Claim Missing Document
Check
Articles

Enhancing Tomato Leaf Disease Detection via Optimized VGG16 and Transfer Learning Techniques Siregar, Sandy Putra; Akbari, Imam; Poningsih, Poningsih; Wanto, Anjar; Solikhun, Solikhun
Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi) Vol 9 No 3 (2025): June 2025
Publisher : Ikatan Ahli Informatika Indonesia (IAII)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29207/resti.v9i3.6410

Abstract

Identification of tomato leaf disease remains difficult because standard approaches are frequently incorrect in identifying distinct signs. Convolutional Neural Networks (CNNs) perform well in image classification and pattern identification, although they are prone to overfitting. Thus, max pooling was employed to reduce dimensionality while retaining crucial information. This paper offers an improved CNN through hyperparameter tuning and compares it to Transfer Learning models such as InceptionV3, NASNetMobile, and VGG16, which were chosen for their efficiency and accuracy. The dataset comprises 7,178 photos classified as Healthy, Leaf Late Blight, Septoria Leaf Spot, and Yellow Leaf Curl Virus, collected from Kaggle.. The dataset is separated into three sections: training, validation, and testing, with a ratio of 70:15:15. The results of this study revealed that the proposed method achieved the highest accuracy of 98.24%. In the application of transfer learning, the inceptionV3 model achieved an accuracy of 96.94%, whereas NASNetMobile obtained 97.50%, and VGG16 showed an accuracy of 96.76%. The evaluation is based on accuracy, precision, recall, F1-score and Inference time to determine the optimum model for accuracy and computing efficiency. This project uses the proposed method and Transfer Learning Techniques to categorize illness images on tomato leaves. These findings will drive further research to improve tehe performance of the proposed method for foliar disease classification and comparable applications.
Quantum Perceptron in Predicting the Number of Visitors to E-Commerce Websites in Indonesian Solikhun, Solikhun; Carissa Arishandy, Dinda; Batubara, Ela Roza; Poningsih
Jurnal Sisfokom (Sistem Informasi dan Komputer) Vol. 14 No. 2 (2025): MEY
Publisher : ISB Atma Luhur

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32736/sisfokom.v14i2.2334

Abstract

In the current digital era, e-commerce has become the backbone of Indonesia's digital economy, which is experiencing rapid growth. However, competition in this industry is becoming increasingly fierce, indicating the importance of predicting the number of website visitors for an effective marketing strategy. Quantum Perceptron, the latest quantum computing innovation, promises a more accurate and efficient approach compared to conventional methods such as classical Perceptron. This research proposes the use of Quantum Perceptron to predict the number of visitors on large e-commerce platforms in Indonesia. The data used in the research is data on the number of e-commerce visitors obtained from the katadata.com website. Data from Shopee, Tokopedia, Lazada, Blibli, and Bukalapak were used to analyze and compare predictions with classical perceptron methods, showing the significant potential of Quantum Perceptron in supporting the development of more efficient business strategies. The research results show that the Quantum Perceptron algorithm can make predictions very well compared to the classical perceptron, proven by the Quantum Perceptron having a perfect accuracy of 100% with a total of 2 epochs while the classical perceptron has 100% accuracy with a total of 10 epochs. Quantum perceptron has better performance and shorter time, this can be seen from the smaller number of epochs.
Strategi Pendekatan Supervisi Pendidikan Kepala Sekolah dalam Pembinaan Kinerja Guru di MTs Hidayatullah Bintan Solikhun, Solikhun; Rahayu, Fitri; Gusfirullah, Icmi
RIGGS: Journal of Artificial Intelligence and Digital Business Vol. 4 No. 2 (2025): Mei - Juli
Publisher : Prodi Bisnis Digital Universitas Pahlawan Tuanku Tambusai

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31004/riggs.v4i2.971

Abstract

Penelitian ini bertujuan untuk mendeskripsikan strategi pendekatan supervisi pendidikan yang diterapkan oleh kepala sekolah dalam membina kinerja guru di MTs Hidayatullah Bintan. Penelitian ini menggunakan pendekatan kualitatif dengan metode studi kasus, di mana data diperoleh melalui observasi, wawancara mendalam, dan dokumentasi. Hasil penelitian menunjukkan bahwa kepala sekolah menerapkan kombinasi pendekatan supervisi direktif, non-direktif, dan kolaboratif secara adaptif sesuai dengan karakteristik guru dan kebutuhan pembinaan. Pelaksanaan supervisi dilakukan secara terstruktur melalui observasi kelas, pertemuan reflektif, dan pembinaan berkelanjutan. Supervisi tidak hanya meningkatkan kemampuan teknis guru dalam perencanaan, pelaksanaan, dan evaluasi pembelajaran, tetapi juga menumbuhkan sikap profesional, motivasi kerja, dan budaya kolaboratif di lingkungan sekolah. Faktor pendukung yang menonjol meliputi gaya kepemimpinan partisipatif kepala sekolah, budaya kerja religius, dan semangat guru untuk berkembang. Sementara itu, hambatan yang dihadapi antara lain keterbatasan waktu kepala sekolah, perbedaan kompetensi guru, dan minimnya pelatihan profesional berkelanjutan. Meski demikian, kepala sekolah mampu mengatasi kendala tersebut melalui penjadwalan prioritas supervisi dan kerja sama dengan pihak eksternal. Temuan ini menegaskan pentingnya peran strategis kepala sekolah sebagai supervisor pendidikan dalam mendorong peningkatan mutu kinerja guru dan pembelajaran secara menyeluruh.
Enhancing Lung Cancer Prediction Accuracy UsingQuantum-Enhanced K-Medoids with Manhattan Distance Solikhun, Solikhun; Pujiastuti, Lise; Wahyudi, Mochamad
MATRIK : Jurnal Manajemen, Teknik Informatika dan Rekayasa Komputer Vol. 24 No. 3 (2025)
Publisher : Universitas Bumigora

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30812/matrik.v24i3.4190

Abstract

Lung cancer is a leading cause of cancer-related deaths worldwide, and early detection plays a crucialrole in improving treatment outcomes. This study proposes an enhancement of the K-Medoids clusteringmethod by integrating a quantum computing approach using Manhattan distance to improveprediction accuracy for lung cancer diagnosis. The research was conducted using a publicly availablelung cancer dataset consisting of 309 patient records with 14 diagnostic attributes. Comparative experimentswere carried out between the classical K-Medoids and the quantum-enhanced K-Medoids, withperformance evaluated based on clustering accuracy, precision, recall, and F1-score. The results showthat the quantum-based method has the same accuracy as the classical method, namely 88%. Thissuggests that quantum-based clustering can match the accuracy of classical methods after adequatetraining, although consistency and parameter stability remain areas for further refinement. Furtherresearch is recommended to test the model on larger datasets and to explore real-world deployment inclinical decision support systems.
Optimizing convolutional neural network hyperparameters to enhance liver segmentation accuracy in medical imaging Purnama, Iwan; Windarto, Agus Perdana; Solikhun, Solikhun
International Journal of Electrical and Computer Engineering (IJECE) Vol 15, No 4: August 2025
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijece.v15i4.pp3876-3887

Abstract

Liver segmentation in medical imaging is a crucial step in various clinical applications, such as disease diagnosis, surgical planning, and evaluation of response to therapy, which require a high degree of precision for accurate results. This research focuses on increasing the accuracy of liver segmentation by optimizing hyperparameters in the convolutional neural network (CNN) model using the developed ResNet architecture. The uniqueness of this research lies in the application of hyperparameter optimization methods such as random search and Bayesian optimization, which allow broader and more efficient exploration than conventional approaches. The results show that the DeepLabV3Plus model (the proposed model) significantly outperforms the standard ResNet in the image segmentation task. DeepLabV3Plus shows excellent performance with an MIoU score of 0.965, a PA Score of 0.929, and a meager loss value of 0.011. These results show that DeepLabV3Plus is able to recognize and predict segmentation areas very accurately and consistently and minimize prediction errors effectively. In conclusion, the results of this study show a significant improvement in segmentation accuracy, with the optimized model providing better performance in the evaluation.
Enhancing Autonomous Vehicle Navigation in Urban Traffic Using CNN-Based Deep Q-Networks Windarto, Agus Perdana; Solikhun, Solikhun; Wanto, Anjar
Journal of Applied Data Sciences Vol 6, No 4: December 2025
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jads.v6i4.896

Abstract

This research proposes a CNN-based Deep Q-Network (CNN-DQN) model to enhance the navigation capabilities of autonomous vehicles in complex urban environments. The model integrates CNN for spatial abstraction with reinforcement learning to enable end-to-end decision-making based on high-dimensional sensor data. The primary objective is to evaluate the impact of CNN-DQN state abstraction on the quality and stability of the resulting policy. Using a grid-based simulator, the agent is trained on a synthetic dataset representing urban traffic scenarios. The CNN-DQN model consistently outperforms standard DQN in multiple metrics: cumulative reward increased by 14.3%, loss convergence accelerated by 22%, and mean absolute error (MAE) reduced to 0.028. Furthermore, the model achieved a Pearson correlation coefficient of 0.94 in predicted actions and demonstrated superior robustness under Gaussian noise perturbation, with reward loss limited to 6.18% compared to 18.7% in the baseline. Visualizations of CNN feature maps reveal spatial attention patterns that support efficient path planning. The action symmetry index confirms that the CNN-DQN agent exhibits consistent left-right decision behavior, validating its policy regularity. The novelty of this study lies in its combined use of deep spatial encoding and value-based reinforcement learning for structured, rule-based environments with real-time control implications. These findings indicate that CNN-enhanced reinforcement learning architectures can significantly improve autonomous navigation performance and robustness in dynamic urban settings.
Implementation of Random Forest Optimized with Ant Colony Optimization (ACO) for Breast Cancer Prediction Ht. Barat, Ade Ismiaty Ramadhona; Siregar, Sandy Putra; Poningsih, Poningsih; Windarto, Agus Perdana; Solikhun, Solikhun; Sembiring, Rahmat Widia
Journal of Computer System and Informatics (JoSYC) Vol 6 No 4 (2025): August 2025
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/josyc.v6i4.7116

Abstract

Breast cancer is a significant disease impacting women globally, highlighting the necessity for precise and dependable diagnostic models. This study aims to improve breast cancer prediction by optimizing the Random Forest algorithm using Ant Colony Optimization (ACO). This study uses datasets containing various cell characteristics to build and evaluate models. The ACO algorithm is applied to fine-tune the hyperparameters of the Random Forest model and improve its predictive performance. The experimental results showed that the optimized Random Forest model outperformed the baseline model in all evaluation metrics. The optimized model achieved an accuracy of 94.74%, precision of 97.92%, recall 90.38%, an F1 score of 92.93%, and an AUC score of 0, 9449 compared to the basic Random Forest model, with lower scores across all metrics. This improvement highlights the effectiveness of ACOs in improving model performance, especially in reducing false negatives, which are critical for medical diagnosis. This study demonstrates that ACO successfully fine-tunes Random Forest hyperparameters, achieving superior accuracy compared to baseline and outperforming previous optimization methods such as PSO. These findings confirm that the combination of Random Forest and ACO offers a powerful and effective approach to improving the accuracy of breast cancer predictions, making them a valuable tool for clinical decision-making.
New Innovation: Predicting Anemia with the K-Medoids Method and Quantum Computing Using Manhattan Distance Hartama, Dedy; Putri, Adelia; Solikhun, Solikhun
JST (Jurnal Sains dan Teknologi) Vol. 13 No. 2 (2024): July
Publisher : Universitas Pendidikan Ganesha

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.23887/jstundiksha.v13i2.83457

Abstract

The low accuracy of anemia diagnosis with the classical K-Medoids method shows the need for alternative, more effective techniques in processing medical record data. This research aims to analyze the effectiveness of the quantum computing approach as a solution to develop an anemia diagnostic method by integrating the K-Medoids algorithm and Manhattan distance calculation. This research is an experimental study with a comparative design. The research subjects comprised anemia patient medical record data covering 5 attributes and 1 target, with 20 samples taken from the Kaggle.com platform. Data collection was conducted using data mining techniques, while the instrument used was computational modeling software. The data was analyzed using the accuracy comparison method between the classical and quantum computing-based K-Medoids methods. The analysis results show that the quantum computing-based K-Medoids method can achieve 80% accuracy, which is equivalent to the classical K-Medoids method, but with higher data processing efficiency. This research confirms that integrating quantum computing in the K-Medoids method can be an alternative in diagnosing anemia, offering the potential for broader application to more complex medical record data. The implication of this research is the creation of opportunities for innovation in quantum computing-based medical decision support systems that are more efficient.
Application of Numerical Measure Variations in K-Means Clustering for Grouping Data Buaton, Relita; Solikhun, Solikhun
MATRIK : Jurnal Manajemen, Teknik Informatika dan Rekayasa Komputer Vol. 23 No. 1 (2023)
Publisher : Universitas Bumigora

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30812/matrik.v23i1.3269

Abstract

The K-Means Clustering algorithm is commonly used by researchers in grouping data. The main problem in this study was that it has yet to be discovered how optimal the grouping with variations in distance calculations is in K-Means Clustering. The purpose of this research was to compare distance calculation methods with K-Means such as Euclidean Distance, Canberra Distance, Chebychev Distance, Cosine Similarity, Dynamic TimeWarping Distance, Jaccard Similarity, and Manhattan Distance to find out how optimal the distance calculation is in the K-Means method. The best distancecalculation was determined from the smallest Davies Bouldin Index value. This research aimed to find optimal clusters using the K-Means Clustering algorithm with seven distance calculations based on types of numerical measures. This research method compared distance calculation methods in the K-Means algorithm, such as Euclidean Distance, Canberra Distance, Chebychev Distance, Cosine Smilirity, Dynamic Time Warping Distance, Jaccard Smilirity and Manhattan Distance to find out how optimal the distance calculation is in the K-Means method. Determining the best distance calculation can be seen from the smallest Davies Bouldin Index value. The data used in this study was on cosmetic sales at Devi Cosmetics, consisting of cosmetics sales from January to April 2022 with 56 product items. The result of this study was a comparison of numerical measures in the K-Means Clustering algorithm. The optimal cluster was calculating the Euclidean distance with a total of 9 clusters with a DBI value of 0.224. In comparison, the best average DBI value was the calculation of the Euclidean Distance with an average DBI value of 0.265.
Moral dan Etika Sains dalam Islam Solikhun, Solikhun; Rahman, Rahman; Futihatus Sirriyah; Imam Subekti
ULIL ALBAB : Jurnal Ilmiah Multidisiplin Vol. 4 No. 2: Januari 2025
Publisher : CV. Ulil Albab Corp

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.56799/jim.v4i2.7206

Abstract

Penelitian ini bertujuan mengetahui moral dan etika sains dalam Islam. Hasil dalam penelitian ini menunjukkan bahwa Islam sangat mendukung perkembangan sains, penggunaannya harus selalu berlandaskan prinsip-prinsip etika yang telah digariskan dalam Al-Qur'an dan Hadis, agar sains dapat memberikan manfaat yang sebesar-besarnya tanpa menimbulkan kerusakan. Dengan demikian, Islam mengajarkan bahwa sains yang berkembang dengan nilai moral dan etika dapat menjadi sarana untuk meningkatkan kualitas hidup, baik di dunia maupun di akhirat. Dalam pandangan Islam, ilmu pengetahuan yang tidak memperhatikan etika dan moral dapat menimbulkan berbagai dampak negatif, seperti penyalahgunaan ilmu, dehumanisasi, kerusakan lingkungan, ketidakadilan sosial, dan penyimpangan dari tujuan hidup yang lebih tinggi. Para ahli juga mengingatkan bahwa tanpa pengawasan moral, sains dapat berfungsi sebagai kekuatan yang merusak, bukan untuk kebaikan umat manusia. Oleh karena itu, Islam dan para pemikir etika menekankan pentingnya menghubungkan ilmu pengetahuan dengan prinsip-prinsip moral yang baik.