Article Per Year (5 Year)
p-Index From 2021 - 2026
0.751
P-Index
This Author published in this journals
All Journal Journal of Social Science Utilizing Technology Journal of Social Science Utilizing Technology Journal of Computer Science Advancements
Sahirin, Rohmat
Unknown Affiliation
Religion Arts Humanities Computer Science & IT Education Environmental Science Languange, Linguistic, Communication & Media Mathematics Social Sciences Other

Published : 4 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search
Journal : Journal of Computer Science Advancements

 1 
Implementation of Deep Learning in a Voice Recognition System for Virtual Assistants Apriyanto, Apriyanto; Sahirin, Rohmat; Bradford, Snyder
Journal of Computer Science Advancements Vol. 2 No. 6 (2024)
Publisher : Yayasan Adra Karima Hubbi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.70177/jsca.v2i6.1533

Abstract

Voice recognition technology has become a vital component in virtual assistants, enabling more natural and efficient user interactions. However, traditional voice recognition systems face challenges in accurately interpreting diverse accents, dialects, and background noise, which can limit their usability. This study investigates the implementation of deep learning techniques to improve the accuracy and adaptability of voice recognition systems within virtual assistant applications. The research aims to enhance voice recognition performance by leveraging deep learning models that can process complex speech patterns and adapt to varied linguistic nuances. A convolutional neural network (CNN) architecture combined with recurrent neural networks (RNN) was used to train the voice recognition model on a large, diverse dataset of audio samples. The dataset included multiple languages, accents, and noisy environments to test the robustness of the model. Results indicate a 25% improvement in word error rate (WER) and a significant increase in recognition accuracy across diverse voice inputs compared to traditional voice recognition systems. The model demonstrated high adaptability, accurately interpreting speech in varying acoustic conditions, thus improving user experience with virtual assistants. These findings suggest that deep learning can significantly enhance voice recognition systems, offering more reliable performance in real-world applications. Implementing deep learning models in voice recognition systems can bridge the gap between human and machine communication, making virtual assistants more accessible and user-friendly.
Sentiment Analysis on Social Media Using Data Mining for Mapping Community Satisfaction Usup, Usup; Sahirin, Rohmat; Lucas, Laura; Qingjun, Chu
Journal of Computer Science Advancements Vol. 3 No. 1 (2025)
Publisher : Yayasan Adra Karima Hubbi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.70177/jsca.v3i1.1536

Abstract

Social media has become a significant platform for individuals to express opinions, including satisfaction and dissatisfaction with services and policies, making it a valuable source of community sentiment data. Understanding public sentiment can assist policymakers and organizations in responding to community needs effectively. This study aims to conduct sentiment analysis on social media using data mining techniques to map community satisfaction levels. By analyzing sentiment patterns, this research seeks to provide actionable insights for improving public services and enhancing community engagement. The research applies data mining methodologies, including text mining and machine learning algorithms, to analyze posts and comments collected from various social media platforms. Sentiment classification was performed using natural language processing (NLP) and a supervised machine learning approach to categorize sentiments as positive, neutral, or negative. The model was trained on a large dataset and validated to ensure accuracy in sentiment detection. Results indicate that social media sentiment analysis can reliably reflect community satisfaction trends, with findings showing 70% positive, 15% neutral, and 15% negative sentiments regarding local services. The study concludes that data mining for sentiment analysis provides a robust method for assessing community satisfaction on social media, offering a real-time understanding of public opinion. By implementing this approach, organizations and policymakers can identify areas of improvement and proactively address community concerns, ultimately fostering a responsive and community-centered approach to public service.  
Co-Authors Apriyanto, Apriyanto Bradford, Snyder Khan, Mariam Lucas, Laura Ngr. Putu Raka Novandra Asta Qingjun, Chu Tu, Nguyen Minh Usup, Usup