Shofwatul Uyun

Shofwatul Uyun

Auto-generated profile, based on publicly available information

Shofwatul Uyun

UIN Sunan Kalijaga Yogyakarta

Research Focus & Metrics

Academic Career Length

16Years (2011-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 April 2011 in TELKOMNIKA (Telecommunication Computing Electronics and Control)
Verified
Journal Article
Open access
PDF file
S‘Shofwatul ‘Uyun
Shofwatul ‘Uyun
IRImam Riadi
Imam Riadi
As the education fees are becoming more expensive, more students apply for scholarships. Consequently, hundreds and even thousands of applications need to be handled by the sponsor. To solve the problems, some alternatives based on several attributes (criteria) need to be selected. In order to make a decision on such fuzzy problems, Fuzzy Multiple Attribute Decision Making (FMDAM) can be applied. In this study, Unified Modeling Language (UML) in FMADM with TOPSIS and Weighted Product (WP) methods is applied to select the candidates for academic and non-academic scholarships at Universitas Islam Negeri Sunan Kalijaga. Data used were a crisp and fuzzy data. The results show that TOPSIS and Weighted Product FMADM methods can be used to select the most suitable candidates to receive the scholarships since the preference values applied in this method can show applicants with the highest eligibility.

1 February 2018 in International Journal of Electrical and Computer Engineering (IJECE)
Verified
Journal Article
Open access
PDF file
SUShofwatul Uyun
Shofwatul Uyun
LCLina Choridah
Lina Choridah
The very dense breast of mammogram image makes the Radiologists often have difficulties in interpreting the mammography objectively and accurately. One of the key success factors of computer-aided diagnosis (CADx) system is the use of the right features. Therefore, this research emphasizes on the feature selection process by performing the data mining on the results of mammogram image feature extraction. There are two algorithms used to perform the mining, the decision tree and the rule induction. Furthermore, the selected features produced by the algorithms are tested using classification algorithms: k-nearest neighbors, decision tree, and naive bayesian with the scheme of 10-fold cross validation using stratified sampling way. There are five descriptors that are the best features and have contributed in determining the classification of benign and malignant lesions as follows: slice, integrated density, area fraction, model gray value, and center of mass. The best classification results based on the five features are generated by the decision tree algorithm with accuracy, sensitivity, specificity, FPR, and TPR of 93.18%; 87.5%; 3.89%; 6.33% and 92.11% respectively.

8 May 2023 in Jurnal Nasional Teknik Elektro dan Teknologi Informasi
Verified
Journal Article
Open access
PDF file
QFQorry Aina Fitroh
Qorry Aina Fitroh
S'Shofwatul 'Uyun
Shofwatul 'Uyun
Benign and malignant cancers are the most common skin cancer types. It is essential to know skin cancer symptoms with an early diagnosis to provide an appropriate treatment and reduce the mortality rate. Dermoscopic image is one of the diagnostic media that many researchers have developed. It provides more optimal results in computational-based diagnosis than visual detection. Deep learning and transfer learning are two models that have been used successfully in computational-based analysis, although optimization is still needed. In this study, transfer learning was used to separate dermoscopic images of skin cancer into two categories: benign and malignant. This study used 2,000 images to increase previous research’s accuracy conducted on the Kaggle public dataset containing 3,297 images. Two pretrained models, namely VGG-16 and residual network (ResNet)-50, were compared and used as feature extractors. Fine-tuning was conducted by adding a flatten layer, two dense layers with the ReLU activation function, and one dense layer with the Softmax activation function to classify images into two categories. Hyperparameter tuning on the optimizer, batch size, learning rate, and epoch were performed to get each model’s best performance parameter combination. Before hyperparameter tuning, the model was tested by resizing the input image using different sizes. The results of model testing on the VGG-16 model gave the best performance at an image size of 128 × 128 pixels with a combination of Adam parameters as an optimizer, batch size of 64, learning rate of 0.001, and epoch of 10 with an accuracy value of 91% and loss of 0.2712. The ResNet-50 model yielded better accuracy of 94% and a loss of 0.2198 using the optimizer parameter RMSprop, batch size of 64, learning rate of 0.0001, and epoch of 100. The results indicate that the proposed method provides good accuracies and can assist dermatologist in the early detection of skin cancer.

31 December 2022 in JOIV : International Journal on Informatics Visualization
Verified
Journal Article
Open access
PDF file
DFDony Fahrudy
Dony Fahrudy
S'Shofwatul 'UYun
Shofwatul 'UYun
Class-imbalanced data with high attribute dimensions in datasets frequently contribute to issues in a classification process as this can affect algorithms’ performance in the computing process because there are imbalanced numbers of data in each class and irrelevant attributes that must be processed; therefore, this needs for some techniques to overcome the class-imbalanced data and feature selection to reduce data complexity and irrelevant features. Therefore, this study applied random oversampling (ROs) method to overcome the class-imbalanced data and two feature selections (information gain and forward selection) compared to determine which feature selection is superior, more effective and more appropriate to apply. The results of feature selection then were used to classify the student graduation by creating a classification model of Naïve Bayes algorithm. This study indicated an increase in the average accuracy of the Naïve Bayes method without the ROs preprocessing and the feature selection (81.83%), with the ROs (83.84%), with information gain with 3 selected features (86.03%) and forward selection with 2 selected features (86.42%); consequently, these led to increasing accuracy of 4.2% from no pre-processing to information gain and 4.59% from no pre-processing to forward selection. Therefore, the best feature selection was the forward selection with 2 selected features (GPA of the 8th semester and the overall GPA), and the ROs and both feature selections were proven to improve the performance of the Naïve Bayes method.

26 February 2021 in Jurnal Teknologi dan Sistem Komputer
Verified
Journal Article
Open access
PDF file
MIMuhammad Edi Iswanto
Muhammad Edi Iswanto
MSMaria Ulfah Siregar
Maria Ulfah Siregar
S'Shofwatul 'UYun
Shofwatul 'UYun
MNMuhammad Taufiq Nuruzzaman
Muhammad Taufiq Nuruzzaman
The selection of students' interests based on the 2013 curriculum (K-13) is carried out before students start learning in class X. Accuracy in its determination is required to ensure that students learn according to their interests and talents. This study applies three DSS methods, namely profile matching, SAW, and a combination of both, to provide accurate recommendations for determining these students' interests. The three methods are compared using the same alternatives and criteria to find the most dominant method. The results of this study indicate that the application of SPK can assist PPDB activities with an accuracy of 79.2 %. In determining interest for students, the combination method is the most dominant, with an accuracy of 78 %. The application of DSS not only helps the specialization process to be faster but also accurate. This is indicated by only 6 out of 122 students who chose specialization based on the DSS recommendation getting a score below the KKM.

29 December 2023 in IJID (International Journal on Informatics for Development)
Verified
Journal Article
Open access
PDF file
DRDiniati Ruaika
Diniati Ruaika
SUShofwatul Uyun
Shofwatul Uyun
Image classification is a fundamental task in vision recognition that aims to understand and categorize an image under a specific label. Image classification needs to produce a quick, economical, and reliable result. Convolutional Neural Networks (CNN) have proven effective for image analysis. However, problems arise due to factors such as the model’s quality, unbalanced training data, overfitting, and layers’ complexity. ResNet50 is a transfer learning-based convolutional neural network model frequently used in many areas, including Lepidopterology. Studies have shown that ResNet50 performs with lower accuracy than other models for classifying butterflies. Therefore, this study aims to optimise the accuracy of ResNet50 using an augmentation approach and ensemble deep learning for butterfly image classification. This study used a public dataset of butterflies from Kaggle. The dataset contains 75 different butterfly species, 9.285 training images, 375 testing images, and 375 validation images. A sequence of transformation functions was applied. The ensemble deep learning was constructed by incorporating ResNet50 with CNN. To measure ResNet50 optimisation, the experimental results of the original dataset and the proposed methods were compared and analysed using evaluation metrics. The research revealed that the proposed method provided better performance with an accuracy of 95%.