuk mendapatkan makna sentimen yang terkandung dalam ulasan. Penelitian ini mengimplementasikan klasifikasi Long Short-Term Memory (LSTM) dan ekstraksi fitur Word2Vec variasi skip-gram serta CBOW pada dataset ulasan aplikasi Ferizy. Hasil pengujian dari model menghasilkan nilai akurasi sebesar 88,20% untuk variasi skip-gram dan 74,20% untuk variasi CBOW.
- Open Access
- 1 March 2019
- Published by IOP Publishing in Journal of Physics: Conference Series
- Vol. 1192 (1) , 012029
Cyberbullying identification on twitter using random forest classifier
Cyberbullying is a repeated act that is done to harass other people online in which it is common to happen amongst adolescents although there are some who don't understand the harm of cyberbullying. The incomprehension of cyberbullying itself can cause ignorance of cyberbullying's harmful effects. By using a classification technique, a cyberbullying identification can be done on social media to provide a better understanding regarding this particular activity. In Indonesia, researches that involved cyberbullying have not solved the problem related to identification of tweets containing cyberbullying and non-cyberbullying content. Therefore, this research proposed the usage of Random Forest's classification to help in identifying cyberbullying on one of the affected social media, Twitter. The proposed method was applied on 50 group of tweets using training/test split. The usage of Random Forest in this research is optimized by the consideration of general rules from previous study with some needed adjustments. Random Forest results in the highest F1-Score which is 0.90. The wrong prediction is caused by the inconsistency in which some rules that determine whether a group of tweets are considered to be cyberbullying the most are justified as non-cyberbullying in some cases as well as the other way around.
- 1 March 2019
- Published by IOP Publishing in Journal of Physics: Conference Series
- Vol. 1192 (1) , 012029
Open AccessCyberbullying identification on twitter using random forest classifier
Cyberbullying is a repeated act that is done to harass other people online in which it is common to happen amongst adolescents although there are some who don't understand the harm of cyberbullying. The incomprehension of cyberbullying itself can cause ignorance of cyberbullying's harmful effects. By using a classification technique, a cyberbullying identification can be done on social media to provide a better understanding regarding this particular activity. In Indonesia, researches that involved cyberbullying have not solved the problem related to identification of tweets containing cyberbullying and non-cyberbullying content. Therefore, this research proposed the usage of Random Forest's classification to help in identifying cyberbullying on one of the affected social media, Twitter. The proposed method was applied on 50 group of tweets using training/test split. The usage of Random Forest in this research is optimized by the consideration of general rules from previous study with some needed adjustments. Random Forest results in the highest F1-Score which is 0.90. The wrong prediction is caused by the inconsistency in which some rules that determine whether a group of tweets are considered to be cyberbullying the most are justified as non-cyberbullying in some cases as well as the other way around.