Kinetik: Game Technology, Information System, Computer Network, Computing, Electronics, and Control
Not a member yet
424 research outputs found
Sort by
Various Implementation of Collaborative Filtering-Based Approach on Recommendation Systems using Similarity
Get PDFThe Recommendation System plays an increasingly important role in our daily lives. With the increasing amount of information on the internet, the recommendation system can also solve problems caused by increasing information quickly. Collaborative filtering is one method in the recommendation system that makes recommendations by analyzing correlations between users. Collaborative filtering accumulates customer item ratings, identifies customers with common ratings, and offers recommendations based on inter-customer comparisons. This study aims to build a system that can provide recommendations to users who want to order or choose fast food menus. This recommendation system provides recommendations based on item data calculations with customer review data using a collaborative filtering approach. The results of applying cosine similarity calculation to determine fast food menu recommendations obtained for the item-based recommendation is Pizza Frankfurter BBQ Large with a value of 1.0, item-based with genre recommendation is Calblend Float with value 1.0 and user-based recommendation is Pizza Black Pepper Beef / Chicken Large with mean score 2.5
XGBoost and Network Analysis for Prediction of Proteins Affecting Insulin based on Protein Protein Interactions
Get PDFProtein Interaction Analysis (PPI) can be used to identify proteins that have a supporting function on the main protein, especially in the synthesis process. Insulin is synthesized by proteins that have the same molecular function covering different but mutually supportive roles. To identify this function, the translation of Gene Ontology (GO) gives certain characteristics to each protein. This study purpose to predict proteins that interact with insulin using the centrality method as a feature extractor and extreme gradient boosting as a classification algorithm. Characteristics using the centralized method produces features as a central function of protein. Classification results are measured using measurements, precision, recall and ROC scores. Optimizing the model by finding the right parameters produces an accuracy of and a ROC score of . The prediction model produced by XGBoost has capabilities above the average of other machine learning methods
Comparison Analysis Of Social Influence Marketing For Mobile Payment Using Support Vector Machine
Get PDFThere are many digital-based financial services today, one of them is mobile payment service. Users can deposit money and make online transaction with their smartphone through mobile application. Five mobile payment service providers with the most users in Indonesia, according to Dailysocial are GOPAY, OVO, LinkAja, DANA, and PayTren. This study uses sentiment analysis to classify user’s opinion into positive and negative classes. The classification method used is Support Vector Machine. This study utilizes three metrics, namely Net Sentiment, Share of Voice, and Social Influence Marketing Score. Those metrics are useful for knowing reputation, reach, and influence of brands in social media. The findings in this study indicate that GOPAY, OVO, DANA, and PayTren have a positive dominant sentiment, while LinkAja has a negative dominant sentiment. The brand with the biggest influence and reaches in the mobile payment industry is GOPAY. While the highest reputation brand is PayTren. The implication of this research is to encourage mobile payment providers to be able to monitor their brand conditions among their competitors by utilizing social network analysis method
Data Pattern Of Computer Maintenance Management System With Eclat Algorithm
Get PDFDecision support system, basically used to help choosing some solution for stakeholde to take the best decision in manufacturer. In manufacturer company using Enterprise Resource System (ERP) that has Work Oder (WO) modul as request maintenance from user. But many of data from WO still didn’t use to help decision making and only as warehouse data about infrastructure maintenance from last time. Because that, author use that data to help technician to take decision making by using association rule as pattern processing. This is because WO has unique pattern that has problem (p), symptom (s), and root cause (r). Previous research (Sukmana, Rozi, 2017) was proved if association rule can use to help people to take decison making, it is just involved two variable, that is symptom (s), root cause (r) and using apriori algorithm as association rule. And focussing in this research is using that three variable and eclat algorithm as association rule methode. Result of this research has purpose to take the best pattern when using eclat algorithm
Transient Analysis And Optimization Of A Knuckle Joint
Get PDF Knuckle joint is used to provide movement between rods while transferring force along the pin axis. It has a range of applications such as in robotics, reciprocating engine valve, fulcrum, and suspension bridge. Various cases have been reported of failures in a Knuckle joint due to poor design and strenuous loading condition. For a guaranteed safety of the structure, analysis and optimization of a knuckle joint are required. A cheaper and qualitative production of the knuckle joint can be achieved in a short period through optimization. In this paper, Finite Element Method (FEM) using ANSYS workbench was used to carry out topology optimization, and transient analysis of a knuckle joint where its dynamic response is observed and its weight is reduced through optimization under certain design loading conditions. Weight reduction of 20%, 35%, and 50% using a structural steel material under a static loading of 1000N. The optimization process successfully identifies the mass that needs to be removed to minimize both weight and cost without compromising its reliability and durability. The structural design was carried out using SolidWorks software and then imported into the ANSYS workbench for analysis. By the results obtained, it is proved that ANSYS software can be employed by production companies to minimize material wastages and maximize profits while at the same time maintaining product quality and reliabilit
Educational Media Design for Learning Basic Programming in Branching Control Structure Material Using Problem-Posing Learning Model
Get PDFBasic programming is one subject that tends to be difficult for students to learn. Along with the development of technology, several researchers have provided solutions to solve this problem, by developing educational games, educational media, interactive learning media, and other auxiliary media. However, on average they have not used or adhered to the syntax of various existing learning models. This study focuses on designing educational media that uses the problem-posing learning model to study the material of branching control structures in basic programming learning which is recommended as a learning medium for vocational high school students. Educational media named TOLSYASUPI-EduMed. We use the highest type of research and development (R&D), the level 4 that we adopted to be adapted into a number of steps that are in line with the needs of this research area. Observation techniques are used as a form of generative research which is a type of user experience research, to explore information before designing a product/application. The side that we highlight here is how the form of educational media design by following the syntax of the problem-posing learning model. Then do an A/B testing which is assessed by experts to choose the best design with results that are type B designs with a percentage of 90.9%. We also state the analysis of the functional aspects of educational media to strengthen the validity of this design idea
Pattern Recognition Bird Sounds Based on Their Type Using Discreate Cosine Transform (DCT) and Gaussian Methods
Get PDFTo know the type of bird, most people know from the shape of bird species and the sound of birds. In this study, it identified the pattern of bird sounds. The bird sounds studied were Canary Trills, Vulture and Crow birds. In the introduction of the type of bird sound pattern in this study using the Discrete Cosine Transform (DCT) method and Gaussian value. The researcher conducted several steps to get the sound model of birds, among others, namely (1) bird sound input in the form of WAV file, (2) Hamming Windowing, (3) DFT / FFT, (4) Mel Bank Filter, (5) DCT, and (6) Value Gaussian. The output obtained is in the form of vector values and represented in graphical form. The results obtained in the study of pattern recognition of bird sound types get the results of observations in the same bird sound duration and frequency of the same, then the same pattern is obtained in the same bird as evidenced by calculating the closest distance value with Bray Curtis method. For the same duration of time and the length of the frequency that is not the same; it found that the pattern of bird sounds is not the same
Performance Improvement of Non Invasive Blood Glucose Measuring System With Near Infra Red Using Artificial Neural Networks
Get PDFMeasurement of body blood sugar levels is one of the important things to do to reduce the number of people with diabetes mellitus. Non-invasive measurement techniques become a blood sugar measurement technique that is more practical when compared to invasive techniques, but this technique has not shown too high levels of accuracy, specificity and sensitivity. For this reason, the non-invasive measurement model using NIR and ANN is proposed to improve the performance of non-invasive gauges. Non-invasive blood sugar measuring devices will be built using a nodemcu board with photodiaodes and NIR transmitters whose data is then processed using ANN models compared to invasive blood sugar data obtained from 40 data. 40 data obtained then used as raw data to build ANN models which 75% percent of it use as training data and 25% od it will be use as testing data to validate accuration of the model been built, the split of data doing randomly without any interference from programmer or model designer. All the data gathered are data collected from all volunteers which willingly to test their blood glucose using invasive glucose meter and non invasive glucose meter which been built. The invasive glucose meter used to gather raw data of blood glucose is SafeAccu-2 with 95% level of accuracy so the accuracy and error parameter calculated in this research are based on that 95% level accurcy of the invasive device