Claim Missing Document
Check
Articles

Found 6 Documents
Search

From Algorithms to Cures: AI's Impact on Drug Discovery Karimah, Fitrah; Amirah
Journal of Computer Science Application and Engineering (JOSAPEN) Vol. 1 No. 2 (2023): JOSAPEN - July
Publisher : PT. Lentera Ilmu Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.70356/josapen.v1i2.12

Abstract

This study explores the paradigm-shifting fusion of artificial intelligence (AI) and pharmaceuticals, heralding a new era of innovation in drug development. AI's transformative potential revolutionizes the traditionally arduous drug discovery process by seamlessly assimilating vast data volumes encompassing molecular structures, genetics, and disease pathways. This synergy expedites the identification of potential drug candidates with heightened precision and efficiency, propelling breakthrough treatments. The exploration navigates through AI-driven computational models, showcasing their role in expediting drug validation and optimization. AI's iterative learning enhances predictive capabilities, forecasting medication efficacy and safety profiles, thereby minimizing clinical trial risks and boosting success rates. Beyond acceleration, AI reshapes drug development strategies toward personalized medicine. Analyzing expansive patient datasets, AI tailors treatments based on genetic variations and disease characteristics, promising optimized therapeutic outcomes and minimized adverse effects, marking a departure from traditional healthcare approaches. The methodology employed various research techniques, including literature reviews, data collection, surveys, case studies, synthesis, and recommendations, offering comprehensive insights into AI's impact on drug discovery. In conclusion, the study emphasized AI's transformative potential in revolutionizing drug discovery, advocating for continued exploration and integration to optimize pharmaceutical research and development practices.
A Review for the Mechanism of Research Productivity Enhancement in the Higher Education Institution Sanmorino, Ahmad; Karimah, Fitrah
International Journal of Advanced Science Computing and Engineering Vol. 3 No. 1 (2021)
Publisher : SOTVI

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (294.333 KB) | DOI: 10.62527/ijasce.3.1.43

Abstract

The main purpose of this review is to find out the mechanism of research productivity enhancement proposed by each researcher in the papers they have published. The availability of these various mechanisms raises the desire of the authors to compare each mechanism. The focus of the review lies in the mechanism, characteristics, source of data, and evaluation methods used by each researcher. The review then jumps to the results obtained by each mechanism. The author also compares the types of data used by each researcher and the parties involved in the mechanism. There are some differences in the use of terminology between one to another mechanism, but in essence, it has the same goal, research productivity enhancement.
Preliminary Study for Cyber Intrusion Detection Using Machine Learning Approach Amirah; Karimah, Fitrah
Jurnal Sistem Informasi dan Teknik Informatika (JAFOTIK) Vol. 1 No. 1 (2023): JAFOTIK - February
Publisher : PT. Lentera Ilmu Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.70356/jafotik.v1i1.4

Abstract

This article discusses the importance of information system security in the current technological era and how the increasingly complex threat of cyber attacks demands a more sophisticated approach to detection and prevention. This initial study explores the potential of applying Machine Learning in cyber intrusion detection as a first step to developing detection systems that are adaptive and responsive to evolving threats. Through a methodology involving the collection of representative data on cyber attacks, data preparation, and Machine Learning model selection, this article describes the initial stages for understanding and testing the potential of this technology in the context of cyber security. Although it includes an example dataset, data preparation steps, and the selection of several Machine Learning algorithms, this study only gets to the model selection stage, while the model training process and performance evaluation are the focus of future work. The conclusions of this initial study emphasize the importance of selecting appropriate algorithms with specific features for effective intrusion detection against growing cyber threats.
Leveraging Open Data with Machine Learning Algorithms Amirah; Karimah, Fitrah
Jurnal Sistem Informasi dan Teknik Informatika (JAFOTIK) Vol. 1 No. 2 (2023): JAFOTIK - August
Publisher : PT. Lentera Ilmu Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.70356/jafotik.v1i2.19

Abstract

In the evolving landscape of technology, the amalgamation of open data and machine learning stands as a powerful catalyst for innovation. This study explores the dynamic synergy between these domains, where open data's accessibility and transparency converge with machine learning's pattern recognition and predictive capabilities. The fusion holds immense promise across diverse sectors, from healthcare to finance, urban planning, and environmental science. By leveraging advanced algorithms on openly available information, organizations can gain unprecedented insights into trends, correlations, and anomalies, fostering a culture of innovation. The methodology involves a comprehensive literature review, knowledge enrichment, case studies, and conclusion, providing a systematic approach to understanding the intersection of open data and machine learning. The results showcase practical applications in predictive policing, healthcare resource allocation, smart traffic management, and more. Each application is supported by relevant machine learning algorithms, emphasizing their role in addressing complex challenges. The study culminates with a simplified example of predictive policing using a Support Vector Machine (SVM) algorithm, showcasing its pseudocode and decision function equation. This example illustrates how machine learning can predict crime occurrences based on patrol data and historical crime rates. Overall, this fusion marks a pivotal chapter in technological progress and societal advancement.
Short Communication: Drug Discovery Advancements in The Artificial Intelligence Era Karimah, Fitrah; Ahmad
Jurnal Sistem Informasi dan Teknik Informatika (JAFOTIK) Vol. 2 No. 1 (2024): JAFOTIK - February
Publisher : PT. Lentera Ilmu Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.70356/jafotik.v2i1.29

Abstract

Artificial Intelligence (AI) is significantly transforming drug discovery by enhancing efficiency and reducing costs. Traditional drug development has been slow and expensive, but AI's integration accelerates the process by predicting molecular interactions, identifying drug candidates, and optimizing formulations. Recent advancements highlight AI's role in molecular interaction prediction, target identification, lead optimization, and toxicity prediction. AI models, particularly deep learning algorithms, improve drug efficacy predictions and streamline virtual screening. They also address challenges in toxicity prediction by analyzing historical data to foresee adverse reactions, thus reducing late-stage failures. Despite its potential, AI faces challenges such as data quality and model interpretability. Future developments include advancements in explainable AI and the integration with personalized medicine, promising a revolution in creating more effective, tailored treatments while minimizing side effects. This short communication emphasizes AI's growing impact and the transformative opportunities it presents in modern medicine.
E-learning platforms for boarding schools Karimah, Fitrah
Jurnal Sistem Informasi dan Teknik Informatika (JAFOTIK) Vol. 2 No. 2 (2024): JAFOTIK - August
Publisher : PT. Lentera Ilmu Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.70356/jafotik.v2i2.40

Abstract

This study examines the adoption and impact of e-learning platforms in boarding schools. Boarding schools present unique educational challenges, making them a compelling case for exploring the effectiveness of digital learning solutions. The study begins with a comprehensive needs assessment involving stakeholders such as administrators, teachers, students, and IT staff. Key areas of focus include technological infrastructure, internet accessibility, and digital literacy levels. The study then moves to the implementation phase, piloting an e-learning platform in select schools to evaluate its effectiveness. Results indicate significant challenges, including outdated infrastructure and varying levels of digital literacy. However, e-learning platforms show promise in enhancing educational outcomes by offering interactive content and improving communication. The study compares different platforms, such as Google Classroom, Moodle, and Microsoft Teams, to determine the most suitable options for these institutions. Findings suggest that while some platforms are more user-friendly, others offer robust customization and scalability, making them better suited for the specific needs of boarding schools.