cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
Rikie Kartadie
Contact Email
ojs@akakom.ac.id
Phone
+6282135469911
Journal Mail Official
ojs@akakom.ac.id
Editorial Address
Universitas Teknologi Digital Indonesia (d.h STMIK AKAKOM) Jl. Raya Janti Jl. Majapahit No.143, Jaranan, Banguntapan, Kec. Banguntapan, Kabupaten Bantul, Daerah Istimewa Yogyakarta 55918
Location
Kab. bantul,
Daerah istimewa yogyakarta
INDONESIA
Journal of Intelligent Software Systems
Published by Universitas Teknologi Digital Indonesia
ISSN : -     EISSN : 29627702     DOI : https://doi.org/10.26798/jiss
Core Subject : Science,
Computer Science & IT
Journal of Intelligent Software Systems (JISS) is open access, peer-reviewed international journal that will consider any original scientific article that expands the field of Intelligent Software Systems. The journal publishes articles in all Intelligent Software Systems specialities of interest to Intelligent Software Systems, physicians, and researchers.
Arjuna Subject : Ilmu Komputer (semua kategori) - Kecerdasan Buatan
Articles 49 Documents
 1   2   3   4   5 
Designing a Digital Marketing Strategy Through Keyword Weighting on Ten Main Competitors of PT Oemah Solution Indonesia with the TF-IDF Approach Yani Aji Susilo; WIDYASTUTI ANDRIYANI
Journal of Intelligent Software Systems Vol 4, No 2 (2025): Desember 2025
Publisher : LPPM UTDI (d.h STMIK AKAKOM) Yogyakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26798/jiss.v4i2.2313

Abstract

Information on the PT Oemah Solution Indonesia website is currently not easily accessible to internet service users, because it still uses conventional methods in terms of product marketing, in addition to that, the optimization of the PT Oemah Solution Indonesia website has so far only focused on adding website content without involving keyword elements as part of the search optimization element with the implication that the search results are not optimal, the use of internet media and search engines to increase pagerank to become number one in the search index should be implemented in order to help in marketing and can increase company turnover. Analyzing keywords competitor PT Oemah Solution Indonesia website with TFIDF method as a form of digital marketing strategy of PT Oemah Solution Indonesia website. The highest TF-IDF value, URL https://bamai.uma.ac.id/.../software-membuat-aplikasi android/ gets the highest TF value at number 1 and IDF at number 109,531 and TF-IDF value 1 for android keyword, the second order is for URL https://www.slimfaq.com/hugaf/...mobile gets TF value 1 and IDF value 109 and TF-IDF with a value of 1 with website keyword. From the results of the research that has been done it can be recommended for the preparation of company web content, android and website keywords must be involved in the preparation of content material, including in the creation of articles, service descriptions, and other important elements. In addition, the use of keyword variations of more than 2 words can also increase visibility in online searches. Optimizing the results of digital marketing strategies, collaboration between the marketing team and the content development team is necessary. TF-IDF analysis must be translated into content that is engaging and meaningful to the target audience. Integration with SEO (Search Engine Optimization) techniques is necessary to ensure content is easily found by Google
AGORA-BIM: An Agentic, Retrieval-Augmented and Spatially-Aware Framework for Natural Language Querying of BIM Knowledge Graphs Wijang Widhiarso; Alfiarini Alfiarini; Dytha Ananda Widhiarso; Jamaludi Salim
Journal of Intelligent Software Systems Vol 5, No 1 (2026): July 2026
Publisher : LPPM UTDI (d.h STMIK AKAKOM) Yogyakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26798/jiss.v5i1.2810

Abstract

Building Information Modeling (BIM) consolidates heterogeneous building information into a single digital model, yet retrieving meaningful insights from Industry Foundation Classes (IFC) files remains difficult and demands specialised expertise. Recent work has shown that converting IFC into a Linked Building Data (LBD) knowledge graph (KG) and pairing it with Large Language Models (LLMs) enables natural language querying, but three limitations persist: (i) the full ontology cannot be supplied to the LLM because of prompt-token constraints, restricting holistic reasoning; (ii) single-pass SPARQL generation fails on complex queries, which is the dominant error mode; and (iii) implicit spatial concepts that are not explicitly modelled as KG entities, such as rooms, corridors and enclosed areas, are frequently answered incorrectly. We introduce AGORA-BIM, a framework that extends the KG-plus-LLM paradigm with three coordinated modules: a retrieval-augmented ontology-grounding module that injects only the semantically relevant schema fragments into the prompt; an agentic, self-correcting SPARQL-generation module built around an executor–validator–repair loop with verbal reflection; and a spatial-reasoning module that reconstructs implicit building entities from element geometry, adjacency and opening relationships. We describe the architecture, the prompting design, and an evaluation protocol on a real multi-storey office building in Barcelona with an extended question set, comparing AGORA-BIM against the single-pass baseline and across commercial and open-source LLM backbones. The illustrative results reported here indicate that grounding LLM reasoning in retrieved schema, bounded self-correction, and explicit spatial inference together substantially improve correctness on indirect and reasoning-intensive questions, while preserving interpretability. AGORA-BIM thus advances intuitive, scalable and cost-aware natural language access to BIM data.Keywords: Building Information Modeling (BIM), Large Language Models (LLMs), Retrieval-Augmented Generation (RAG), Knowledge Graphs, Agentic AI
Integration of Constraint-based Mining in Frequent Closed Itemset Mining using CEG&REP Approach Bambang Purnomosidi Dwi Putranto; Yuli Astuti; Muhammad Haries; Wiwi Widayani; Ali Impron; Rikie Kartadie
Journal of Intelligent Software Systems Vol 4, No 2 (2025): Desember 2025
Publisher : LPPM UTDI (d.h STMIK AKAKOM) Yogyakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26798/jiss.v4i2.2315

Abstract

Frequent Closed Itemset Mining is an important approach in discovering hidden patterns inlarge-scale data. The CEG&REP (Concurrent Edge Prevision and Rear Edge Pruning)algorithm has previously been proven to improve the efficiency of the pattern mining processthrough parallel edge projection mechanisms and selective pruning of sequence graphstructures. However, the search space exploration can still be very large when the datasetcontains many items, high sequence lengths, or complex pattern variations. This research is animprovement of CEG&REP through the integration of constraint-based mining, namely theapplication of various types of constraints that can direct the mining process only to relevantpatterns. Three main types of constraints are introduced: temporal constraints (time-basedconstraints), length constraints (pattern length constraints), and item constraints (itemexistence or attribute constraints). This integration allows the pruning process to occur earlier,reducing the exploration of irrelevant branches, and improving the quality of the resultingpatterns. This approach aims to make CEG&REP more adaptive, efficient, and suitable forvarious application domains such as user activity logs, IoT sensor data, retail transactions, andbioinformatics analysis.
Retrieval-Augmented Generation Architecture for Indonesian Academic Regulation Question Answering: A Microservices-Based Implementation Hendarman Lubis; Istiqoomatun Nisaa; Annas Rifai; Erlangga Erlangga
Journal of Intelligent Software Systems Vol 5, No 1 (2026): July 2026
Publisher : LPPM UTDI (d.h STMIK AKAKOM) Yogyakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26798/jiss.v5i1.2814

Abstract

Indonesian higher-education institutions operate under a dense and frequently updated body of academic regulations—national standards, institutional statutes, and study-program handbooks—that students and staff must consult accurately. General-purpose large language models answer such questions fluently but without provenance, and they hallucinate rules that do not exist. This paper proposes a retrieval-augmented generation (RAG) architecture for Indonesian academic-regulation question answering, deployed as a set of loosely coupled microservices: an API gateway, a RAG orchestrator, an embedding and retrieval subsystem backed by a FAISS vector index, a generation subsystem, and an offline ingestion pipeline. The retrieval layer combines lexical BM25 and vector similarity through reciprocal rank fusion, so that regulation passages are grounded and citable. We evaluate the retrieval core—the component that determines whether generated answers can be grounded—on an original corpus of 20 synthetic Indonesian academic-regulation passages and 30 labeled questions. Vector retrieval attains Recall@1 of 0.867, MRR@10 of 0.902, and nDCG@5 of 0.915, outperforming BM25 (0.800, 0.865, 0.898) on early precision, while BM25 is an order of magnitude faster and reaches perfect Recall@5. Retrieval latency for all configurations remains below one millisecond on a single node. The microservices decomposition lets the index be rebuilt when regulations change without redeploying the generation service. Results indicate that a hybrid retrieval core is a sound and inexpensive foundation for grounded, citable regulation question answering, and that the architecture is deployable on modest institutional infrastructure.
Comparison of Self-Organizing Maps and K-means Algorithms in Grouping Divorce Cases in Yogyakarta City Priyo Purnomo; Domy Kristomo; Widyastuti Andriyani; Bambang Purnomosidi Dwi Putranto
Journal of Intelligent Software Systems Vol 4, No 2 (2025): Desember 2025
Publisher : LPPM UTDI (d.h STMIK AKAKOM) Yogyakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26798/jiss.v4i2.2309

Abstract

The Religious Court of Yogyakarta serves as the primary judicial body that processes, examines, and renders decisions on civil cases involving adherents of Islam at the initial legal level. The court manages approximately 900 cases annually, with marital dissolution proceedings constituting the predominant category. Categorizing these cases according to marital age and divorce causation is crucial for illustrating the distinctive patterns of divorce within the jurisdiction, thereby informing governmental initiatives promoting family wellness education. For the purpose of data clustering, a comparative analysis was conducted between self-organizing maps (SOM) and K-means methodologies. The research utilized secondary data from 2021, comprising 45 entries of concluded divorce cases. The internal silhouette validation metric established two as the optimal cluster quantity for both analytical approaches. The distribution and attributes of clusters remained largely consistent across both models. Evaluation using the standard deviation ratio revealed superior clustering performance by the SOM method when applied to the divorce case dataset. The analytical results demonstrated that cluster 1 encompassed 11 neighborhoods characterized by elevated divorce rates, while cluster 2 contained 34 regions exhibiting lower to moderate divorce frequencies. The principal determinants shaping divorce case characteristics in Yogyakarta City were identified as "ongoing conflicts and arguments," "abandonment by one spouse," and "financial circumstances."
Classification of Cancer Based on RNA Data Using Elman Recurrent Neural Network Eka Alifia Kusnanti; Anisa Nur Azizah; Alven Safik Ritonga
Journal of Intelligent Software Systems Vol 5, No 1 (2026): July 2026
Publisher : LPPM UTDI (d.h STMIK AKAKOM) Yogyakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26798/jiss.v5i1.2754

Abstract

Cancer is one of the deadliest diseases whose number of sufferers continues to increaseevery year. The development of cancer cells can quickly spread to all parts of the bodythrough the bloodstream or from the lymphatic system so that it can cause death. Thiscan happen because there is a disorder that exists in the gene. The basic thing in geneticsis the monitoring of gene expression itself, namely by measuring from mRNA not fromprotein because the sequence of mRNA will hybridize with complementary DNA and RNA.The purpose of this study is to classify cancer based on RNA data using the ElmanRecurrent Neural Network method. In the recurrent network there are two inputs, namelythe actual input and the contextual input. The iteration process is much faster due tofeedback, so parameter updates and convergence are also faster. The data used are RNAdata with four classes, namely BRCA or breast adenocarcinoma (breast cancer), KIRC orkidney renal clear cell carcinoma (kidney cancer), UCEC or uterine corpus endometrialcarcinoma (uterine cancer), and LUAD or lung adenocarcinoma (lung cancer). The datawill be preprocessed using a minmax scaler then classified using ERNN with trials ofdata sharing, learning rate, and the number of hidden layers. The best combinationof parameters was obtained at 20 nodes hidden layer I, 50 nodes hidden layer II, andlearning rate 0.1. In this model, the accuracy reached 99.19 %, sensitivity of 99.03 %and specificity of 99.72 %. The time required for the model is 19 seconds.
LACM-Tree: Exact Similarity Search via Learning-Enhanced Clustered Metric Trees with Distance-Table Compression Techniques Ali Impron; Linda Sutriani; Adi Kusjani; Agung Wilis Nurcahyo; Fadhlih Girindra Putra; Muhammad Haries
Journal of Intelligent Software Systems Vol 5, No 1 (2026): July 2026
Publisher : LPPM UTDI (d.h STMIK AKAKOM) Yogyakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26798/jiss.v5i1.2816

Abstract

Unmanned aerial vehicle (UAV)-based agricultural and forestry monitoring generates large volumes of imagery, creating a need for efficient similarity search over the high-dimensional embeddings produced by deep learning models. Classical metric indexing methods such as the M-tree, Slim-tree, and the Clustered Metric Tree (CM-tree) provide exact search with support for dynamic operations, but their performance degrades in high dimensions because the quadratic size of the pairwise distance table erodes the effective node capacity. This paper identifies distance-table compression as the key enabler for extending the CM-tree to high-dimensional embeddings. The core contribution is the Compressed Distance Table (CDT): 8-bit quantization of the pairwise distance table combined with an error-margin pruning rule that provably preserves result completeness (Theorem 1). Under an iso-memory configuration in which the per-node table budget is held constant, CDT halves the total distance-table footprint and reduces node accesses by up to 53% at n = 20,000 (and 61% on skewed data) while maintaining recall of exactly 1.000, verified by an auditing protocol that found zero unsafe prunings across more than 24,000 quantized pruning decisions. Three lighter-weight extensions—cost-oriented pivot selection (LPS), density-adaptive splitting (DAS), and multi-pivot bound tightening (MBT)—are evaluated in a controlled component-wise ablation that serves as a diagnostic study of learning-augmented metric trees. The ablation shows that these components are not additive: DAS in particular degrades performance through split-induced fragmentation, and the mechanism of this negative interaction is analyzed in detail. All results are obtained from a complete open prototype with directly measured distance computations and node accesses, on datasets up to 20,000 objects and 256 dimensions. The findings position distance-table compression—rather than learned heuristics—as the most robust path toward exact, dynamic, memory-efficient metric indexing for embedding workloads.
CLASSIFICATION OF OIL PALM FRUIT CROSS-SECTIONS USING HSV FEATURE EXTRACTION AND GAUSSIAN NAÏVE BAYES Teguh Junian Kuswanto; WIDYASTUTI ANDRIYANI; Rikie Kartadie; Bambang Purnomosidi D.P; Danny Kriestanto
Journal of Intelligent Software Systems Vol 4, No 2 (2025): Desember 2025
Publisher : LPPM UTDI (d.h STMIK AKAKOM) Yogyakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26798/jiss.v4i2.2310

Abstract

Accurate identification of oil palm fruit varieties is essential for supporting breeding programs and optimizing seed quality in plantation operations. Manual approaches often lead to inconsistencies due to the high visual similarity among fruit types, particularly between dura and tenera. This study proposes an automatic classification model for oil palm fruit cross-sections using HSV-based color feature extraction combined with a Gaussian Naïve Bayes classifier. A dataset of 186 cross-sectional fruit images was used, consisting of 90 training samples and 96 testing samples representing the dura, pisifera, and tenera varieties. The methodology includes preprocessing, segmentation, HSV feature extraction, model training, and performance evaluation through a confusion matrix. Experimental results show that the proposed model achieves an accuracy of 85%, with misclassifications primarily occurring in the tenera class due to its close resemblance to the dura variety. Compared to Linear Discriminant Analysis (LDA), the proposed approach demonstrates faster computation time and competitive accuracy. These findings indicate that Gaussian Naïve Bayes, supported by HSV feature descriptors, provides an efficient solution for lightweight and cost-effective digital classification of oil palm fruit varieties
Sentiment Analysis of Comments on the School Zoning System in Karanganyar on X Using Naive Bayes Meilia Eka Trisna; Cucut Hariz Pratomo
Journal of Intelligent Software Systems Vol 5, No 1 (2026): July 2026
Publisher : LPPM UTDI (d.h STMIK AKAKOM) Yogyakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26798/jiss.v5i1.2741

Abstract

Social media provides a space for the public to express their views on various public policies, including the implementation of zoning regulations in PPDB. This study aims to evaluate public opinion on the PPDB zoning system in Karanganyar Regency by analyzing comments on platform X using a Naive Bayes classification method. The dataset utilized in this research was obtained through web scraping with Python in Google Colab, which focused on public comments regarding the application of the PPDB zoning policy. Before classification, the text data underwent preprocessing, including text cleaning, converting to lowercase, word separation, removing common words, and stemming to improve data quality and facilitate analysis. Then, word weighting was carried out using the Term Frequency–Inverse Document Frequency (TF-IDF) method, followed by sentiment classification through a Multinomial Naive Bayes classifier. This study also utilized the SMOTE method to handle the imbalance in the amount of sentiment data to make the data distribution more balanced. The test results showed that the model achieved an accuracy rate of 92.77%, with a precision value of 75.00%, a recall rate of 37.50%, and an F1-measure of 50.00%. Based on the analysis, the majority of public comments regarding the PPDB zoning system in Karanganyar tended to be negative. The results of this research are expected to serve as a consideration to the Karanganyar Regency Education and Culture Office in improving the application of the PPDB zoning policy to make it fairer, transparent, and responsive to community needs.

Page 5 of 5 | Total Record : 49

 1   2   3   4   5 

Filter by Year

2022 2026


Reset
Filter By Issues
All Issue Vol 5, No 1 (2026): July 2026 Vol 4, No 2 (2025): Desember 2025 Vol 4, No 1 (2025): Juli 2025 Vol 3, No 2 (2024): December 2024 Vol 3, No 1 (2024): July 2024 Vol 2, No 2 (2023): December 2023 Vol 2, No 1 (2023): July Vol 1, No 2 (2022): Desember Vol 1, No 1 (2022): July