Claim Missing Document
Check
Articles

Found 3 Documents
Search
Journal : Science, Technology, and Communication Journal

Performance comparison of the Naive Bayes algorithm and the k-NN lexicon approach on Twitter media sentiment analysis Azhar, Azhar; Masruroh, Siti Ummi; Wardhani, Luh Kesuma; Okfalisa, Okfalisa
Science, Technology and Communication Journal Vol. 3 No. 2 (2023): SINTECHCOM Journal (February 2023)
Publisher : Lembaga Studi Pendidikan and Rekayasa Alam Riau

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59190/stc.v3i2.229

Abstract

Sentiment analysis or opinion mining is a natural language that processes words to find out opinions, attitudes, or moods about certain things. Word processing in this study related to the process of classification in textual documents, which was classified into three classes, positive, negative, and neutral. Data obtained from social media Twitter were related to netizens' comments as many as 1000 comments. These data were crawled using keywords of the “Pilpres2019” and “Jokowi”. This study compared the performance of the Naive Bayes and k-Nearest Neighbor (k-NN) algorithms with the lexicon approach in classification. The aim of this study was to compare the level of accuracy, precision, and recall of Naive Bayes and the k-NN algorithm with the lexicon approach. From the evaluation, we concluded that the combination of the k-NN algorithm and the lexicon approach could improve accuracy in this sentiment analysis case. Generally, the k-NN algorithm with lexicon approach in which the k value is k = 5 has better performance with a 77% of accuracy level, followed by Naive Bayes with an accuracy of 81% of accuracy level.
Chatbot AI Riau tourism towards society 5.0 success Ningrum, Melia Vivi; Ahmad, Sayyidina Anshari; Belantara, M Abyan; Al-Sa’ban, M Dafa; Santana, Qistan Alif; Alfitri, Rahmat; Okfalisa, Okfalisa
Science, Technology, and Communication Journal Vol. 5 No. 1 (2024): SINTECHCOM Journal (October 2024)
Publisher : Lembaga Studi Pendidikan dan Rekayasa Alam Riau

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59190/stc.v5i1.283

Abstract

Riau Province stands out as one of the most rapidly developing regions in Indonesia, showcasing significant advancements economy, population, and tourism. The strategic positioning of Riau Province along the world's busiest routes, establishes this province as a focal point for diverse activities from multiple nations. Furthermore, the natural resources of Riau Province provide it as a significant business hub by substantial activities. This undoubtedly impacts the cross section both business and tourist visitors from local and international. Unfortunately, it is found that information pertaining to Riau tourism is under-addressed the visitors’ need. Therefore, the chatbot artificial intelligent (AI) is developed to deliver comprehensive information about tourism in Riau, encompassing tourist attractions, culinary, and merchandise destinations. This chatbot is adopted a prototyping approach for software development with natural language processing, employing tools such as Xampp, Visual Studio Code, PHP for programming language, and the Botman for the library. This chatbot is then integrated with a customized website platform utilizing an accessible API. The questions are derived from commonly asked questions on the tourism platform, limited into five tourist attractions, culinary options, or merchandise’s spot. To evaluate, a black-box and user acceptance testing techniques will be employed to verify that the application operates as intended and receives favorable feedback from users. This chatbot grows into the smart and responsive tool tailored to tourism needs in Riau. This tool supports the Society 5.0 that concerns on human centric, technology, and resilience social and community while preserving the cultural elements of Riau’s Malay heritage.
UI/UX design thinking adoption for integrated AI point-of-sale system (Case study: Plastic Poultry Wholesale Store) Okfalisa, Okfalisa; Fahruddin, Fahruddin; Setiaji, Haris; Pratama, M Farhan Aulia; Finaldhi, Harry; Delifah, Nur
Science, Technology, and Communication Journal Vol. 5 No. 3 (2025): SINTECHCOM Journal (June 2025)
Publisher : Lembaga Studi Pendidikan dan Rekayasa Alam Riau

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59190/stc.v5i3.311

Abstract

The development of artificial intelligence (AI) technology drives the need for a point of sales (POS) system that is not only efficient, but can also provide adaptive information according to the user's sudden wishes. This research implements the design thinking method in designing a user interface (UI/UX) for a smart POS system integrated with conversational AI features. A case study was conducted at a Plastic Poultry Wholesale Store to gain in-depth insights related to field needs and operational challenges faced. The design thinking method was chosen because of its user-centered approach, through the stages of empathize, define, ideate, prototype, and testing, it is hoped that the final results obtained can be aligned with the concrete needs of users, so that the output of the system that has been designed will not be abandoned, but will always be used. In this design, the implementation of conversational AI is used to enhance the user experience through a virtual assistant feature that is able to answer dynamic questions according to the wishes of the user, so that users can freely explore any information in detail related to their overall business performance. The implementation results show that this system not only increases operational efficiency, but also improves user experience through more intuitive interactions when they want to see their business performance. This research contributes to integrating AI technology with a user-centered design approach for smart, responsive, and adaptive POS system solutions.