cover
Contact Name
Budi Hermawan
Contact Email
-
Phone
+62081703408296
Journal Mail Official
info@kdi.or.id
Editorial Address
Jl. Flamboyan 2 Blok B3 No. 26 Griya Sangiang Mas - Tangerang 15132
Location
Kab. tangerang,
Banten
INDONESIA
bit-Tech
ISSN : 2622271X     EISSN : 26222728     DOI : https://doi.org/10.32877/bt
Core Subject : Science,
The bit-Tech journal was developed with the aim of accommodating the scientific work of Lecturers and Students, both the results of scientific papers and research in the form of literature study results. It is hoped that this journal will increase the knowledge and exchange of scientific information, especially scientific papers and research that will be useful as a reference for the progress of the State together.
Articles 642 Documents
Hybrid Prediction Model Fuzzy Time Series-LSTM on Stock Price Data with Volatility Variation Alfi Hidayatur; Mohammad Idhom; Wahyu Syaifullah
bit-Tech Vol. 8 No. 2 (2025): bit-Tech
Publisher : Komunitas Dosen Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32877/bt.v8i2.3014

Abstract

Predicting stock prices in volatile markets remains a major challenge in financial analysis because irregular fluctuations often undermine the reliability of conventional models. Traditional methods such as ARIMA struggle to capture nonlinear dynamics and the complex dependencies that characterize financial time series. To address this gap, this study proposes a hybrid forecasting model that integrates Fuzzy Time Series (FTS) with Long Short-Term Memory (LSTM). The FTS component helps manage uncertainty and simplifies volatility patterns, while the LSTM network captures sequential dependencies across time. Together, these elements provide a more adaptive representation of stock price behavior under different volatility levels. The model was applied to datasets representing both high and low volatility in the Indonesian stock market. Performance was assessed using Mean Absolute Error (MAE) and Mean Absolute Percentage Error (MAPE). Results show that the hybrid model achieved high accuracy in low-volatility data with an MAE of 284.36 and a MAPE of 0.039%. In high-volatility conditions it also maintained robust performance with an MAE of 885.85 and a MAPE of 0.53%. These outcomes indicate that combining fuzzy logic with deep learning offers a promising approach for stock prediction under volatility variation. The integration not only enhances the reliability of forecasting but also provides a basis for future exploration of risk-aware applications in financial analysis.
Evaluation of E-Learning Website Quality of Universitas Islam Majapahit Using Webqual 4.0 and IPA Nur Diniyanti; Anita Wulansari; Rafika Rahmawati
bit-Tech Vol. 8 No. 2 (2025): bit-Tech
Publisher : Komunitas Dosen Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32877/bt.v8i2.3015

Abstract

The rapid development of digital technology in education requires higher education institutions to provide online learning systems that are effective, responsive, and user-oriented. The e-learning website of Universitas Islam Majapahit (UNIM) serves as the main platform supporting student learning activities, yet issues remain regarding response speed, information availability, and interface design. This study aims to evaluate the quality of UNIM’s e-learning website using the WebQual 4.0 model, modified by adding the User Interface Quality variable, and analyzed with the Importance Performance Analysis (IPA) method. Data were collected through a five-point Likert scale questionnaire involving 281 screened respondents. The findings indicate an average conformity level of 95%, yet several indicators reveal negative gap values, reflecting performance that has not met user expectations. Six indicators SIQ1, SIQ2, SIQ3, SIQ4, SIQ5, and UIQ2 were identified in Quadrant I, marking them as top priorities for improvement. The implications of this study highlight the need to enhance system responsiveness, service interaction quality, and interface usability to meet user expectations. Practically, the results provide guidance for improving the development of UNIM’s e-learning website, while theoretically, this study reinforces the applicability of the modified WebQual 4.0 as a comprehensive evaluation method for assessing the quality of online education services.
Evaluation of the POSKetanmu Website Using E-GovQual and Importance Performance Analysis (IPA) Nur Dinanita; Anita Wulansari; Rafika Rahmawati
bit-Tech Vol. 8 No. 2 (2025): bit-Tech
Publisher : Komunitas Dosen Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32877/bt.v8i2.3016

Abstract

The rapid advancement of information and communication technology has driven governments to provide digital public services that are efficient, transparent, and user-centered. One such initiative is the POSKetanmu website developed by the Mojokerto Regency Population and Civil Registration Office, aimed at facilitating online processing of population documents. Despite its purpose, the platform still faces several challenges, including difficulties in understanding service procedures, lengthy account verification, and errors in document submission. This study evaluates the quality of the POSKetanmu website using the E-GovQual model, which consists of six dimensions: Ease of Use, Reliability, Trust, Functionality of the Interaction Environment, Information Content and Display, and Citizen Support. To determine service priorities, the Importance–Performance Analysis (IPA) method was employed. Data were collected from 391 valid respondents using a five-point Likert scale. The results reveal an average conformity level of 96% and an overall gap score of –0.17, indicating that performance lags slightly behind user expectations. Indicators located in Quadrant I, particularly RB1–RB4, EU1, and FI2, were identified as priority areas requiring immediate attention. These findings suggest that improvements in system reliability, interface navigation, and interactive features are essential. The integration of E-GovQual and IPA not only provides practical insights for policy-making but also offers a valuable framework for advancing digital government evaluation.
Identification of Ginger Varieties Using Manhattan Distance on Image Pixel Vectors and Histograms Rauditha Putri Cahyani; Rudi Heriansyah; Gasim Gasim
bit-Tech Vol. 8 No. 2 (2025): bit-Tech
Publisher : Komunitas Dosen Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32877/bt.v8i2.3019

Abstract

The integration of digital image processing and pattern recognition has opened new opportunities for improving agricultural product classification. This study focuses on the identification of three economically important ginger varieties red ginger, elephant ginger, and Emprit ginger through an image-based classification system. Unlike conventional manual inspection, which is prone to subjectivity and error, the proposed method applies a distance-based similarity measure to enhance consistency and reliability. Central to this approach is the use of the Manhattan Distance metric, chosen for its computational efficiency and robustness in high-dimensional data spaces. Two types of image features were explored: global intensity histograms and pixel vector representations. Comparative evaluation demonstrates that histogram-based classification achieves an accuracy of 86.6%, substantially outperforming the pixel vector approach at 76.6%. Novelty this research lies in demonstrating that lightweight, interpretable techniques can deliver competitive accuracy while avoiding the data and computational demands of more complex machine learning or deep learning models. This makes the system particularly suitable for smallholder farmers, local cooperatives, and resource-limited agricultural environments. Moreover, the study highlights the potential of histogram-based representation as a practical solution to variability in lighting and texture, offering improved robustness over traditional visual inspection or pixel-level methods. By contributing a simple yet effective framework, this research advances the field of agricultural informatics and supports the development of low-cost, automated tools for crop identification. Beyond academic significance, the findings have practical implications for supply chain management, post-harvest quality control, and precision agriculture, fostering transparency and value optimization in ginger production and distribution
Classification of Feline Skin Diseases Based on Severity Using Type-2 Fuzzy Dhevi Puspitasari; Anggraini Puspita Sari; Firza Prima Aditiawan
bit-Tech Vol. 8 No. 2 (2025): bit-Tech
Publisher : Komunitas Dosen Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32877/bt.v8i2.3027

Abstract

Cats are among the most popular pets due to their friendly nature and relatively easy care, yet limited health attention often makes them vulnerable to skin diseases. Clinical examinations at veterinary clinics provide accurate results but require considerable time and cost. This study develops an early detection system for feline skin diseases, defined as a computational tool to help owners recognize symptoms at an early stage prior to advanced clinical diagnosis. The system integrates Dempster–Shafer Theory (DST) for disease classification and Fuzzy Type-2 for severity classification, where severity is categorized into mild, moderate, or severe based on symptom intensity. Fuzzy Type-2 was selected over type-1 fuzzy logic due to its superior ability to manage uncertainty and linguistic variability in veterinary assessments. The hybrid approach combines decision tree-based questioning with DST to identify the most probable disease, followed by Fuzzy Type-2 to evaluate severity. Validation was conducted using 100 medical records from the Easy Pet Care Animal Clinic in Tulungagung. For DST-based disease classification, evaluation with a confusion matrix on 100 cases achieved 83% accuracy, 93% precision, 86% recall, and an F1-score of 89%, demonstrating strong statistical performance. For severity prediction using fuzzy type-2, testing on 20 cases resulted in 85% correct classifications. These findings confirm that integrating DST with Fuzzy Type-2 provides an effective and statistically validated model for decision support in feline dermatology. The system offers a low-cost, fast, and reliable screening method that accelerates decision-making and minimizes delays in responding to potentially severe cases
K-Means Approach to Identifying High-Risk Stunting Areas in Indonesia Hairul Aji; Femi Dwi Astuti
bit-Tech Vol. 8 No. 2 (2025): bit-Tech
Publisher : Komunitas Dosen Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32877/bt.v8i2.3042

Abstract

Stunting is one of the chronic nutritional problems that remains a serious issue in Indonesia, particularly in Bulungan Regency, North Kalimantan. This condition not only affects children's physical growth but also their cognitive development, which has implications for their quality of life and future prospects. This study aims to classify regions based on stunting risk levels to provide a more targeted framework for local governments in setting intervention priorities. The method used is K-Means Clustering, an effective data mining algorithm for non-hierarchical data clustering. The data used are secondary data from the Bulungan District Bappeda in 2021, covering 81 locations with 29 stunting risk factor variables. The analysis process was conducted through data processing, centroid initialization, Euclidean distance calculation, and the formation of convergent clusters. The results of the study show the formation of two main clusters: a cluster with moderate vulnerability and a cluster with high vulnerability. The moderate cluster is in a transitional state with fluctuating risks, while the high cluster has low health and sanitation indicators, requiring special attention. These findings indicate that the K-Means method can provide data-driven insights to support stunting prevention policies. This study is expected to serve as a reference for local governments in developing more targeted intervention programs and contribute academically to the application of data mining methods in public health.
Implementation of PXP in Developing a Website for Computer Sales and Services Maulana Bryan Syahputra; Nur Cahyo Wibowo; Rizka Hadiwiyanti
bit-Tech Vol. 8 No. 2 (2025): bit-Tech
Publisher : Komunitas Dosen Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32877/bt.v8i2.3047

Abstract

Technology is crucial for enhancing competitiveness and service quality for Micro, Small, and Medium Enterprises (MSMEs) in Indonesia. However, many MSMEs, including CV. Tecomp’99, a computer sales and repair company, still rely on manual processes, resulting in inefficiencies and reduced customer satisfaction. This study designs and implements an integrated web-based sales and service information system for CV. Tecomp’99 using the Personal Extreme Programming (PXP) methodology, a lightweight adaptation of Extreme Programming for individual developers. PXP was applied iteratively through requirement gathering, planning, implementation, testing, refactoring, and retrospectives. The system was developed using Laravel, Tailwind CSS, and MySQL, and validated through 67 test scenarios. Development was completed in four iterations, with 19 user stories implemented and a total of 71 story points completed in 63 working days, demonstrating stable velocity and reliable estimation. The resulting system integrates product sales, service order management, payment processing, reporting, live chat, and service tracking, improving workflow efficiency by 20% and reducing order processing time by 30%. These improvements are backed by measurable outcomes from testing and user feedback. Findings indicate that PXP is an effective framework for small-scale projects in MSMEs, supporting iterative development under resource constraints. Future work should include integrating AI-driven analytics for decision-making, enhancing mobile platform access, and developing customer feedback loops for continuous system improvement.
Stock Price Forecasting of Meta Platforms Inc (META) Using ARIMA Method Esra Rombeallo; Thesya Atarezcha Pangruruk
bit-Tech Vol. 8 No. 2 (2025): bit-Tech
Publisher : Komunitas Dosen Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32877/bt.v8i2.3055

Abstract

This study examines the application of the Autoregressive Integrated Moving Average (ARIMA) method in forecasting the stock price of Meta Platforms Inc. (META) using historical time series data from 2021 to 2025. The main objective is to identify the best-fit ARIMA model and evaluate its predictive accuracy in capturing stock price dynamics. Data preprocessing was conducted through differencing to achieve stationarity, followed by model identification using autocorrelation and partial autocorrelation analysis. Among several candidate models, ARIMA (2, 2, 1) was selected as the most appropriate, supported by the lowest Akaike’s Information Criterion (AIC) value of 530.9496 and favorable diagnostic tests. Residual analysis indicated no significant autocorrelation and normally distributed errors, confirming the model’s adequacy. Further accuracy evaluation demonstrated that ARIMA (2, 2, 1) achieved the lowest Root Mean Squared Error (RMSE) of 32.49 and Mean Absolute Percentage Error (MAPE) of 7.07%, indicating reliable forecasting performance. These findings underscore the effectiveness of ARIMA in modeling linear stock price patterns and offer valuable insights for investors and analysts in short- to medium-term decision-making. The practical significance of forecasting META’s stock price lies in the company’s strategic position within the global technology ecosystem, where accurate predictions serve as a crucial foundation for risk management, portfolio development, and policy formulation in highly volatile markets. Nevertheless, limitations persist, as ARIMA is unable to account for nonlinear dynamics or external shocks.  Future studies are recommended to integrate hybrid models, such as ARIMA-LSTM, or macroeconomic indicators to enhance forecasting accuracy.
Decision Support System For Selecting Smart Indonesia Card Candidates Using Preference Selection Index Method M. Reza Fhalepi; Herri Setiawan; Nazori Suhandi
bit-Tech Vol. 8 No. 2 (2025): bit-Tech
Publisher : Komunitas Dosen Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32877/bt.v8i2.3068

Abstract

The Kartu Indonesia Pintar (KIP) program is a government initiative designed to ensure equal access to higher education for students from low-income families. However, the selection process remains challenging due to the large number of applicants, diverse evaluation criteria, and reliance on manual judgment, which can lead to inefficiency and bias. This study develops a decision support system (DSS) using the Preference Selection Index (PSI) method to improve transparency and objectivity in selecting KIP recipients at Universitas Indo Global Mandiri. Data were obtained through observation, structured interviews, documentation, and secondary records from the BKABK finance division. Five main criteria were used in the evaluation process: parents’ occupation, housing condition, number of siblings, academic achievements, and interview performance. The PSI method was implemented through data normalization, calculation of mean and deviation, automatic weight generation, and computation of each applicant’s final PSI score. A total of 270 valid applicants were processed, with most achieving scores between 0.80 and 0.90 (mean = 0.86; SD = 0.04), reflecting a high level of competition. The top five candidates scored between 0.8787 and 0.9179, led by Christopher Nathan Tanugraha and Kiagus Deru Cahyadi. These results demonstrate that PSI can reduce subjectivity in weight assignment, increase efficiency, and minimize human error, while ensuring fair scholarship distribution. More broadly, the proposed PSI-based DSS can be applied in other universities and scholarship programs, offering a scalable solution for equitable and data-driven decision-making in higher education.
Application of Preference Selection Index (PSI) Method in Selecting Outstanding Students Yusuf Akhlun Nazzar; Shinta Puspasari; Lastri Widya Astuti
bit-Tech Vol. 8 No. 2 (2025): bit-Tech
Publisher : Komunitas Dosen Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32877/bt.v8i2.3085

Abstract

This study aims to develop an objective and transparent system for selecting outstanding students at MTsN 2 Palembang by applying the Preference Selection Index (PSI) method. Conventional evaluation in schools often prioritizes academic performance while overlooking extracurricular involvement and religious achievements. To address this limitation, three main criteria academic achievement, non-academic activities, and tahfidz were integrated into a multi-criteria decision-making framework. The research involved several stages: determining criteria and weights (0.5 for academics, 0.3 for non-academics, and 0.2 for tahfidz), collecting and normalizing data, calculating preference values, and ranking students. Data from 189 students were processed, producing PSI scores ranging from 0.6703 to 0.8329, with an average of 0.7251 and narrow gaps among the top five students. The results showed that although grade IX students dominated the top ten rankings, a grade VIII student achieved the highest score of 0.8329, demonstrating that balanced performance across criteria can outweigh seniority. These findings highlight the effectiveness of PSI in ensuring fair and holistic rankings by recognizing diverse dimensions of student performance, unlike conventional teacher-based assessments. Furthermore, the study illustrates the broader applicability of PSI as a decision-support tool that can be adopted by other schools and educational institutions to align evaluation practices with institutional values and educational goals. This approach not only improves fairness and transparency in student selection but also encourages the design of balanced development programs that foster intellectual, social, and spiritual growth.