Claim Missing Document
Check
Articles

Found 2 Documents
Search

ANALISIS SENTIMEN DATA TWITTER TERKAIT CHATGPT MENGGUNAKAN ORANGE DATA MINING Pahtoni, Tri Yuli; Jati, Handaru
Jurnal Teknologi Informasi dan Ilmu Komputer Vol 11 No 2: April 2024
Publisher : Fakultas Ilmu Komputer, Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.25126/jtiik.20241127276

Abstract

Perkembangan teknologi bergerak begitu cepat, diikuti dengan popularitas media sosial yang semakin meluas. Platform media sosial mampu membangun profil big data pengguna, dengan melacak setiap aktivitas seperti partisipasi, pengiriman pesan, dan kunjungan situs Web. Saat ini banyak orang sering membagikan kritik terhadap sesuatu melalui platform media sosial seperti Facebook, Twitter, Instagram, dan lainnya. Sehingga perlu diketahui bagaimana komentar dari pengguna media sosial yang menghasilkan reaksi masyarakat terhadap chatGPT yang dirilis oleh OpenAI. Banyaknya komentar di Twitter menyebabkan sulitnya mengetahui kecenderungan respon masyarakat. Tujuan dari penelitian ini yaitu melakukan analisis sentimen postingan publik di Twitter untuk memberikan wawasan tentang sikap dan persepsi orang tentang suatu peristiwa yang terjadi. Penelitian ini memberikan ilustrasi peran Twitter dalam menampung postingan pengguna Twitter terkait chatGPT. Hasil penelitian ini dapat digunakan oleh pemangku kepentingan untuk menentukan kebijakan dalam penggunaan chatGPT. Penelitian ini menganalisis sebanyak 5.192 postingan tweet bahasa Inggris dan 641 tweet bahasa Indonesia, mulai dari tanggal 27 April hingga 8 Mei 2023. Tanggapan positif, negatif dan netral diolah menggunakan perangkat lunak orange data mining, yaitu tools machine learning, data mining, dan visualisasi data. Hasil menunjukan bahwa chatGPT mendapatkan tanggapan netral berbahasa Inggris dengan nilai sebesar 54,72%, tanggapan positif sebesar 31,64%, dan tanggapan negatif sebesar 13,64%. Hasil analisis sentimen berbahasa Indonesia tidak jauh berbeda, dengan nilai tanggapan netral sebesar 63,96%, tanggapan positif 23,56%, dan tanggapan negatif 12,48%. Sehingga dapat disimpulkan bahwa, rilisnya chatGPT mayoritas publik memberikan tanggapan netral atau tidak terdapat penolakan.AbstractTechnological developments move so fast, followed by the increasingly widespread popularity of social media. Social media platforms can build big-data profiles of users by tracking every activity such as participation, messaging, and website visits. Currently, many people often share criticism of something through social media platforms, such as Facebook, Twitter, Instagram, and others. So it is necessary to know how comments from social media users generate public reactions to chatGPT released by OpenAI. A lot of comments on Twitter make it difficult to know the trend of people's responses. This study aims to analyze the sentiment of public postings on Twitter to provide insight into people's attitudes and perceptions of an event that has occurred. This research illustrates Twitter's role in accommodating Twitter user posts regarding chatGPT. The results of this study can be used by stakeholders in making policies on the use of chatGPT. This study analyzed 5,192 posts in English and 641 tweets in Indonesian from April 27 to May 8, 2023. Positive, negative, and neutral responses were processed using orange data mining software, namely machine learning tools, data mining, and data visualization. The results show that chatGPT received neutral responses in English with a value of 54.72%, positive responses of 31.64%, and negative responses of 13.64%. The results of sentiment analysis in Indonesian were not much different, with neutral responses of 63.96%, positive responses of 23.56%, and negative responses of 12.48%. So it can be concluded that after the release of chatGPT, the majority of the public gave neutral responses or no rejection.
Design and Development of Industrial Practice Monitoring and Assessment Systems using Tsukamoto Fuzzy Logic Pahtoni, Tri Yuli; Arifin, Fatchul
Elinvo (Electronics, Informatics, and Vocational Education) Vol. 8 No. 2 (2023): November 2023
Publisher : Department of Electronic and Informatic Engineering Education, Faculty of Engineering, UNY

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21831/elinvo.v8i2.57669

Abstract

Vocational high schools are given flexibility for their students to carry out direct learning in the industry as part of the practical education activities of implementing student skills. The implementation of industrial practice requires a special way to find out and monitor each student's activities so that the achievements of the implementation of industrial practice can be carried out properly.  The implementation of industrial work practice assessment has several assessment criteria. These criteria include attendance, neatness, attitude, skills, and knowledge. The problems found in the assessment system are still done manually so that the effectiveness is minimal. This study aims to create a system that can monitor and assess the implementation of industrial practices.  The system developed will be tested as a medium for monitoring and assessing industrial practices.  This research uses Fuzzy Tsukamoto's logic approach as a scoring logic  model and  uses the waterfall method as a development model consisting of analysis, design, coding, and testing. The results of the research conducted resulted in a system that can monitor and assess the implementation of industry practices.  The test was carried out by 24 people consisting of guidance teachers and students. Testing is done by testing aspects of functionality and aspects of usability. Based on the test results, the functionality aspect scored 100% (very feasible) and the usage aspect got a score of 84.8% (very feasible)