JUTI: Jurnal Ilmiah Teknologi Informasi
Not a member yet
407 research outputs found
Sort by
KEYWORD IDENTIFICATION IN SCIENTIFIC JOURNAL PUBLICATION CONTENT FOR CASE STUDY ITS ONLINE PUBLICATION (POMITS) SEARCHING
Full text linkITS Online Publication (POMITS) is a publication journal for ITS undergraduate students. Many articles are published in it, and they are often needed as reference material for other student research. The search process is still based on title, abstract, author\u27s name, and keywords. The data is still entered manually by the author. This process allows the selection of less appropriate keywords. So an effort is needed so that the choice of these keywords can be more precise and represent the article.The purpose of this research is to identify keywords in articles automatically. These keywords are distinguished into the software used, methods, and other representative keywords. With this identification, article searches can return more precise search results. This problem can be solved by using Named Entity Recognition (NER). However, the Indonesian language NER model owned by SpaCy is still not available, so it is necessary to develop the NER model.This study identifies each keyword annotation in POMITS content into metadata by detecting named entities in the form of software, methods, and representative keywords using the NER model. The NER annotation results are stored as triplet pairs in the Apache Jena Fuseki triple store. Furthermore, the triple store can answer searches about software, methods, and keywords. Based on the test results, the system successfully detects NER entities and saves annotations as triplet pairs on Apache Jena Fuseki. Keywords identification produce an average value of 84.76% precision and 63.59% recall.
MALICIOUS TRAFFIC DETECTION IN DNS INFRASTRUCTURE USING DECISION TREE ALGORITHM
Full text linkDomain Name System (DNS) is an essential component in internet infrastructure to direct domains to IP addresses or conversely. Despite its important role in delivering internet services, attackers often use DNS as a bridge to breach a system. A DNS traffic analysis system is needed for early detection of attacks. However, the available security tools still have many shortcomings, for example broken authentication, sensitive data exposure, injection, etc. This research uses DNS analysis to develop anomaly-based techniques to detect malicious traffic on the DNS infrastructure. To do this, We look for network features that characterize DNS traffic. Features obtained will then be processed using the Decision Tree algorithm to classifyincoming DNS traffic. We experimented with 2.291.024 data traffic data matches the characteristics of BotNet and normal traffic. By dividing the data into 80% training and 20% testing data, our experimental results showed high detection aacuracy (96.36%) indicating the robustness of our method
THE DEVELOPMENT OF QR CODE BASED MOBILE ATTENDANCE INFORMATION SYSTEM USING SCRUM FRAMEWORK
Full text linkAs one of State ‘s Higher Education Institutions, the Kalimantan Institute of Technology (ITK) must perform the education and teaching function as mandated by the tri dharma perguruan tinggi, then the function is regulated in academic regulations and implemented in business processes of attendance. Currently, the attendances data are recapitulated manually at week 15 by Academic Staff. The attendance process that has been running has several problems, namely prone to violations of the actual meeting realization and attendance count, recapitulation time that takes a long time, risk of data input errors and loss of presence sheet. Based on these problems, the attendance information system is developed (SIAP ITK). This research was conducted with the agile software development methodology with the scrum framework. The results of this research is an android application following the business processes of attendance in ITK. Based on the testing result which was carried out during this research, SIAP ITK is considered capable of optimizing the attendance process that has been running at ITK
EVALUATION OF THE SUCCESSFUL APPLICATION OF MOVEAPS AT PT. PIXEL RESEARCH
Full text linkEvaluation of success in IS is an important aspect that must be done to develop information systems. Over time, the paradigm of evaluating the success of information systems continues to change according to the objectives, context, and impact of information technology. The information system evaluation models that can be used include the DeLone and McLean Information Systems Success Model (DM IS Model), Technology Acceptance Model, Unified Theory of Acceptance and Use of Technology, and others. Each model has a different purpose, so the evaluator must choose the model that suits his needs. Moveaps application is an application developed by PT. Pixel Research. The Moveaps application is used to support every research project and online research needs. The evaluation of these systems adopts the DM IS Model. The model adopted in this study used all the variables in the DM IS Model and added intrinsic motivation. Thus, the variables become 8 variables consisting of 1) Information quality, 2) System quality, 3) Service quality, 4) Intrinsic motivation, 5) Perceived interaction, 6) Usage, 7) User satisfaction, and 8) Net impact. Previously identified variables, then a model of the relationship between variables was developed. The relationship between variables produces 16 hypotheses. This hypothesis is then compiled into several questions that are arranged in the form of a questionnaire. Questionnaires were distributed to 41 Moveaps Users. The results of 16 hypotheses found five hypotheses that have a positive effect, namely: information quality has a positive effect on service quality, system quality has a positive effect on information quality, system quality has a positive effect on service quality, user satisfaction has a positive effect on net benefits, and usage has a positive effect on benefits. Clean. While 11 hypotheses have no effect
ENERGY EFFICIENT SLEEP WAKEUP SCHEDULING METHOD FOR P-COVERAGE AND Q-CONNECTIVITY MODEL IN TARGET BASED WIRELESS SENSOR NETWORKS
Full text linkEnergy limitations are the problem that gets the most attention in the term of Wireless Sensor Networks (WSN). Sleep wakeup scheduling method is one of the most efficient techniques to increase sensor node operational time on WSN. However, in the target-based WSN environment with p-coverage and q-connectivity models, the use of wake-up scheduling has to consider the constraints on the number of connectivity on the sensor and coverage on the target. Genetic Algorithm is a solution to the problem of sleep-wake scheduling with multi-objective problems. This study proposes a new method of sleep wakeup scheduling based on Genetic Algorithm for energy efficiency in target-based WSN with p-coverage and q-connectivity models. This new method uses the sensor range, connectivity range and energy as an objective function of the fitness function in the Genetic Algorithm. With the presence of energy as a factor of the objective function can increase energy efficiency in target-based WSN with p-coverage and q-connectivity models
IMPROVING ROBUSTNESS OF FACE EXPRESSION RECOGNITION USING MULTI-CHANNEL LOCAL BINARY PATTERN AND NEURAL NETWORK
Full text linkABSTRACTFacial Expression Recognition (FER) is a subset of Artificial Intelligence (AI) that relates to human non-verbal communication. The development of Convolutional Neural Network (CNN) based FER is subject to noise, mainly because of the usage of RGB Original Image as training data. Many research explored texture feature methods which noise resistant, such as Local Binary Pattern (LBP) and Gray Level Co-occurrence Matrix (GLCM), which mainly worked on grayscale images. Multi-Channel Local Binary Pattern (MCLBP) is derived from LBP which analyzes texture on color images.This research aims to develop FER using MCLBP as a method of hand-crafted texture feature and NN as a classification method. The combination of MCLBP and Neural Network (NN) is expected more robust to noise. First, preprocessing is applied to the facial image for contrasting with Adaptive Gamma Correction Weighted Distribution (AGCWD). Next, the facial image is converted to MCLBP images. Then MCLBP images are converted to vectors as a NN architecture training data with 5 Fully Connected layers. Batch Normalization and Rectified Linear Unit (ReLu) activation are used in every Fully Connected layer. At the last Fully Connected Layer, ReLu activation was replaced with SoftMax activation. This NN uses Stochastic Gradient Descend (SGD) optimizer with a learning rate of 0.005.Performance testing was held by comparing the epoch required to reach F1-score 1 and F1-Score from many scenarios in FER with LBP + NN with 140 × 190 image size, LBP + NN with 70 × 85 image size, and MCLBP + NN with 70 × 85 image size approaches. From all scenarios we have tried, the best method is MCLBP with F1-Score =1 in 22 epochs. The method of hand-crafted texture feature with NN can increase the desirable FER performances. Keywords: Local Binary Pattern, Multi-Channel LBP, Neural Network, Face Expression Recognition, Gamma Correctio
KUBERNETES CLUSTER MANAGEMENT FOR CLOUD COMPUTING PLATFORM: A SYSTEMATIC LITERATURE REVIEW
Full text linkKubernetes is designed to automate the deployment, scaling, and operation of containerized applications. With the scalability feature of Kubernetes technology, container automation processes can be implemented according to the number of concurrent users accessing them. Therefore, this research focuses on how Kubernetes as cluster management is implemented on several cloud computing platforms. Standard literature review method employing a manual search for several journals and conference proceedings. From 15 relevant studies, 5 addressed Kubernetes performance and scalability. Seven literature review addressed Kubernetes deployments. Two articles addressed Kubernetes comparison and the rest is addressed Kubernetes in IoT. Regarding the cloud computing cluster management challenges that must be overcome using Kubernetes: it is necessary to ensure that all configuration and management required for Docker containers are successfully set up on on-premises systems before deploying to the cloud or on-premises. Data from Kubernetes deployments can be leveraged to support capacity planning and design Kubernetes-based elastic applications
FRECOMTWEET: PRODUCT RECOMMENDATION APPLICATION USING FRIENDSHIP CLOSENESS ON TWITTER
Full text linkThe information and communication technology development makes someone interact with each other easier. This convenience is used to exchange ideas, like using social media Twitter for product recommendations before buying it. It brings up a trend that consumers seek product recommendations through other people on social media. Social media, especially Twitter, has several features such as tweets, ReTweet and mentions to interact with other people. Users can describe the product, attach a link, and give a positive or negative rating in a tweet. These types of tweets can be used as an alternative to product recommendations. FrecomTweet is an Android-based product recommendation application that can detect close friendships based on the user’s ReTweet and mentions. This application also detects a product recommendation that appears in a conversation between users. This detection uses the keyword filtering method, which matches the conversation content with the markers in the database. If the conversation has a positive rating, it will recommend the user’s closest friends. This research uses a crawling method with the Twitter API streaming filter built using the CodeIgniter framework. The results of the black box test show that Twitter user conversations can be used as a product recommendation with a precision and recall value of 0.94 and 0.81, respectively
OFFLOADING DECISION SELECTION METHOD FOR ENERGY EFFICIENCY AND LOW LATENCY IN HETEROGENE SIMUATION ENVIRONMENTS
Full text linkMobile Cloud Computing (MCC) is a technology that can overcome the problems of high computing and limited resources owned by mobile devices. However, in practice, MCC has a very long transmission distance from the mobile device, resulting in a large latency. Mobile Edge Computing (MEC) is a technology that exists to overcome this problem. However, new problems arise from the presence of this MEC.One of the problems that arise is the selection of offloading decisions from mobile devices. Several studies consider energy efficiency / large latency or both in determining offloading decisions. However, there are not many studies that consider the movement of mobile devices in determining offloading decisions. Even though the movement of mobile devices is also very influential on latency because tasks need to be migrated to another edge server when a mobile device has moved. Several studies that have addressed this issue apply the solution to smaller, less heterogeneous simulation environments.This study used a new method of offloading decision-making that pays attention to the movement of mobile devices in a heterogeneous environment. This proposed method uses Black Widow Optimization in solving the problem of decision selection when offloading. From the simulation results, the performance of the proposed method is better than the comparison method in terms of the amount of energy consumption and delay latency.
K-MEANS AND XGBOOST FOR CUSTOMER ELECTRICITY ACCOUNT PAYMENT BEHAVIOR ANALYSIS (CASE STUDY: PLN ULP PANAKKUKANG)
No full textRevenue Acceleration from electricity account receivables is one of the energy companies\u27 efforts to maintain cash flow so that they can carry out operational activities and carry out investment activities to develop company assets. Factors that influence electricity bill payment behavior include the location of consumers, the amount of the bill, payment point facilities located around consumers\u27 homes, the use of digital technology as a media of payment, as well as consumer awareness and understanding regarding the time limit for paying electricity bills. Therefore, it is necessary to conduct an analysis so that the company can determine a special strategy for customers who have the potential to be in arrears in electricity bills. To get the characteristic of electricity bill payments, several previous studies have used various classification methods of machine learning such as random forest, nave bayes, SVM, CART, etc. to get the best accuracy. In this research, to increase the accuracy of the model, author using the cluster method with the k-means technique and combining it with the eXtreme Gradient Boosting (XGBOOST) classification method based on data on the characteristics of consumer electricity bill payments. In this study also used hyperparameter adjustment with hillclimbing, random search, and bayesian techniques to increase the accuracy of the model. The model simulation carried out in this thesis gives the result that the combination of the k-means cluster with the XGBoost classification and by adjusting the bayesian technique hyperparameters has a much better model accuracy rate with a value of 89.27% and an Area Under Curve (AUC) value of 0.92 when compared to gradient boosting method with an accuracy rate of only 74.76% and an AUC value of 0.75. Based on the simulation results on ULP Panakkukang customer data, it was found that the subsidy category customer group and customers who often experience power outages have a tendency to be in arrears on electricity bills