Publication Search

Organisasi Kesehatan Dunia (WHO) secara resmi menyebut virus Covid-19 sebagai pandemi global, oleh karena itu semua negara di dunia berusaha meminimalkan dampak yang ditimbulkan oleh virus tersebut, yaitu dengan mengembangkan vaksin. Dalam konteks pandemi Covid-19, pemerintah Indonesia juga meminta dan mendorong masyarakat untuk turut serta mendukung vaksinasi, namun upaya tersebut sebenarnya memiliki kelebihan dan kekurangan, sehingga banyak masyarakat yang mengutarakan pendapatnya di jejaring sosial salah satunya Twitter. Penelitian ini bertujuan untuk mengetahui hasil penerapan analisis sentimen dan mengukur performansi algoritma Naïve Bayes Classifier (NBC) dan Support Vector Machine (SVM) terhadap data vaksin Covid-19 dengan cara mengklasifikasikan data tersebut ke dalam kelas positif dan negatif. Data tweet yang didapat kemudian dilakukan text preprocessing untuk mengoptimalkan pengolahan data. Terdapat 4 tahapan text preprocessing antara lain Case Folding, Tokenizing, Filtering, dan Stemming. Penelitian ini mengkaji kinerja Naïve Bayes Classifier (NBC) dan Support Vector Machine (SVM) dengan menambahkan teknik TF-IDF (Term Frequency-Inverse Document Frequency) yang bertujuan untuk memberikan bobot pada hubungan kata (term) sebuah dokumen. Kemudian melakukan splitting data yaitu membagi data training 80% dan data testing 20% dengan harapan mendapatkan model dengan performansi terbaik dan yang terakhir melakukan visualisasi data tweet dengan menggunakan Word Cloud agar bisa menarik sebuah kesimpulan. Hasil klasifikasi data tweet vaksin Covid-19 menggunakan algoritma Naïve Bayes Classifier mendapatkan nilai accuracy sebesar 81%, precision sebesar 80%, recall sebesar 99%, dan f1-score sebesar 89%, Sedangkan untuk algoritma Support Vector Machine mendapatkan nilai accuracy sebesar 87%, precision sebesar 88%, recall sebesar 96%, dan f1-score sebesar 92%.

Sentiment analysis of comments from visitors to tourist attractions and the public on tourist attractions in Sukabumi Regency which is one of the areas with various categories of tourist objects and is a sector of economic income for the surrounding community or for related parties such as the government and managers, in sentiment analysis research This includes using the Nave Bayes classification algorithm to examine the sentiment of tourist visitors and the performance of the classification model used. The data used in this research was taken from the website from Tripadvisor and Google Maps using a crawling technique, which then processed the data by a pre-processing process and then applied a classification to the data and got a sentiment visualization by processing word frequency on tourist visitor sentiment data. The results of the accuracy of the model used were re-tested with the k-fold cross validation method and the results of sentiment visualization got the frequency of words that most often appear on negative sentiment labels are garbage, beaches, lacking, places, roads, parking, dirty, entering, caring, clean , expensive, pay, manage, good and water.

Recording financial transactions is an important activity that must be carried out by a business. In accounting, recording financial transactions is a fundamental thing that must be done. With these records, all business financial transactions can be accounted for and brought to the financial statements. Computerized technology using barcodes in payment transactions has the aim of increasing speed, accuracy, and service quality. In carrying out payment administration activities still using the manual system will have an impact on the Administrative officer having difficulty finding data on students who want to pay and the network (LAN) has not been implemented on the computer is also a problem at the research site, because to share data with the treasurer and principal becomes constrained, data can be damaged, lost or misused because its security is not guaranteed. In solving these problems, the researchers collected data and facts that existed at the research site, then designed a payment administration information that was in accordance with the needs of the school. In making Administrative Information System Integrated With Barcode Reader, researchers used Microsoft Visual Basic 6.0 programming language, MySQL Server database as data storage media and Subtime Text as tools to create source code. In the preparation of this application using object-oriented methods.

Laibu Samsaman : with the title "Effectiveness of the Process of Determining Village Devices in the Administrative Village of Otademan, Wakate District, Eastern Seram Regency". Accompanied by Advisor I: Z. A. Rengifurwarin and Advisor II: I. S. Ponto. This research is a descriptive qualitative research. Aims to describe and explain an overview of the Effectiveness of the Process of Determining Village Devices in the Administrative Village of Otademan Kec. Wakate Kab. SBT. Informants of this study amounted to 10 people. The data needed in this study were collected through observation, interviews and documentation techniques. Data was successfully collected, processed and analyzed with the help of descriptive qualitative analysis techniques. The results of this study indicate that the effectiveness of determining Otademan Administrative village apparatus is generally ineffective or not optimal, which is seen based on a) accuracy of age/age, b) accuracy of education, c) accuracy of work experience, d) accuracy of work skills, e) accuracy as villagers. This is influenced by supporting and inhibiting factors such as: a) likes and dislikes, b) group fanaticism, c) nepotism, and d) loyalty. The implications of this research, if you want to increase the effectiveness of determining village officials in the Otademan Administrative Village, Kab. SBT. district Wakate, the things that need to be considered are the level of accuracy of age/age, accuracy of education, accuracy of work experience, accuracy of work skills, and accuracy as a villager.

For an organization, speed and accuracy in obtaining information has become a necessity. The existence of an information system becomes a necessity to solve the problem of speed and accuracy in accessing information. XYZ Electronic Store before implementing the information system, had difficulty in knowing quickly and accurately about sales information and how much the store's debt had not been paid. The research method used in this research is observation, interviews and design and implementation of the system. Design system design with DFD, ERD and implementation with PHP programming language. Design system design with the prototype method. After implementing the sales accounting information system and debt recording, the problems faced by XYZ Electronic Store can be resolved. Information on the number of sales in a certain period can be quickly identified, and can schedule debt payments. Debt payment time can be set because the system records all complete debt lists along with the due date.

Starting from the development of medical technology that is increasingly sophisticated and rapidly growing,researchers conduct medical research, namely about patient infusion handling services. In handling patient infusion, currently it is still manual which is carried out by nurses / medical personnel. Infusion handling services for patients still have shortcomings, namely the process of monitoring and replacing infusion fluids which are often late. If the problem is not treated quickly, it can lead to problems, namely the presence of air embolism in the blood vessels (the entry of foreign objects into the blood vessels, for example air). From that problem, the researchers made a new innovation in medical technology in handling infusions automatically and based on IoT. In this study, the smart online infusion device that has been made has good features and is very effective in handling infusions. This device has 3 main functions, namely: it can monitor the remaining infusion, it can change the infusion fluid automatically and it can indicate a blocked patient's infusion. This device already has a method for processing data with fuzzy logic. Media monitoring has been supported by a website that can be controlled remotely and in real time. Tests have been carried out and the effectiveness of the system is found to have an error rate of 0.2% - 0.7% and has an accuracy of 98%. Thus this tool can be used in terms of handling patient infusion automatically.