Journal of Computer Networks, Architecture and High Performance Computing
Not a member yet
473 research outputs found
Sort by
Analysis of Public Sentiment Towards The TikTok Application Using The Naive Bayes Algorithm and Support Vector Machine
Get PDFIn the current digital era, social media applications such as TikTok have become an important aspect of people's lives. TikTok allows users to create and share short videos, making it a global phenomenon with millions of active users. However, this application has also been the subject of various responses and opinions from the public. This research aims to classify public sentiment towards the TikTok application based on comments on Playstore using the Naïve Bayes algorithm and Support Vector Machine (SVM). This research method involves collecting comment data from Playstore using scraping techniques, resulting in 5,000 review data. Data pre-processing stages include case folding, tokenization, normalization, stopword removal, stemming, and data labeling using a lexicon. The data that has been processed is then weighted using Term Frequency - Inverse Document Frequency (TF-IDF) before being classified using the Naïve Bayes and SVM algorithms. Algorithm performance evaluation is carried out using the Confusion Matrix to measure accuracy, precision and recall. The research results show that the SVM algorithm has higher accuracy (84%) compared to Naïve Bayes (79%). SVM also shows better precision and recall values in classifying positive and negative sentiment from user reviews. From the results of the tests that have been carried out, the SVM algorithm is more effective than Naïve Bayes in sentiment analysis of the TikTok application. This research provides insight into how public sentiment can be measured and analyzed, and underscores the importance of choosing the right algorithm for data sentiment analysis on social media platforms
Classification of Watermelon Ripeness Levels Using HSV Color Space Transformation and K-Nearest Neighbor Method
Get PDFWatermelons had high appeal due to their sweet taste, refreshing nature, and numerous benefits. However, consumers often faced difficulties in selecting suitable fruit because of the subtle differences between fully ripe and half-ripe watermelons. One important indicator of a watermelon’s ripeness was the yellowish pattern on its skin. In this study, the proposed use of digital image processing methods, specifically the HSV Color Space Transformation, was aimed at extracting watermelon images and employing the K-Nearest Neighbor (K-NN) method to classify them into two categories: "Ripe" and "Half-Ripe." HSV (Hue Saturation Value) was a color extraction method used to convert colors from the RGB model. The Hue component indicated the type of color, Saturation measured the purity of the color, and Value measured the brightness of the color on a scale from 0 to 100%. In this research, the K-Nearest Neighbor (K-NN) method was applied to classify watermelon images based on the extraction of skin color features. This method compared a new image (test data) with training images to determine classification based on the nearest distance with a parameter of k=3. The data used consisted of 120 images, with 92 images used as training data and 28 images as test data. Experimental results showed an accuracy of 89%, with 25 images correctly classified and 3 images misclassified
Industry Class Clustering of Software Expertise Competency at SMKN 2 Kraksaan Using Constrained K-Means Clustering Algorithm
Get PDFAddressing the gap between school education and industry needs is a recurring concern, as many graduates struggle to enter the workforce due to lacking practical skills. Industry Classes aim to bridge this gap by preparing students with relevant skills and knowledge aligned with real-world industry demands. This study focuses on the application of Constrained K-Means Clustering to categorize students in the software engineering competency classes at SMKN 2 Kraksaan. This algorithm modifies traditional K-Means by utilizing Linear Programming Algorithm (LPA), ensuring each cluster meets predefined subject requirements. The research involves analyzing academic proficiency test data (TKDA) from 96 X-grade students, evaluating their abilities in analogy, series, figural, mathematical, and recall skills. Using a 3-cluster approach, each with 32 to 60 student capacity constraints, the study aims to optimize student distribution for effective learning outcomes. Evaluation through silhouette method yielded a score of 0.3199, indicating satisfactory separation between clusters with overlap to address. Cluster analysis revealed Cluster 2 as the most proficient, showcasing strengths in recall and series attributes critical for software engineering. These findings suggest that Constrained K-Means Clustering is effective in classifying students, highlighting Cluster 2 as optimal for software engineering competencies at SMKN 2 Kraksaan. Future research should focus on enhancing data quality, expanding sample size, and refining algorithms for improved clustering accuracy and effectiveness
Implementation of Intrusion Detection System with Rule-Based Method on Website
Get PDFThe aim of this research is to implement an intrusion detection system using rule-based methods on websites. The approach in this research is the development of an intrusion detection system (IDS). research results after implementation, testing, and acceptance of test results, conclusions can be drawn. The detection system can be implemented well in website-based applications using a rule-based method
Sentiment Analysis of Starlink on Twitter Using Support Vector Machine Algorithm
Get PDFIndonesia faces unique challenges in the provision of internet services. Cable and fiber optic infrastructure is often difficult and expensive to implement in many areas. Based on data from the Asosiasi Penyelenggara Jasa Internet Indonesia (APJII), by 2024 internet users will reach 221.5 million. Starlink, Elon Musk's satellite-based internet service through his SpaceX company, offers an innovative solution to fill the void in Indonesia's telecommunications infrastructure. However, Starlink's presence has raised concerns among local service providers, particularly regarding potential market disruption and existing regulations. Starlink could also pose a potential threat to Indonesia's security and sovereignty. Starlink has been a hot topic on various social media platforms, including Twitter. Twitter is a very popular social media platform with millions of active users who often share their opinions in real-time. The number of public responses in assessing the presence of Starlink in Indonesia, became a reference for a sentiment analysis. The Support Vector Machine algorithm is used to classify opinions into positive and negative categories. Based on testing that has been done using the Cross Validation technique with a K-Fold value with a total of 1976 tweets data. The results show that 1112 tweets contain positive sentiment and 864 tweets contain negative sentiment. This shows that 56.3% of people agree with the presence of starlink in Indonesia. While from the use of the Support Vertor Machine algorithm, the Accruracy value is 76.22%, Precision is 77.48%, and Recall is 81.38%
Implementing Preference Selection Index for Optimal Employee Ranking in Organizational Decision-Making
Get PDFThe rapid development of information technology has affected various aspects of life, including in the world of work. This research aims to apply the Preference Selection Index (PSI) method in determining the best employees at Bina Karya Utama Company. The assessment is based on four main criteria: Attendance, Tardiness, Overtime, and Length of Service. Data is obtained through observation and interviews, then processed using the PSI method which involves the normalization process and the calculation of preference values. The results showed that employees with alternative code A8 obtained the highest score, followed by A5 and A9. The PSI method proved to be effective in helping companies make objective and fair decisions, as well as motivating employees to improve their performance. This research concludes that a PSI-based decision support system can improve transparency and fairness in employee evaluation at Bina Karya Utama Company
Product Layout Analysis Based on Consumer Purchasing Patterns Using Apriori Algorithm
Get PDFIn every self-service store, it is certain to have a sales transaction data, where the data will continue to grow every day. But in self-service stores the data is only a record of sales at the store. Whereas transaction data can be used as information on how consumer purchasing patterns when shopping at the store, but not all supermarkets know this. So this research aims to find information on these purchasing patterns, where to do this research using the apriori algorithm which is part of the association technique which is also part of data mining, where in its application it will calculate the support value, confindence value and will be tested using the lift ratio. And after the calculation is carried out, optimization will be carried out using the high utility itemset mining variable which will calculate the highest profit value on the product, so that based on the calculation, the final result is obtained with a support value of 85%, a confidence value of 86%, a lift ratio test of 1.01 and the high utility gets the highest result of Rp. 567,000
Use of QRCode and Digital Signature Using The DSA Method to Authenticate Student Academic Documents
Get PDFVerification of digital documents is no longer done conventionally but is also done digitally, such as signatures on documents. A signature is a means of authentic evidence as well as proving a person's identity, is a means of proving the authenticity or validity of an agreement or approval for something in a document issued by two or more parties so that it can be used as a solution to verify the integrity of valid data in a document. Application of signatures Digital is also commonly used at the North Sumatra Islamic University (UINSU) Medan, especially in student study documents. Even though UINSU Medan has implemented digital signatures, this application is still a signature image obtained through scanning or photos and manually inserted into a document so that other people can easily reuse the signature image and it can even be misused. To facilitate the digital signature process, the code generated from the signature formation process is entered into a QR code so that it is easy to use to carry out the electronic document authentication process. Researchers will use QR codes and digital signatures to authenticate student documents using the DSA method. This research will be implemented into an application model with PHP and MySQL programming. The student document authentication application model using the DSA method works well as evidenced by the QRCode which contains document token information which is different for each document so that it cannot be duplicated by non-owners of the document
Analysis of the Selection of the Best Household Ceramics Using the Complex Proportional Assessment (COPRAS) Method
Get PDFRapid advances in communication and information technology due to globalization have had a significant impact on a number of industries, including the industrial sector. The industry is taking great advantage of the capabilities of this technology to search, store, distribute and present information. The ceramic sector in Indonesia looks increasingly promising every year. One type of building material that functions to cover the floor and beautify its appearance is ceramic. When choosing ceramics, consumers become confused because of the availability of various brands (vendors) with different themes and quality. When deciding on product quality, a decision support system can be implemented to offer a structured evaluation that assists stakeholders in the business and consumers in assessing high-quality ceramic options. DSS The complex proportional assessment method, or COPRAS, is used in system design. In improving the accuracy and efficiency of decision making, the COPRAS approach can evaluate several options and estimate them based on their utility level when attribute values ??are expressed in intervals. Based on the findings of this research, the application of the COPRAS method in the decision-making process to determine the best household ceramics can be used in selecting the best household ceramics by collecting data on ceramic criteria and the alternative used is the type of ceramic. The weights obtained for each criterion are then normalized which are then used to determine the Ui for each alternative, so that based on the results of this research the best household ceramics are obtained, namely Redhorse type ceramics with a Ui value of 100%, Fortuna type with a Ui value of 99.27%. , Prato type with a Ui value of 98.82%, Crystal type with a Ui value of 98.71%, Mulia type with a Ui value of 88.50%, Vancouver type with a Ui value of 88.24%, Murano type with a Ui value of 84.97% and the Virginia type with a Ui value of 79.77%
Case Study: Gradient Boosting Machine vs Light GBM in Potential Landslide Detection
Get PDFAn increasing demand for precise forecasts concerning the likelihood of landslides served as the impetus for this investigation. Human life, infrastructure, and the environment are all profoundly affected by this natural occasion. Constructing models capable of discerning intricate patterns among diverse factors that impact the likelihood of landslide occurrences constitutes the primary obstacle in landslide detection. Predicting potential landslides requires algorithms that are both accurate and efficient in their processing of vast quantities of data encompassing a variety of geographical, environmental, and ecological characteristics. An evaluation of the efficacy of both Gradient Boosting Machine and Light Gradient Boosting Machine in identifying patterns associated with landslides is accomplished by comparing their performance on a large and complex dataset. In the realm of potential landslide detection, the primary aim of this research endeavor is to assess the predictive precision, computation duration, and generalizability of Gradient Boosting Machine and Light Gradient Boosting Machine. This research aims to enhance comprehension regarding the comparative benefits of these two approaches in surmounting the obstacles associated with risk assessment and modeling pertaining to potential landslides, with a specific emphasis on efficiency and precision. The research findings are anticipated to serve as a valuable reference in the identification of more efficient approaches to reduce the likelihood of landslide-induced natural catastrophes. The accuracy of the GBM experiment reached 82% and LGBM reached 81%