Jurnal Informatika: Jurnal Pengembangan IT
Not a member yet
437 research outputs found
Sort by
Analisis Sentimen Aplikasi Get Contact di APP Store Menggunakan Metode SVM (Support Vector Machine)
No full textCurrent technological developments have led to a variety of new innovations to create applications that make it easier for users to manage phone calls, one of which is the Get Contact application. By managing phone calls, it is hoped that it will help users to minimize the occurrence of fraud or the like. The goal is to analyze user sentiment towards the Get Contact application by classifying user reviews into positive and negative categories through sentiment analysis. The Support Vector Machine method is used in this analysis process with a linear kernel approach to determine the accuracy of the Get Contact application review classification. The stages used in this research include data collection, preprocessing, labeling, split data, SVM model training, and model evaluation. This study shows that the Support Vector Machine (SVM) method classification of sentiment analysis of Get Contact application reviews on the App Store produces an accuracy value of 95.50%, negative precision 0.96, positive precision 0.95, negative recall 0.95, positive recall 0.96, positive and negative f-1 scores are the same, amounting to 0.95. As for the results of the most reviews are negative reviews with a negative review percentage of 94.8%, while for positive reviews it is 5.2%
Utilization of Mobile Applications for Digital Marketing of MSMEs: Challenges and Opportunities in Central Lampung
No full textThis study aims to identify and recommend the development of a digital marketing application for MSMEs in Central Lampung, with a focus on addressing operational challenges and limited market access. Using the Systems Development Life Cycle (SDLC) approach with the Waterfall method, this research examines the necessary features to improve operational efficiency, expand market reach, and support data-driven decision-making. The key features recommended include a digital product catalog, payment integration, customer notifications, order tracking, and an analytics dashboard. This application is expected to reduce reliance on costly traditional promotional methods, while enhancing product visibility and customer interaction. However, challenges such as low digital literacy and limited internet infrastructure in some areas remain barriers that must be addressed through training, infrastructure improvement, and incentive strategies. The study suggests further research to evaluate the long-term impact and the potential application of advanced technologies to support more personalized and effective marketing strategies for MSMEs
A Supervised Learning Model for Sentiment Analysis Based on Regional Dialects in Tourism-Related Issues
No full textIndonesia has an exceptionally rich diversity of regional languages, one of which is the Bekasi dialect, often used in social media communication. The uniqueness of this dialect presents specific challenges in extracting public opinion, especially in text-based sentiment analysis. This study aims to develop a sentiment analysis framework that incorporates regional dialects from social media data and evaluate the effectiveness of various supervised learning algorithms. Data were collected from the Facebook group “Explore Bekasi Tourism,” totaling 1,257 posts and comments, which were filtered down to 1,000 relevant instances. A manual validation process was conducted by linguistic experts to convert non-standard terms and regional dialects into standardized Indonesian, followed by translation into English for annotation purposes. The analysis method involved preprocessing steps (tokenizing, case folding, stemming), feature weighting using TF-IDF, and sentiment classification using four algorithms: Naive Bayes, K-Nearest Neighbor, Support Vector Machine, and Decision Tree. The evaluation results show that Naive Bayes achieved the best performance with an accuracy of 76%, followed by K-Nearest Neighbor (67.5%), SVM (65.5%), and Decision Tree (28%). These findings highlight the crucial role of expert judgment in processing dialect-based data to ensure accurate sentiment classification. The study recommends developing a broader annotated corpus of regional dialects and exploring deep learning methods in future research to enhance classification performance and generalizability
Evolutionary Fuzzy Rule-Based Classification System dalam Analisis Sentimen terhadap Danantara
No full textIn order to improve the Indonesian economy, the government launched Danantara, a Sovereign Wealth Fund that serves to manage data owned in investment instruments such as stocks, bonds or property where the profits from this are returned to the state into the APBN, besides that Danantara aims to stabilize the economy and drive national development. However, with the huge amount of money being managed and its vital purpose, the existence of danantara has been the subject of much debate. Some people support the existence of danantara because it can help raise the Indonesian economy, but others reject the existence of danantara because they are afraid of being a place for corruption if there is mismanagement of large funds and can disrupt the current Indonesian economy. For this reason, the research aims to analyze public sentiment using the Evolutionary Fuzzy Rule-Based Classification System which has an approach to fuzzy rules that can overcome the level of ambiguity in sentiment analysis. The stages carried out in this research start from data collection using the webscraping method on platform x, data cleaning, data pre-processing, data labeling, classification of Evolutionary Fuzzy Rule-Based Classification System and at the end of the evaluation stage. The results obtained in this study are the accuracy and recall rates of 69%, then precision 72% and f1-score 70%. This shows that the Evolutionary Fuzzy Rule-Based Classification System model is less suitable in analyzing and classifying public sentiment regarding the existence of danantara
Evaluasi Gemini Flash pada Ekstraksi Jadwal Skripsi Terstruktur dan Tidak Terstruktur
No full textThe administration of thesis seminar and defense scheduling is often hampered by unstructured PDF formats, which increases manual workload and the risk of errors. This study aims to evaluate and compare the performance of three Gemini Flash model variants, namely Gemini 2.0 Flash-Lite, Gemini 2.0 Flash, and Gemini 2.5 Flash Preview, in automating schedule information extraction using a zero-shot prompting approach. The dataset consists of 87 PDF files containing thesis seminar and defense schedules (588 entries) from the 2023/2024 academic year, alongside 200 question scenarios executed in two different context formats: raw extracted text (TXT) and structured JSON data. Performance evaluation metrics include Precision, Recall, F1-score, Exact-Match, and inference latency per request. Experimental results indicate that Gemini 2.5 Flash Preview achieves average F1-scores above 0.98 in both contexts with approximately 3.9 seconds latency. Conversely, smaller-capacity variants (Gemini 2.0 Flash and Flash-Lite) showed more significant performance gains using the JSON format compared to raw text, especially on complex question types such as multi-attribute filtering and list retrieval. Through error analysis, the primary challenge identified was tasks requiring numeric aggregation and determination of superlative values, accounting for approximately 78% of total extraction failures, particularly for lightweight models. A paired t-test indicated no statistically significant difference between the two context formats (average F1 difference = 0.0077; p=0.48). This study recommends the use of explicit numeric prompting or rule-based post-processing when employing lightweight models to significantly improve the accuracy of academic schedule information extraction
Perancangan dan Evaluasi Desain Antarmuka Pengguna pada E-Commerce Peralatan Medis Berbasis User Centered Design
No full textIn todays rapidly evolving digital era, e-commerce platforms play a key role in distributing information, marketing, ordering, and online product sales. However, many platforms still struggle with suboptimal user interface and user experience design, such as unclear product and company information, complex ordering and payment processes, and visually unappealing interfaces, such as layout, colour, images, and typography. This study aims to designing the UI/UX design of a healthcare e-commerce platform at PT. Wanbass Timur Persada is using the User Centered Design (UCD) approach. UCD emphasizes active user involvement throughout the design process to ensure that the final product aligns with user needs, preferences, and behaviours. The research involved observing business process flows, conducting competitor UI/UX analysis, creating user personas and journey maps, developing prototypes, and evaluating the design. The result is a website-based e-commerce prototype that enhances navigation, improves information search efficiency, and features a clean and readable interface. User testing and in-depth interviews indicate that the prototype improves business process clarity and efficiency, while also meeting user expectations regarding interaction control and perceived security
Rancang Bangun Aplikasi Dashboard Laporan Pengaduan Kendala Sistem Internal pada PT. Telkom Indonesia Witel Purwokerto berbasis Website menggunakan Metode Prototype
No full textThe development of Internet technology has increased the number of internet users and affected people's lives at large. As a telecommunications company, PT. Telkom Indonesia Tbk. (Telkom) provides Indihome services, has an increasing number of customers. To meet customer needs, Telkom gave responsibility to the Access Service Operation (ASO) unit at Witel Purwokerto to supervise and control Indihome services. However, the recording of complaints by the Access Service Operation (ASO) unit is still done manually, resulting in duplicate orders and recording imperfections. Based on the existing problems, the author designed a complaint report system that uses methods such as the System Development Life Cycle (SDLC) Prototype model to overcome these problems and uses black box and white box testing methods. The purpose of this study is to build a system that will facilitate Witel Purwokerto's Access Service Operation (ASO) unit in accommodating data on complaint reports made by whistleblowers. The system is built using PHP programming language with the help of Laravel framework, and as database management using MySQL. The test results of the system with the black box and white box methods showed 99.74% success for black box testing, and the white box test results showed 100% success. It can be concluded that every function on the dashboard website reports complaints of internal system constraints in this study runs well
E-Assessment untuk Menentukan Tipe Gaya Belajar Menggunakan Algoritma Rule-Based Reasoning
No full textSetiap manusia dilahirkan dengan perbedaan fisik, psikologis, genetik, dan eksternal yang signifikan yang mungkin mempengaruhi karakter individu. Perbedaan karakter ini terkadang diabaikan oleh individu, terutama selama proses pembelajaran, karena kemampuan siswa dalam menyerap pengetahuan pasti berbeda-beda. Akibatnya, diperlukan sistem untuk mengidentifikasi jenis gaya belajar siswa dan membantu guru dalam menghasilkan kualitas pengajaran yang mudah diterima oleh siswa. Penelitian ini menciptakan sistem e-assessment gaya belajar yang menggunakan algoritma penalaran berdasarkan aturan untuk menentukan gaya belajar siswa seperti visual, auditori, kinestetik, dan campuran. Teknik evaluasi melibatkan perhitungan setiap jawaban tergantung pada kategori gaya belajar, yang menghasilkan persentase untuk menunjukkan jenis pembelajaran yang berlaku. Sistem ini dibangun menggunakan model penelitian ADDIE, dengan VueJS di frontend dan PHP Laravel di backend. Pengujian dilakukan dengan menggunakan metode blackbox dan User Acceptance Testing (UAT). Hasil pengujian blackbox dengan alat Selenium IDE menunjukkan bahwa sistem bekerja sesuai dengan skenario yang ditentukan, dan UAT dengan 30 siswa memperoleh skor 86%, yang menunjukkan bahwa sistem ini dapat digunakan. Metode ini dirancang untuk membantu pengembangan cara belajar yang lebih tepat bagi siswa dan guru
Modifikasi Gain Ratio Pada Algoritma C4.5 dengan Nilai Koefisien Determinasi untuk Prediksi Kelulusan Mahasiswa (Studi Kasus: Universitas Islam Madura)
No full textC4.5 is a decision tree algorithm that can be used for making predictions. The stages start from forming a decision tree through splitting attributes, pruning and extracting rules or knowledge to then be used for prediction. However, one of the weaknesses of the C4.5 algorithm is the occurrence of overfitting and misclassification costs which result in low prediction performance. The development of the C4.5 algorithm has been carried out in terms of split attributes such as the imprecise info-gain ratio (Credal-C4.5) method using Imprecise Probability Theory, bossing gain ratio (C5.0) and average gain. This research applies the termination coefficient value (R2) as a method for modifying the gain ratio in selecting attributes as decision tree nodes which is then implemented to predict student graduation on time using a case study at the Universitas Islam Madura (UIM). Testing of the decision tree model rule for predicting student graduation on time at UIM shows that the performance values of accuracy, precision and recall are 70.49%, 77.14% and 72.97%. This performance is higher compared to the C4.5 algorithm without making modifications to the coefficient of determination, especially in accuracy and recall performance, while the precision is lower but the difference is below 1%. The difference in performance values was 11.48% (positive) for accuracy and 27.03% (positive) for recall. Meanwhile, precision performance has a difference of -0.13% (negative). The application of the Knowledge Model Rule for student graduation on time at SIMAT UIM shows very good results because it displays a prediction results page
Peningkatan Keberagaman Data untuk Klasifikasi Penyakit Diabetes Berbasis Stacking Ensemble Learning
No full textDiabetes cases are becoming more common in the late years. Diabetes attacks not only parents, but also children. The development of diabetes is not far from the lifestyle and diet that we live on a daily basis. Therefore, early detection of diabetes is essential because the earlier the disease is detected, the easier it is to treat. In the process of detecting disease based on factors, the cause can be predicted with data mining. The aim of this research is to increase data diversity so that it can be processed to the maximum in data mining. In the process of data upgrading, we used the imbalance learning method SMOTE-ENN combined with the method Stacking Ensemble Learning. In the search for a powerful stacking model, seven classification algorithms were involved in the experiments carried out on this study, namely: Random Forest, Decision Tree, Gradient Boosting, Naïve Bayes, Extreme Gradiant Boost, Logistic Regression, and k-Nearest Neighbor. Four algorithms were used to be classifiers level 0 (base model), namely kNN, Gradient Boosting, decision tree, and random forest, while Random Forest was used again to be classifier level 1. (meta model). With these combinations, the accuracy obtained is 97.3%. These are the highest results when compared to individual algorithms