Jurnal Teknik Informatika dan Sistem Informasi

Hybrid System Recommendations Using the Switching Method

Muhammad Rizki, Rianto Rianto — Tue, 27 Aug 2024 00:00:00 +0000

Technological developments force businesses to provide the best service by making recommendation systems a solution to maintain consumer loyalty. Many studies have been carried out on recommendation systems to overcome Cold-Start or Serendipitous Problems. This study conducted Hybrid Collaborative Filtering and Content-Based filtering using the Switching method as a medium for selecting the correct data and attributes. Furthermore, the data is processed using the TF-IDF and KNN algorithms. This study conducted several tests using various K values and the training and testing data composition. The test results show that the highest accuracy produced by the model that has been developed is 83.62 percent for the switching method with the product category attribute as the variable label and 74.9 percent for the switching method with the rating attribute as the variable label. The training and testing data ratio used in this study is 70:30, with a K equals 3. The study's results also found a significant correlation between the K value and the accuracy value, where a high K value would also result in high accuracy.

Analysis of Aquvaprn.exe Malware for Operating System Investigation Using Memory Forensics Method

Hafish Naufal Aditya, Nur Widiyasono, Alam Rahmatulloh — Tue, 27 Aug 2024 00:00:00 +0000

In today's digital age, data has become a valuable asset. Various techniques are used to steal personal data that could potentially be misused by irresponsible parties. The object used in this study is AQUVAPRN.exe, which is a type of malware known as a Remote Access Trojan (RAT). When this malware runs, the creator of the malware can access personal data from the infected operating system without the user's knowledge. AQUVAPRN.exe works in the background when an application is executed, creating several processes such as modifying the registry, creating files, reading files, and making continuous internet connections to a specific IP address without the user's knowledge. The result obtained from the AQUVAPRN.exe malware is an IP address of 109.51.76.80, with the domain located in Lisbon, Portugal, and has an MD5 hash value of 55c2c12970cda52f58bfad7b8c7d37d5. It is also known that the AQUVAPRN.exe malware uses an anti-reverse engineering technique, specifically obfuscation, which obstructs or hinders the malware from being analyzed or reverse-engineered to determine the code used to create the malware. The PID of the AQUVAPRN.EXE process is 8332 with a virtual tool (Virtual Address) of 0x8e0f57042080.

Design Customer Relationship Management with K-Means For Web-Based Customer Rewards

Angela Odetio Barimbing, Rahmat Rian Hidayat — Tue, 27 Aug 2024 00:00:00 +0000

The continuous growth of the industry drives PT Porta Griya Indah, an industrial company, to retain their customers. Currently, the company still employs manual methods such as telephone, WhatsApp, and email to manage customer relationships, making it difficult for the marketing team to foster good relations. To overcome this challenge, the company needs to implement Customer Relationship Management (CRM) strategies. To increase sales and customer loyalty, the appropriate choice is to provide rewards to customers. Therefore, the aim of this research is to develop a simple web-based Customer Relationship Management application with K-Means Clustering algorithm, used to determine rewards for customers accurately and quickly. Data will be grouped into two clusters, namely eligible for rewards and not eligible for rewards, based on criteria and assessment weights. The research results indicate that this system helps the marketing team in communicating with customers, facilitating the conveyance of customer issues and needs to the directors, and facilitating the selection of the best customers eligible for rewards to enhance loyalty. The implementation of this system enhances the efficiency of PT Porta Griya Indah in retaining customers and increasing customer satisfaction amidst the continuously growing industry.

Development of Financial Data Mart for Nonprofit Organization

Susiana Sunjaya, Agata Filiana, Lussy Ernawati, Gloria Virginia — Tue, 27 Aug 2024 00:00:00 +0000

Finance is a critical component of organizational long-term viability. Non-profit organizations must pay close attention to their financial health. This is a form of accountability for funds entrusted to the organization by its stakeholders. Additionally, organizations must prepare for potential future financial risks. As a non-profit organization, SATUNAMA Foundation currently administers its financial data separately for each funding source. This complicates financial analysis and decision making, particularly when attempting to comprehend the organization’s financial health holistically. A financial data mart was built in order to assist data analysis as well as store historical data based on a predetermined set of analysis criteria. The Kimballs method and the starflake schema were used in the development of this data mart. The data mart was successfully implemented in accordance with the schema design and was able to fulfill the given analysis criteria. The data mart has fulfilled the four characteristics of data warehouse based on the data characteristic evaluation. From the operational perspective evaluation, data mart is proven faster than operational database thus making the data mart more efficient in providing information. Dashboard and pivot table are also successfully built to visualize the results.

Web-Based Design Order Management System with Design Thinking Method

Louis Godson Sinaga, Meliana Christianti Johan — Tue, 27 Aug 2024 00:00:00 +0000

Company in Bandung conducts online sales in various marketplace to reach all customer. Transactions that have been made by customer will enter each marketplace used by the company, but the sales admin must open each marketplace to find out the transactions that have been made and also have to open one marketplace to add product and add voucher and others. This is less effective and makes it difficult for sales admin from company in Bandung if there are many transactions from customer and many product additions in each marketplace. Therefore, Order Management System (OMS) is very functional for company in Bandung to the facilitate the work of sales admin. Order Management System will be made based on the needs of the sales admin or user by using web-based design thinking method, so that the sales admin can access through the website. The research conducted only focuses on designing a web-based design of the order management system, so that the results of the study will discuss the design of the order management system which contains features based on the needs of the sales admin or user.

Evaluation of Information Technology Services Using the Information Technology Infrastructure Library Framework

Arya Pratama, Wella Wella — Tue, 27 Aug 2024 00:00:00 +0000

This study aims to evaluate PT Gramedia Asri Media to find out whether the company's IT services meet the best standards of the ITIL V3 Service Operations framework. Service Operations focus includes Problem Management, Incident Management, Access Management, Request Fulfillment, and Event Management. Evaluation is carried out in four stages: planning, field inspection, reporting, and follow-up. The results indicate the maturity level of IT service implementation. For data collection, questionnaires were distributed to five people who were surveyed and interviewed with the heads of engineering services and support staff. The results show that the company's IT service maturity level is at level one. In addition, this research produced seventy suggestions for companies to improve their IT services. The results show that almost all suggestions were well received by the company, and some of them have been implemented at PT Gramedia Asri Media. However, of course, with the problems discovered from this research, there will be a desire to improve service and performance.

Human Centered Design-Based Information Architecture Implementation On Profile Websites

Febrianty Rakhmana, Dian Prawira, Nurul Mutiah — Tue, 27 Aug 2024 00:00:00 +0000

In the era of globalization, the use of the internet facilitates users in obtaining information. One of the media for obtaining information is by accessing websites. Website is a part of the internet that consists of pages displaying text, images, sound, and videos. One type of website is a profile website that presents specific information about organization or educational institution. However, not all educational institutions fully utilize websites, and one example is Information Systems Program (Prodi Sistem Informasi), as their existing website only showcases a landing page and does not meet the users' information needs. Therefore, this research conducted to provide a profile website implementing Information Architecture based on Human-Centered Design to effectively convey website information, aligning with user needs. The applied Information Architecture encompasses aspects such as organization system, labeling system, navigation system, and searching system. Human-Centered Design approach includes analysis of usage context, analysis of user needs, design solution creation, and design evaluation. Design evaluation uses tree testing, resulting in a success rate of 94.71%, directness 82.65%, time spent 3 minutes 59 seconds, first click 73.75%, and destination 80.5%. Thus, the design solution's outcomes can be implemented in the form of a website, as the majority of users can find all tested categories. Subsequently, searching system test conducted through tree testing, yielding success rate of 98%, directness 96%, time spent 79% falling into fast category, first click 97%, and destination 96%. It's concluded that searching system can assist users in quickly and accurately finding the desired pages.

Digital Course Application Development Using Multiplatform Framework and Clean Architecture

Erico Darmawan Handoyo, Sulaeman Santoso, Daniel Jahja Surjawan — Tue, 27 Aug 2024 00:00:00 +0000

The 2020 pandemic taught many people how the digital world can play a big role in life. One area that is affected is the world of learning. Various types of Learning Management Systems (LMS) and learning software were developed. However, many of these learning applications are not designed for one type of learning but to accommodate learning in general. This makes the LMS and software flexible for different types of learning but also makes the system cumbersome and difficult to navigate or manage. Therefore, LMS is generally not appropriate to accommodate learning in a course. A course differs from a formal institution of learning because it is simpler in nature and the needs of its users are also different. This research will design and develop learning software specifically made for digital courses. This software was developed using a multiplatform framework so that it can be used on various platforms.

Strategic Business Planning for Information Systems to Improve Public Health Services

Rita Ambarwati, Dwi Fitri Widyasari — Tue, 27 Aug 2024 00:00:00 +0000

The system management strategy plays a vital role in maintaining profitability to achieve revenue with a strategic plan that starts in all sectors of life to replace conventional technology-based services. This study aims to improve public health services through hospital environmental conditions so that performance increases and provides added value to increase the hospital's competitive advantage. There are several systems that need to be optimized, so it is necessary to group data with the ease of application systems. The study uses a descriptive descriptive approach with IS business strategy methods, internal and external environmental analysis, SWOT analysis, and McFarlan Strategic Grid analysis on IS applications. From the results of the research that has been done, IS planning is influential in helping management make decisions about a policy in a hospital that is useful for evaluating IS applications that continue to change for the better in the future.

Public Service Responsiveness based on E-Government with Digital Population Identity Innovation

Allgreat Meidelia Clarista Salamena, Andi Wahju Rahardjo Emanuel — Tue, 27 Aug 2024 00:00:00 +0000

This study aims to analyze the service responsiveness at the Population and Civil Registration Office of Ambon City, particularly regarding the implementation of Digital Population Identity (DPI). Public service responsiveness is considered crucial in meeting society's expectations towards technological advancements. The research method applied is qualitative, with data collection through direct observation and interviews at the Population and Civil Registration Office of Ambon City. The research results are expected to provide insights into the service responsiveness of this institution towards the needs and expectations of the community. Thus, this research can offer insights into how the office can enhance its responsiveness to technological changes and societal demands.

The k-anonymity using Simple Distribution of Sensitive Values and Aggregation of Sensitive Values

Widodo Widodo, Muhammad Ficky Duskarnaen, Murien Nugraheni — Tue, 27 Aug 2024 00:00:00 +0000

Anonymizing microdata is a matter while the microdata is published or shared. The purpose of anonymization is so
that sensitive data will not be known by unauthorized parties either directly or indirectly. The technique that is widely used is
generalization and suppression in the k-anonymity model, however, this technique has the disadvantage that the level of information loss is quite high. In addition, the generated microdata representation due to anonymization is too large, thus it needs to be simplified. In this research, an anonymity model is built using a sensitive attribute distribution technique, namely Simple Distribution of Sensitive Values (SDSV). The main purpose of this method is to reduce the probability of unauthorized parties guessing the owner of sensitive data. Meanwhile, to simplify the representation of the microdata, the aggregative of sensitive value (ASENVA) technique is applied. The result shows that the SDSV metho has less information loss compared to others, while the use of ASENVA simplifies the representation of anonymized tables to an average of 13.67% for aggregated quasi-identifier tables and 6.35% for sensitive tables.

Design of Tongue Pressure Training Monitoring Application based on Android & Web

Tue, 27 Aug 2024 00:00:00 +0000

The tongue is part of the muscular organs that are essential in the oral (mouth) and maxillofacial systems (jaw, neck, and face). By doing tongue movement exercises can help therapy patients with certain diseases, such as sleep apnea. This research will design and develop LTL applications based on Android and the Web using LTL tools that have been designed by Novie et al. Android-based applications are designed to help monitor the results of FSR sensor output from LTL devices from the user side, and Web-based applications are designed to monitor the progress of the LTL process that has been carried out from the side of medical personnel. The types of tongue exercise movements used in this research are Tongue Slide-1, Tongue Slide-2, and Tongue Press. The Android application will show patient identity data, tongue pressure reading data, and movement repetition data. The Android LTL application design that has been successfully realized presents features to monitor the results of tongue pressure exercise readings. The subjective test results of the appearance and ease of LTL android and LTL web applications get a good assessment because all assessments are in the index of 86.3%-92.5%. Meanwhile, the results of the experimental scenario for LTL System Testing show that the average tongue pressure of all respondents carried out in two experiments for tongue press movement, tongue slide-1 movement, and tongue slide-2 movement, above the set threshold limit of above 500 gf. Further research into experimental scenarios and the use of this application can be applied to real patients.

Designing a Web-Based Interest and Talent Test Application with the Scrum Approach

Jeffry Jeffry, Marcel Marcel — Tue, 27 Aug 2024 00:00:00 +0000

The selection of majors that align with students' interests and talents plays a crucial role in determining their academic success and future careers. However, the Faculty of Psychology faces significant challenges related to the efficiency of the interest-aptitude testing process and the quality of counseling services. Conventional approaches relying on manual forms and booklets have proven to be ineffective and time-consuming. Therefore, this research introduces an innovative solution in the form of a web-based interest test application designed using the agile scrum method. This application is designed to implement the RIASEC theory (Realistic, Investigative, Artistic, Social, Enterprising, and Conventional), which helps identify students' interests based on various types of interests and talents. The RIASEC theory offers a comprehensive framework for understanding students' interest tendencies, subsequently facilitating more accurate major selection. The agile scrum method was chosen to ensure the iterative development of the application and responsiveness to user feedback, thereby making it adaptable to the specific needs of the Faculty of Psychology. In initial testing, this application has shown the potential to improve the efficiency of the interest-aptitude testing process and the quality of counseling services. Features such as an intuitive user interface, automated data processing, and detailed test result reports are expected to reduce the time and resources required in the conventional process. Additionally, this application also allows easier access for students to take interest-aptitude tests anytime and anywhere, thus expanding the reach of counseling services. With this solution, it is hoped that the Faculty of Psychology can address existing constraints and provide more effective and efficient services in assisting students in selecting majors that align with their interests and talents. This application not only provides technological solutions but also contributes to improving the quality of education and career development for students in the future.

Analysis of Laptop Purchase History to Determine Key Factors Using Logistic Regression

Feliks Victor Parningotan Samosir, Hery Hery — Tue, 27 Aug 2024 00:00:00 +0000

E-Commerce is a new concept commonly described as the process of buying and selling goods or services over the World Wide Web or the exchange of products, services, and information through information networks, including the internet. Tokopedia, one of the largest e-commerce platforms in Indonesia, markets many products, including electronics. This study analyzes laptop data from Tokopedia. The vast history of purchase data and customer data in Tokopedia's database offers various ways to process it. One approach is using logistic regression models. The logistic regression results on the three features (price_cat, memory_size, and merk_cat) indicate that all three can be used to create logistic regression equations to determine laptop purchase probability. However, the regression calculations show that the r^2 value for price_cat is the highest at 0.82, indicating that price_cat significantly influences the probability of purchasing a laptop.

The Utilization of Time Series Data Forecasting Techniques on Hospitals Pharmaceutical Inventory

Aziz Mu'min, Setia Budi, Hapnes Toba — Tue, 27 Aug 2024 00:00:00 +0000

Good inventory management is essential in the hospital industry to overcome inventory problems. Ineffective inventory prediction methods can lead to shortages or excess stock inventory. Ultimately, this can impact the budget and availability of pharmaceutical items in the hospital. Previous traditional prediction methods often show inaccuracies. This research utilizes ARIMA (Autoregressive Integrated Moving Average) and FB Prophet methods to predict the demand for pharmaceutical items in hospitals. In an attempt to evaluate the effectiveness of both methods, an experiment was conducted on five pharmaceutical items. The results showed that the ARIMA method produced better performance compared to the FB Prophet method, with the smallest error of 0.07310.

Comparative Analysis Of Machine Learning Algorithms For Forecasting Of Basic Goods Products

Avinash Avinash, Andreas Widjaja, Oscar Karnalim — Tue, 27 Aug 2024 00:00:00 +0000

Abstract —In the era of continuously evolving technology, consumers' demands for everyday needs are becoming more complex.
Retail companies must adopt sophisticated approaches to understand and meet consumer preferences. This research explores the
effectiveness of Machine Learning algorithms in forecasting inventory levels in various types of retail stores using historical
transaction payment data and related variables. One approach used is data transformation using one standard deviation interval
to stationarize data, ensuring statistical consistency that is important for prediction algorithms. The research results show that the
Seasonal Autoregressive Integrated Moving Average (SARIMA) algorithm performs best in predicting inventory levels for both
SMEs and modern retailers. For the original data, the Mean Absolute Percentage Error (MAPE) for SMEs is 1.11% and for
modern retailers is 0.98%. For data modified with one standard deviation interval, the MAPE for SMEs is 0.74% and for modern
retailers is 0.70%. These results indicate superior prediction accuracy, helping companies adjust their inventory levels more
accurately according to market dynamics and consumer expectations. This research is expected to provide a solid guideline for
improving inventory management strategies, enabling companies to prepare inventory levels more accurately according to market
dynamics and consumer expectations.
Keywords—Forecasting Optimization, Machine Learning Algorithm Comparison, Inventory Levels, Modern Retail, SMEs, One
standard deviation interval

Evaluation of Neural Style Transfer Results of Various Pattern Images Using Feature Similarity Index

Vanessa Metayani, Swat Lie Liliawati, Andreas Widjaja — Tue, 27 Aug 2024 00:00:00 +0000

This research was conducted by applying the Neural Style Transfer method to various sets of content and style images, and then calculating the FSIM value for each pair of result and original images. Analysis was done on factors such as art style complexity, resolution and other special characteristics such as colour and texture that can affect the FSIM value. The purpose of this research is to identify whether there are factors that affect FSIM performance such as art style complexity, resolution, or other special characteristics such as colour and texture. This research is expected to be able to help artists who want to change the art style with Neural Style Transfer but still maintain the originality of the image and still be recognised by evaluating the results using FSIM and help artists to develop and produce artistic digital artworks with good quality. The results show that varying FSIM values can depend on the complexity of the art style and image resolution. Simple art styles and high-resolution images tend to produce higher FSIM values, indicating that the image structure is easily preserved. As long as the resolution and colours or textures do not change the main structure, the FSIM results will not decrease significantly. To support the research analysis, the Analysis of Variance (ANOVA) statistical test was used to measure the significance of the effect of complexity and resolution on FSIM and the Cronbach’s Alpha test to test the reliability of the general public and expert surveys. Based on the ANOVA statistical test results, there was insufficient evidence to reject the null hypothesis, so complexity and resolution did not have a significant influence on FSIM. From the Cronbach’s Alpha test results, the public assessment survey received a result of 0.94 and 0.91 for the expert assessment survey. These results indicate that the results from the surveys are reliable as subjective data for the research.