cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
Muhammad Ridha
Contact Email
mridha747@gmail.com
Phone
+6282267370810
Journal Mail Official
jadimijournal@outlook.com
Editorial Address
Echa Residence 2, Blok D, No. 7, Jalan Mendalo Laut, Jambi Luar Kota, Kab. Muaro Jambi, Jambi 36361
Location
Kab. muaro jambi,
Jambi
INDONESIA
JADIMI (Journal of AI-Driven Informatics and Management Information)
Published by PT. Lentera Akademia Indonesia
ISSN : -     EISSN : 31647138     DOI : -
Core Subject :
JADIMI (Journal of AI-Driven Informatics and Management Information) is a peer-reviewed and open-access scientific journal published triannually (April, August and December) by Lentera Akademia Indonesia. The journal serves as an academic platform for scholars, researchers, and practitioners to disseminate high-quality research, case studies, and literature reviews focusing on the application and development of artificial intelligence (AI) in informatics and management information systems. Each manuscript submitted to JADIMI undergoes a rigorous double-blind peer-review process conducted by qualified national and international reviewers to ensure the publication of only the highest quality scholarly work. The journal emphasizes a balanced approach between theoretical development and practical implementation to promote impactful research in the AI and management ecosystem. Submissions must be written in English, original, and not previously published or currently under consideration elsewhere. The editorial board strictly enforces adherence to the journal’s manuscript guidelines and formatting standards. Non-compliance will result in desk rejection prior to review. Through its commitment to academic excellence and ethical integrity, JADIMI aspires to contribute significantly to the global discourse on AI in informatics and management information systems.
Arjuna Subject : -
Articles 7 Documents
 1 
A Hybrid Grey Wolf Optimizer-Tabu Search Approach for Multi-Day Tourist Itinerary Recommendation Based on Content-Based Filtering Benedict Brian Joel Purba; Jian Hazel Sitorus; Dimas Putra Nugroho; Saskia Putri Ananda
Journal of AI-Driven Informatics and Management Information Vol. 1 No. 1 (2026): April - August
Publisher : Journal of AI-Driven Informatics and Management Information

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Planning a multi-day city trip is a hard combinatorial problem: a tourist must choose which attractions to visit, in what order, and on which day, while respecting opening hours, travel time, a daily time budget, and a personal preference and budget profile. The authors present an end-to-end recommendation system for Jakarta that couples a content-based filtering stage with a route optimization stage cast as a Tourist Trip Design Problem (TTDP). The first stage scores candidate attractions using TF-IDF similarity, a Bayesian weighted rating, a category-based budget filter, and a Maximal Marginal Relevance step for diversity. The second stage searches over visit permutations with a time-budget decoder that inserts a lunch break and treats closing hours as a hard constraint. As the optimizer, the authors introduce a hybrid Grey Wolf Optimizer with an embedded Tabu Search local search (GWO–TS), adapted from a continuous formulation to the discrete permutation space using swap sequences. GWO–TS replaces a previously deployed genetic–particle-swarm hybrid as the default optimizer and is exposed through a REST API that powers a live web application. The dataset is built entirely from open sources like OpenStreetMap, the Massive-STEPS check-in corpus, Google Places, Wikipedia, and OSRM—yielding 166 operational attractions and 181 hotels across eight spatial clusters. The authors describe the formulation, the algorithm, and an evaluation protocol that compares GWO–TS against genetic algorithm, particle swarm optimization, and their hybrid over one-, three-, and five-day scenarios using fitness, a user satisfaction score, constraint violations, and runtime.
Daily ISPU Classification in DKI Jakarta Based on Gas Pollutant Parameters Using Machine Learning Andien Swesty; Nayla Raissa Putri; Muhammad Rafi Naufal Hilmy; Vicky Prasetya; Muhammad Pramudya Miftah
Journal of AI-Driven Informatics and Management Information Vol. 1 No. 1 (2026): April - August
Publisher : Journal of AI-Driven Informatics and Management Information

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Air pollution is a persistent environmental problem in urban areas, causing serious impacts on public health and environmental quality. Growing transportation activity, industrial output, and population density lead to daily fluctuations in pollutant concentrations. The Air Pollution Standard Index (ISPU) is used as an indicator of air quality based on several pollutant gas parameters, such as PM2.5, PM10, SO2, CO, O3, and NO2. Machine learning approaches can help classify air pollution levels from historical data patterns. This study examines and compares four machine learning algorithms, namely Logistic Regression, Decision Tree, Random Forest, and XGBoost, to classify the daily air pollution level in DKI Jakarta based on ISPU categories. The dataset used is DKI Jakarta air quality data consisting of 5,538 observations, evaluated under two data-splitting scenarios (70:15:15 and 80:10:10) to test model consistency. Model performance was evaluated using accuracy, precision, recall, and F1-score metrics. The test results show that the 80:10:10 scenario delivers better performance, with XGBoost achieving the highest performance and a test accuracy of 93.68%, outperforming Random Forest, Decision Tree, and Logistic Regression. Based on feature importance analysis, the O3 and PM10 parameters contribute the most to the classification process. These findings indicate that ensemble-learning-based algorithms, particularly XGBoost with a larger training-data proportion, can effectively capture the relationships among pollutant parameters. This approach can support the development of data-driven air quality classification systems, with future research recommended to incorporate weather variables, real-time data streams, further hyperparameter tuning, and deep learning approaches to improve classification accuracy.
Comparative Analysis of Machine Learning with Regression Algorithms in Housing Price Prediction Nayaka Ivana Putra; Kesya Noventa Salsabila; Dimas Pramudya; Naufal Abdirrahman Faiz; Khansa Maritza Runa; Muhammad Ridha
Journal of AI-Driven Informatics and Management Information Vol. 1 No. 1 (2026): April - August
Publisher : Journal of AI-Driven Informatics and Management Information

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Housing price prediction serves as a critical instrument for property industry stakeholders to determine fair market values. This research aims to perform a comprehensive comparative analysis of four machine learning algorithms: Linear Regression, Decision Tree, Random Forest, and Extreme Gradient Boosting (XGBoost). The study utilizes the Melbourne Housing Dataset, comprising 34,857 data rows and 21 features. Experiments were conducted across three primary scenarios: (1) XGBoost pipeline without imputation, (2) model comparison with a 70:15:15 data ratio, and (3) model comparison with an 80:10:10 data ratio. The methodology encompasses systematic preprocessing phases, including ANOVA testing (F-Score > 50) for feature significance, context-based missing value imputation, and selection of the top 10 features based on Pearson correlation coefficients. The results demonstrate that ensemble and boosting-based algorithms deliver superior performance, with XGBoost achieving the highest R² value of 0.7678 at the 80% training data ratio. This research contributes to determining the optimal dataset configuration and model selection for property price estimation, offering minimal error rates for both developers and potential buyers. Furthermore, the findings highlight the critical importance of non-linear algorithms in capturing the complex dynamics of the property market. This study offers actionable technical recommendations for industry stakeholders seeking to enhance the objectivity and precision of their financial decision-making processes through data-driven approaches.
Comparation of Lightweight CNN and Transformer-Based Models for Mammogram Image Classification on CBIS-DDSM Abdullah; Muhammad Varel Arifianta; Frenwin; Gena Darma
Journal of AI-Driven Informatics and Management Information Vol. 1 No. 1 (2026): April - August
Publisher : Journal of AI-Driven Informatics and Management Information

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Detecting malignancy in mammograms requires models capable of recognizing local lesion features while preserving broader breast-tissue context. This study compares three lightweight architectures, DenseNet121, EfficientNetV2-S, and MobileViT-S, for binary classification of benign and malignant lesions in the Curated Breast Imaging Subset of the Digital Database for Screening Mammography (CBIS-DDSM). Region-of-interest images were converted to single-channel grayscale inputs of 224×224 pixels, enhanced using contrast-limited adaptive histogram equalization, and augmented through geometric and intensity transformations. Patient-level stratified group five-fold cross-validation was used to prevent data leakage. Probabilities from the five fold-specific models were averaged through cross-validation bagging and evaluated on an independent test set of 298 images form 145 patients. EfficientNetV2-S achieved the highest accuracy, macro-F1 score, and specificity at 0.6913, 0.6837, and 0.7412, respectively. DenseNet121 achieved the highest sensitivity, ROC-AUC, and PR-AUC at 0.6953, 0.7446, and 0.7127, respectively, indicating fewer false-negative predictions. MobileViT-S achieved 0.6510 accuracy and 0.7368 ROC-AUC with 4.94 million parameters, 2.8250 GFLOPs per model, and an end-to-end bagging latency of 11.8658 ms per image, compared with 15.0013 ms for EfficientNetV2-S and 16.3840 ms for DenseNet121. Overall, EfficientNetV2-S provided the best classification balance, DenseNet121 was preferable when malignancy sensitivity was prioritized, and MobileViT-S offered the greatest computational efficiency for resource-constrained devices. The standardized evaluation framework supports reproducible comparisons across architectures while accounting for patient-level variation. These findings show that model selection should consider predictive performance, computational cost, and the clinical consequences of classification errors.
Controlled Comparison of Multimodal Hybrid Deep Learning for Benign-Malignant Mammogram Classification on CBIS-DDSM Marshall Rasendria Mahendra; Farhan Muamar Fawwaz; Naisya Aghis Nabila; Muhammad Rakha
Journal of AI-Driven Informatics and Management Information Vol. 1 No. 1 (2026): April - August
Publisher : Journal of AI-Driven Informatics and Management Information

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Breast cancer is the leading cause of cancer death among women, and mammography remains the first line of screening. Many deep learning studies on CBIS-DDSM report accuracy above 95%, yet they partition the archive at image level, so views of one patient land in both training and test sets and the figures cannot be trusted. This study builds a controlled comparison of two image representations, the whole mammogram and the lesion region of interest, under one identical training recipe. Six models are trained on each representation: EfficientNet-B3, ConvNeXt-Tiny, a gated hybrid fusion of the two with the Convolutional Block Attention Module, two variants that fuse seven clinical descriptors through a perceptron, and a Mammo-CLIP backbone pretrained on mammograms. The official CBIS-DDSM partition is enforced at patient level and yields 2,148 training, 261 validation and 448 test mammograms from 1,104, 122 and 234 disjoint patients, and the BI-RADS assessment is dropped to avoid circular prediction. On the test set the strongest model, Hybrid Fusion+CBAM+Clinical, reaches an AUC of 0.856 at an accuracy of 78.1% on regions of interest, and an AUC of 0.845 at an accuracy of 78.4% on whole images. Clinical metadata is the largest single source of gain, close to six AUC points on both representations. After its BatchNorm adaptation is corrected, Mammo-CLIP recovers to an AUC of 0.789 and becomes the best image-only whole-image backbone. Grad-CAM shows that attention settles on the lesion rather than on artifacts, and the reported range is what remains once leakage is removed.
Streaming Content Trends for Digital Business Strategy Celvan Putra Dani; Ivan Batara; Firza Septian
Journal of AI-Driven Informatics and Management Information Vol. 1 No. 1 (2026): April - August
Publisher : Journal of AI-Driven Informatics and Management Information

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

The rapid growth of digital platforms has reshaped business strategies, with streaming services such as Netflix, Disney+, and Amazon Prime producing vast amounts of user and content data. This study investigates streaming content trends as a foundation for digital business strategy, emphasizing how genre analysis can guide short video promotion in 2026. The dataset originates from the Netflix Titles dataset on Kaggle, snapshot date March 2024, comprising 8,807 titles with 12 variables including release year, rating, duration, genre classification, and textual descriptions. Historical coverage spans 1925–2021, with the target variable defined as genre (listed_in). Class distribution is imbalanced, dominated by Comedy, Drama, and International Movies. Preprocessing involved imputation, categorical encoding, and TF-IDF vectorization to transform textual features into numerical form. Three analytical scenarios were applied. The first used the K-Nearest Neighbors (KNN) algorithm to classify genres, achieving 75.3% accuracy, 50.1% precision, 52.5% recall, and 44.8% F1-score, with confusion matrices revealing overlaps between Comedy and Drama. The second employed a Greedy Algorithm to heuristically select optimal content, prioritizing high ratings, recent release years, and shorter durations; top selections included The Silent Horizon (Sci-Fi & Fantasy, score 5.32), Family Bonds (Children & Family, score 4.21), and Comedy Nights (Comedy, score 5.10). The third scenario applied linear regression forecasting, yielding coefficients between 1.2–2.1, low forecast errors (RMSE 3.5–5.0, MAE 2.8–3.9), and predicted 2026 counts of 145 Comedy, 130 Action & Adventure, 120 Children & Family, 105 Romantic Drama, and 95 Sci-Fi & Fantasy titles, with uncertainty estimates reported at the 95% confidence level. Results confirm clustering around shorter durations and recurring themes such as family, love, survival, and discovery, reinforcing the suitability of short-form content for marketing. The integration of machine learning, heuristic selection, and forecasting demonstrates the potential of data-driven approaches to inform digital business strategies.
Customer Personality Data for Predicting Consumer Response in Digital Business Firza Septian; Yupriadi; Dini Pratiwi
Journal of AI-Driven Informatics and Management Information Vol. 1 No. 1 (2026): April - August
Publisher : Journal of AI-Driven Informatics and Management Information

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

The rapid growth of digital business has transformed customer engagement, shifting from traditional marketing approaches toward highly personalized, data-driven strategies. Understanding consumer behavior is therefore essential for sustaining competitiveness. This study investigates the use of customer personality data to predict consumer responses in digital business campaigns by comparing four machine learning scenarios: Bald Eagle Search–Random Forest (BES-RF), Bald Eagle Search–KNN (BES-KNN), baseline Random Forest (RF), and baseline KNN. The dataset includes demographic attributes, purchasing behavior, and historical campaign acceptance, offering a comprehensive view of customer characteristics. Evaluation results show that BES-RF achieved the highest accuracy (88.1%) and AUC (0.86), with relatively strong precision (70.3%) but limited recall (27.4%), reflecting challenges in capturing minority classes. RF performed similarly (accuracy 87.6%, AUC 0.85), while BES-KNN improved precision (76.2%) but suffered from very low recall (16.8%). KNN recorded the weakest balance across metrics (accuracy 86.5%, recall 22.1%). Macro-averaged metrics highlight poor sensitivity to responders, weighted averages inflate performance due to majority dominance, and Matthews Correlation Coefficient values confirm only moderate predictive consistency. Confidence intervals for accuracy and AUC suggest that improvements over baselines are modest and not definitive. Consumer pattern analysis revealed that higher income, recent purchases, and product-specific expenditures particularly wine and meat products were influential predictors of responsiveness. However, the current evidence does not confirm effectiveness. The BES algorithm is insufficiently defined, baseline comparisons are limited, validation is unclear, and recall remains very low. Limitations include reliance on a single secondary dataset, class imbalance, possible leakage and outliers, cross-sectional design, lack of external validation, limited generalizability, and inability to infer causality. Future work should focus on methodological improvements such as resampling techniques, cost-sensitive learning, or hybrid ensembles to improve recall without sacrificing precision. Expanding datasets to include psychographic attributes, social media engagement, and real-time behavioral signals, alongside external validation and longitudinal designs, would enrich predictive accuracy and strengthen generalizability.

Page 1 of 1 | Total Record : 7

 1 

Filter by Year

2026 2026


Reset
Filter By Issues
All Issue Vol. 1 No. 1 (2026): April - August