cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
Rizki Wahyudi
Contact Email
rizki.key@gmail.com
Phone
+6281329125484
Journal Mail Official
jcse@icsejournal.com
Editorial Address
Perum Pasir Indah Blok K. No. 22, Pasir Lor, Kec. Karanglewas, Kabupaten Banyumas, Jawa Tengah 53161, Indonesia
Location
Unknown,
Unknown
INDONESIA
Journal of Computer Science and Engineering (JCSE)
Published by Institute of Computer Science and Engineering
ISSN : -     EISSN : 27210251     DOI : https://doi.org/10.36596/jcse
Core Subject : Science,
Computer Science & IT
Computer Architecture, Processor design, operating systems, high-performance computing, parallel processing, computer networks, embedded systems, theory of computation, design and analysis of algorithms, data structures and database systems, theory of computation, design and analysis of algorithms, data structures and database systems, artificial intelligence, machine learning, data science, Information System
Arjuna Subject : Ilmu Komputer (semua kategori) - Ilmu Komputer (Lain-Lain)
Articles 5 Documents
 1 
Search results for , issue "Vol 5, No 2: August (2024)" : 5 Documents clear
Class-Oriented Text Vectorization for Text Classification: Case Study of Job Offer Classification Wabo Tatchum, Ghislain; Nzekon Nzeko'o, Armel Jacques; Sosso Makembe, Fritz; Youh Djam, Xaviera
Journal of Computer Science and Engineering (JCSE) Vol 5, No 2: August (2024)
Publisher : ICSE (Institute of Computer Sciences and Engineering)

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

Abstract

Advances in data science have made it possible to solve many real-life problems using automatic text classification applications. This is the case in e-recruitment, where job offers are classified and recommended to jobseekers. In natural language processing, text classification involves a vectorization step, whereby each document is represented as a vector of coordinates linked to a keyword. Those keywords are obtained by vectorizing the entire corpus, and are used to distinguish one document from another in the corpus. However, it is preferable for each keyword to distinguish one class from another. To obtain these types of keywords, the authors consider the class of documents in the vectorization process. They first create a class-oriented document for each class by merging all documents from the same class, and then apply a vectorization algorithm. Experiments are carried out using datasets from Minajobs, Nigham, and Monster with the classification models Decision Tree, Naive Bayes, Support Vector Machine, and a deep neural network self-attention transformer (TFM). The vectorization methods used on class-oriented documents are Doc2Vec and TF-IDF combined with our class-oriented vectorization strategies, including OC, ZIPF, and OWDC. To evaluate these experiments, we used the precision, MAP, and F1-Score metrics. According to the results, the TFM methods can improve accuracy by 29, 40, and 33% compared to previous work and the traditional way of classifying text documents. The NB methods can improve accuracy by 19, 22, and 20%, while the DT methods can improve accuracy by 34, 37, and 34%. The SVM methods can improve accuracy by 33, 34, and 34% in the Monster, Nigham, and Minajobs datasets. In addition, we validate our contribution by comparing ourselves with three other works in the literature using four datasets (RE'16, Wap, WebKB, and Kla) and obtain improvements in accuracy and F1-score up to 55%.
Enhancing Credit Card Fraud Detection with Modified Binary Bat Algorithm: A Comparative Study with SVM, RF, and DT Olasupo, Yinusa Ademola; Malgwi, Musa Yusuf; Hambali, Moshood Abiola
Journal of Computer Science and Engineering (JCSE) Vol 5, No 2: August (2024)
Publisher : ICSE (Institute of Computer Sciences and Engineering)

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

Abstract

Numerous studies have revealed the problem of irrelevant features, noise, and dimensionality in a dataset, which can inhibit how the classification algorithm performs. In machine learning, feature selection approaches are critical, particularly in the context of credit card fraud detection, where relevant feature selection is critical. We use techniques such as machine learning algorithms, data mining techniques, and data science to stop and detect credit card fraud. These algorithms often classify genuine and fraudulent transactions in credit card datasets. However, the challenge of high dimensionality and irrelevant features persists, hindering improvements in classifier algorithms. This study centered on detecting credit card fraud (CCF) using a Modified Binary Bat Algorithm (MBBA) for feature selection. The MBBA selects the most informative features to improve the classifier algorithm's performance. The classifier algorithms used in this research are Support Vector Machine (SVM), Random Forest (RF), and Decision Tree (DT). We conducted the experiment using the Python programming language, and the results indicate that RF achieves 99.945% accuracy, SVM 99.847%, and DT 99.909%. As a result, RF has the best accuracy. In summary, the optimal performance of a classification algorithm depends on the selection of relevant features for credit card fraud detection. The paper suggests improving the effectiveness of classifier algorithms for credit card fraud detection by employing the Modified Binary Bat algorithm, which outperforms the Genetic Algorithm (GA) in feature selection
Building Reliable Loan Approval Systems: Leveraging Feature Engineering and Machine Learning Shoaeinaeini, Maryam; Shoaeinaeini, Milad; Harrison, Brent; Jasemi, Milad
Journal of Computer Science and Engineering (JCSE) Vol 5, No 2: August (2024)
Publisher : ICSE (Institute of Computer Sciences and Engineering)

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

Abstract

Automating loan approval system is essential in today's banking system.  Even with the shift to online platforms, the traditional method is still cumbersome and needs a lot of customer-related data. This study proposes a robust solution to overcome these challenges. Despite previous studies, new financial indicators in feature engineering stage are introduces to extract more important client information, thereby improving prediction robustness and accuracy. To implement our integrated approach, an online dataset from a finance company, is utilized. The dataset is preprocessed by various data preparation techniques, including cleaning, transformation, and feature engineering. Subsequently, the preprocessed data undergoes a range of powerful machine learning techniques such as K-Nearest Neighbor, Decision Tree, Gaussian Naive Bayes, and Logistic Regression. Additionally, three robust ensemble methods including Random Forest, AdaBoost Classifier, and Gradient Boosting Classifier are employed for further improveness in performance.  The presented approach succeeded to acheive the highest accuracy by AdaBoost Classifier at 88%. A comparison with the original preprocessed model using ROC curve and feature importance analysis demonstrates the superior performance of our approach, with a larger area under the ROC curve and reduced false positive rate. Furthermore, the comparison findings show a stronger reliance of our model on financial features rather than personal customer features, highlighting its robust classification performance. These results indicate the potential strength of our model to replace the current loan approval system in real-world applications.
Accuracy Assessment of Monthly Rainfall Predictions using Seasonal ARIMA and Long Short-Term Memory (LSTM) Akbar, Ahmad Aldizar; Darmawan, Yahya; Wibowo, Arief; Rahmat, Hayatul Khairul
Journal of Computer Science and Engineering (JCSE) Vol 5, No 2: August (2024)
Publisher : ICSE (Institute of Computer Sciences and Engineering)

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

Abstract

Hydro meteorological disasters are common in Indonesia. Rainfall predictions can help mitigate the impact of these disasters. This research aims to compare the accuracy of monthly rainfall prediction models using Seasonal Autoregressive Integrated Moving Average (SARIMA) and Long Short-Term Memory (LSTM) methods. The input data consists of monthly rainfall records from four locations: Sampali, Kualanamu, Belawan, and Tuntungan, located around Medan, North Sumatra. The dataset spans from 2000 to 2020, with training data from 2000 to 2018 and test data from 2019 to 2020. The accuracy assessment reveals that Belawan has the largest RMSE values for both models, measuring 27.68 mm for LSTM and 28.36 mm for SARIMA. Belawan records the highest MAE values, with LSTM and SARIMA yielding 5.65 mm and 5.79 mm, respectively. SARIMA models effectively capture general trends and seasonality in linear time series data with clear patterns but struggle with extreme changes or sharp fluctuations due to their reliance on linear relationships. In contrast, LSTMs are effective at modeling complex, non-linear relationships, making them suitable for capturing general trends, seasonal patterns, and more complicated variations in the data. Understanding the characteristics of the data is crucial before applying SARIMA or LSTM models.
Comparison of Moora and Waspas Methods for Recommendations of Cayenne Pepper Seeds Wardani, Agus Tri; Hamdani, Hamdani; Agus, Fahrul
Journal of Computer Science and Engineering (JCSE) Vol 5, No 2: August (2024)
Publisher : ICSE (Institute of Computer Sciences and Engineering)

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

Abstract

Multiple Criteria Decision Making (MCDM) encompasses several methodologies, including MOORA and WASPAS. These strategies demonstrate unique approaches and produce varying results. The main aim of this work is to provide a comparative analysis of the MOORA and WASPAS procedures. To achieve this objective, we conduct a detailed analysis that specifically examines five parameters related to cayenne pepper seeds: prospective crop yields, optimal harvesting time, recommended conditions for highland cultivation, weight of 1000 seeds, and plant height. The study utilizes the sensitivity test approach in a comparative analysis framework to ascertain the superior method. The computations using both the MOORA and WASPAS methods determine that the Bisi Hp 35 (A3) alternative is the best choice. This alternative has a MOORA preference value of 0.1463, while the WASPAS approach gives it a preference value of 0.8374. Next, we perform a sensitivity test by increasing the weight criteria for each criterion by 0.5 and 1. The sensitivity analysis indicates that the MOORA approach has a level of 380, whereas the WASPAS method has a level of 376. The data suggest that the MOORA method is more effective than the WASPAS method when it comes to making recommendations for cayenne pepper seeds.

Page 1 of 1 | Total Record : 5

 1 

Filter by Year

2024 2024


Reset
Filter By Issues
All Issue Vol 6, No 2: August (2025) Vol 6, No 1: February (2025) Vol 5, No 2: August (2024) Vol 5, No 1: February (2024) Vol 4, No 2: August (2023) Vol 4, No 1: February (2023) Vol 3, No 2: August (2022) Vol 3, No 1: February (2022) Vol 2, No 2: August (2021) Vol 2, No 1: February (2021) Vol 1, No 2: August (2020) Vol 1, No 1: February (2020) More Issue