The rapid advancement of deep learning significantly increases computational demands, making performance optimization essential for model scalability and deployment. While numerous studies optimize neural network architectures, the effect of different programming paradigms on computational efficiency remains insufficiently explored. This study aims to compare Object-Oriented Programming (OOP) and Data-Oriented Programming (DOP) paradigms in TensorFlow-based deep learning workflows, focusing on their performance across four processing phases: build, compile, train, and evaluate, under a controlled experimental environment with repeated iterations and systematic measurements. Both paradigms are implemented using identical Convolutional Neural Network (CNN) architectures trained on the CIFAR-100 image dataset over thirty controlled experimental iterations. A custom profiler integrating the Python System and Process Utilities (psutil) and NVIDIA Management Library (pynvml) monitors real-time system performance, capturing CPU and GPU utilization as well as memory usage. The results reveal that DOP achieves better resource efficiency with lower memory usage (549.98 MB versus 676.25 MB), higher GPU utilization (64.68% versus 61.08%), and faster evaluation execution (1.50 seconds versus 2.59 seconds), while also attaining higher model accuracy (32.38% versus 28.08%). In contrast, OOP benefits from TensorFlow’s Sequential API optimizations, resulting in faster training times but greater CPU and memory consumption. These findings highlight that DOP provides superior runtime efficiency and offers practical benefits for performance-critical deep learning applications.
Digital transformation in the financial industry has encouraged organizations to adopt application integration systems to improve operational efficiency and effectiveness. However, not all information system implementation projects succeed in meeting expectations, particularly in guarantee institutions with complex business processes. This study evaluates the success of an application integration system project in a guarantee company using a combined approach based on the DeLone and McLean (2003) Information System Success Model and Critical Success Factors (CSF). By explicitly incorporating CSF variables as antecedents of system quality, information quality, and service quality, this study extends the conventional application of the DeLone and McLean model by integrating key managerial perspectives into the evaluation of integration success. A quantitative method was employed using survey data collected from 120 respondents of the Penjaminan Application Integration (PAI) system at PT XYZ. The data were analyzed using Partial Least Squares Structural Equation Modeling (PLS-SEM). The results indicate that CSF variables, including top management support, internal communication, user training, and project risk management, significantly affect system quality, information quality, and service quality. Furthermore, these three quality dimensions within the D&M model notably influence the intention to use and user satisfaction, which ultimately impact the perceived net benefits for the organization. In conclusion, considering both managerial and system quality perspectives provides a more comprehensive understanding of the success factors in application integration projects. These findings can serve as a basis for improving IT project implementation strategies in the guarantee sector and other industries with similar characteristics.
Food commodities are essential in developing countries, such as Indonesia. The government rules the food commodity prices in every province. Yet, somehow, there are some issues in certain provinces. Data science and statistics techniques can help the government control food commodity prices. The proposed method applied to predict the commodity price index in East Java is the STACKEL K-MEANS method. This proposed method is a collaborative framework that utilizes cluster analysis and stacking ensemble learning to predict data. Cluster analysis is performed first, using the distance that suits time series data, which is Dynamic Time Warping. Two clusters are formed from each commodity (Rice, Oil, and Flour). Then, the stacking model consists of a base learner and a meta learner. The base learner models used are Ridge Regression, Random Forest, and Support Vector Regression, while the meta learner is Light Gradient Boosting Method. To optimize the parameter, we used a grid search. Following the evaluation process, we compare the proposed method with auto ARIMA from Python. In training and testing data, The proposed method yields superior results to the ARIMA model across all three error metrics: MAPE, MAE, and RMSE. The following scores for flour commodities are 0.042% compared to 0.328%, 4.715 compared to 37.57, and 6.34 compared to 523.99. For rice commodities, the scores are 0.261% compared to 0.392%, 31.585 compared to 48.142, and 41.92 compared to 56.068 For oil commodities, the scores are 0.185% compared to 0.250%, 33.02 compared to 47.571, and 39.35 compared to 56.060.
Balinese language is a local language that is widely use and spoken by Balinese people including in social media. However, the nuances of these politeness levels are often lost in informal digital communication and there is a significant lack of computational model to automatically classify them, especially for low-resource language like Balinese. The primary objective of this study is to evaluate the performance of the Multinomial Naive Bayes method combined with Term Frequency-Inverse Document Frequency (TF-IDF) feature extraction, Chi-square feature selection, and Synthetic Minority Oversampling Technique (SMOTE) in classifying Balinese language levels. The dataset for this study consists of 1,314 annotated social media posts and comments, primarily sourced from Instagram. The annotation was conducted by a Balinese language expert to categorize text into six levels that represent varying degrees of politeness and formality. These levels are alus singgih (polite, used for respecting others), alus sor (polite, used for self-humbling), alus mider (polite, used for both respecting others and self-humbling), alus madia (an intermediate level of politeness), basa andap (casual, commonly used in everyday life), and basa kasar (impolite, often used during arguments or toward animals). The experimental results showed that the model successfully achieved an accuracy of 96.53% on the training data and 61.45% on the test data. Additionally, hyperparameter tuning revealed that the Multinomial Naive Bayes model with 2,720 selected features and SMOTE oversampling achieved an accuracy of 91.78%, significantly outperforming the baseline model without feature selection and oversampling, which obtained only 64.93% accuracy.
Stunting is a form of chronic nutritional deficiency in toddlers and remains a major public health concern due to its impact on child growth and development. Efforts to reduce its prevalence continue to be strengthened in Indonesia, particularly in Sumatra Province. This study aims to evaluate the accuracy of a logistic regression model and three machine learning models—decision tree, random forest, and support vector machine (SVM)—in classifying stunting prevalence. The response variable is defined as the prevalence of stunting among toddlers, categorized into two classes: exceeding the national target and not exceeding the national target, based on the 2024 national threshold. Although classification models can provide accurate predictions, they often lack interpretability. Therefore, this study applies the SHAP method to the best-performing machine learning model to identify the key factors influencing stunting. The use of Shapley values is justified through the uniqueness theorem, which establishes it as the only attribution method satisfying desirable fairness properties. SHAP values are employed to explain the model by referencing both the trained model and the underlying data. The results show that the random forest model achieves the highest accuracy (90.00%), outperforming the other models. SHAP analysis reveals that Underweight is the most influential predictor contributing to stunting prevalence in Sumatra Province. These findings highlight the relevance of machine learning interpretability in supporting policy decisions for stunting reduction.
