— In this pandemic era, the government is forced to implement policies that can reduce the daily positive rate of COVID-19. One of these policies is known as PPKM. It is unclear when the pandemic will end, causing data phenomena to be scattered on social media, one of which is Twitter. Therefore, in this study, we conducted an analysis of sentiment originating from tweets from Twitter social media users in the Jakarta area regarding the government's policy, namely PPKM in the face of the COVID-19 pandemic. In this research, we use a Machine Learning approach, namely LSTM. This modeling produces a classification of positive and negative sentiments. The dataset used is 3000 tweets with a time period of September - November 2021. At the preprocessing stage, the data that are ready to be used for modeling are 2176. The results of this study get an accuracy of 0.943. So the model that we propose, namely LSTM, has succeeded in classifying a satisfactory sentiment with a positive number of 92% and a negative 8% of 2176 sentiments, so it can be concluded that the PPKM policy implemented by the Indonesian government in the DKI Jakarta area is said to be effective or positive.
Pelayanan merupakan deretan peristiwa yang terjadi dalam berinteraksi langsung kepada masyarakat disekeliling atau dengan alat secara fisik dan menyajikan fasilitas yang memuaskan untuk penduduk lebih tepatnya dalam proses pembuatan surat pengantar akta kelahiran, seperti penjadwalan tanggal yang terlalu lama setelah penduduk mendaftar. Pada pelayanan kependudukan membutuhkan inovasi terbaru untuk meningkatkan dalam memberikan pelayanan yang baik tepat dan cepat kepada masyarakat. Proses pelayanan pada pembuatan jadwal pengambilan surat pengantar akta kelahiran yang dicatat dengan proses manual belum bisa dibuat secara maksimal sehingga perlu adanya perbaikan konsep kegiatan yang dikelola secara maksimal melalui penggunaan aplikasi optimalisasi penjadwalan yang tersusun secara otomatis. Oleh karena itu, kebutuhan tentang optimalisasi penjadwalan menjadi penting dan pembuatan aplikasi pada pelayanan tersebut harus dibangun secara lebih lengkap dan meliputi ruang lingkup yang lebih luas. Penelitian ini bertujuan untuk mengimplementasikan penjadwalan pengambilan akta kelahiran menggunakan Algoritma Genetika, serta mendefinisikan Algoritma Genetika dalam proses penjadwalan. Algoritma Genetika yang terdiri dari definisi individu, inisialisasi chromosome, evaluasi chromosome, proses seleksi (crossover) dan proses mutase. Hasil dari algoritma genetika berupa tanggal dan bulan yang digunakan untuk pengambilan surat pengantar akta kelahiran berupa data penjadwalan yang tepat agar tidak terjadi keterlambatan. Hasil dari penelitian ini adalah berupa penjadwalan yang sudah terdapat waktu ruangan yang sudah tersusun lebih cepat dan akurat dari pada proses penjadwalan manual yang memakan waktu yang cukup lama sehingga menyulitkan petugas di catatan sipil serta menciptakan kinerja yang lebih maksimal dalam proses pemilihan dan pengolahan data terhadap aplikasi yang sudah dikelola sehingga menciptakan penjadwalan yang lebih terstruktur.
30 June 2023 in Journal of Applied Computer Science and Technology
This study examines the efforts to develop a model for analyzing public sentiment regarding applying ETLE (Electronic Traffic Law Enforcement) regulations. The method used is the systematic literature review. A systematic literature review (SLR) consists of three stages: planning, conducting, and reporting. The planning stage is the determination of the SLR procedure. This stage includes preparing topics, research questions, article search criteria & inclusion and exclusion criteria. The conducting stage, namely the implementation, includes searching for articles and filtering articles. The reporting stage is the final stage of SLR. This stage includes writing the SLR results according to the article format. The explanation follows: First, hybrid is the most widely used method in developing sentiment analysis models. Apart from hybrid, several methods are used to develop sentiment analysis models, including multi-task, deep, and machine learning. Each has its advantages and disadvantages in the development of sentiment analysis models. Second, this study shows the development of a model with superior performance, namely using XGBoost as a sentiment analysis model, and the stages it goes through are preprocessing data, handling imbalanced data, and optimizing the model. Therefore, the model for analyzing public sentiment regarding the application of ETLE regulations can be an option for hybrid methods, multi-task learning, deep learning, machine learning, and the XGBoost model to obtain superior performance with preprocessing data stages, handling imbalanced data and optimization models.
1 April 2019 in TELKOMNIKA (Telecommunication Computing Electronics and Control)
The Government of the Republic of Indonesia is currently focusing on building logistics infrastructure to facilitate the distribution of logistics to all regions in Indonesia. The Distribution of logistics to each area requires an electronically connected warehouse so that information about goods in the warehouse can be monitored continuously. There are some problems with one of the logistics companies because the existing warehouse management technology is not sufficient enough, so the complex warehouse functions do not become dynamic. One of the solutions to solve the problem need to use RFID technology, conveyor belt, and robotic arm by using an Arduino microcontroller. This research uses Design Science Research Methodology which focuses on developing and improving the model performance of a system and using the prototyping model for system development. The test results state that the built system works well because it could lift, identify and sort goods by type. So, this research could answer the problem that happened at the warehouse.
In the era of big data and cloud computing, digital records of supermarket sales data and other accompanying factors are ubiquitous. However, less than optimal WeeklySale s can occur due to several factors influencing it. This study proposes particle swarm optimization with gradient boosting regression (PSO-GBR) as a solution for optimizing the association rule in supermarket sales based on a regression model that can predict WeeklySales. As a benchmark for this research, we compare our proposed GBR with two legacy prediction methods: linear regression (LR) and AdaBoost Regression (ABR). The first step is data preparation. Then we develop a model that can predict sales from the dataset using GBR. The next step is to evaluate the model with benchmark methods. Then with an optimum regression method, we optimize sales using PSO. The last step is to show that the method provides optimum WeeklySale s results. The results show that our proposed GBR has a higher $R^{2}$ score than LR and ABR, namely 0.95 and 0.94 for train data and test data, respectively. Then PSO is proven to optimize sales using the GBR model as a cost function. PSO can increase the ten lowest WeeklySales of the actual dataset from a total of US$ 45 to a total of US$ 2, 296.9 by selecting a more optimized Department according to the GBR prediction. This research proves that we can use a regression model with good performance for a cost function in PSO optimization for department sales.