SINTECH (Science and Information Technology) Journal

Implementasi K-Medoids Clustering Dalam Pengelompokkan Harga 8 Jenis Minyak Goreng

Fina Nasari, Andri Nofiar Am — Sun, 31 Dec 2023 00:00:00 +0000

In everyday life, cooking oil has become a necessity with international sales prices varying depending on the quality and type. The types of cooking oil sold in the international market are very diverse, including coconut, olive, palm kernel oil, palm oil, peanuts, rapeseed, soybeans and sunflower. Therefore, it is necessary to group data on selling prices of cooking oil on the international market to get the best grouping of cooking oil. The data used in this research is historical price data of 8 edible oils kaggle August 1992 to July 2022. Data grouping in this research uses the k-medoids algorithm. The k-medoids algorithm produces a more balanced group, better performance and accuracy than other algorithms. The aim of this research is that the k-medoids algorithm is able to group cooking oil price data into 4 group models, namely group models 2, 3, 4 and 5 and obtain the best group model based on the dbi value. The research results showed that the cooking oil price data was successfully grouped into group 2, 3, 4 and 5 models with the best group based on the lowest dbi performance value being the group 2 model with a dbi value of 0,000 and olive oil being the cooking oil with the highest price in the world while 7 types other cooking oils have more or less the same price (in the same price group).

Penerapan Metode Clustering Dalam Segmentasi Pelanggan Perusahaan Logistik

Hanan Kishendrian, Nisa Hanum, Cahyo Prianto, Woro Isti Rahayu — Sun, 31 Dec 2023 00:00:00 +0000

Marketing is important in business to compete and maintain market share. The development of technology brings major changes in the industry. In addition to product development as well as required services, and customer segmentation becomes a factor to consider in marketing strategies. Clustering, such as the K-Means method, is used in customer segmentation to divide data into groups based on similarities. This technique helps in useful pattern recognition and customer segmentation. By applying Clustering techniques in Data mining, companies can understand customer behavior, recognize similar customer groups, and plan marketing strategies accordingly. The results showed that the best cluster was generated with a k value of 4, and the data was normalized using the Min-Max Normalization method. Grouping customers in the form of clusters can enable the identification of consumer profiles to guide companies in decision making.

Komparasi Naïve Bayes dan SVM Analisis Sentimen RUU Kesehatan di Twitter

Tetrian Widyanto, Ina Ristiana, Arief Wibowo — Sun, 31 Dec 2023 00:00:00 +0000

This research focuses on sentiment analysis regarding the plan to ratify the Health Bill which has become a hot topic of conversation on social media, especially Twitter. This research aims to classify tweets that reflect various opinions regarding the Health Bill, including support, rejection and neutrality. In this research, the author uses two types of classification algorithms, namely the Multinomial Naïve Bayes Algorithm and the Support Vector Machine (SVM) Algorithm. Previously, tweets were labelled using the Lexicon InSet dictionary. The research was conducted in the Python programming language and using Google Collaboratory. Data validation was carried out using the K-fold cross-validation method. The research results indicate that both algorithms predominantly produce positive sentiments over negative ones. However, SVM with a linear kernel achieves a higher accuracy rate of 0.87, compared to Multinomial Naïve Bayes, which has an accuracy of 0.82. SVM also records a precision of 0.87, recall of 0.97, and an F1-score of 0.91, while Multinomial Naïve Bayes shows a precision of 0.81, recall of 0.98, and an F1-score of 0.89. Overall, this research confirms that SVM excels in text sentiment classification, while Multinomial Naïve Bayes also provides good results in recognising positive and negative sentiment. These results have important implications for understanding public sentiment regarding the Health Bill on the Twitter platform.

Perbandingan Kinerja Algoritma K-Means dan K-Medoids Dalam Klasterisasi Jumlah Tindak Pidana Kejahatan Berbasis Wilayah Kepolisian Daerah

Gelar Nurcahya, Arief Wibowo, Dwi Kristanto — Sun, 31 Dec 2023 00:00:00 +0000

Criminal acts are often a problem that occurs in Indonesia. Where currently the number of reports handled by the police regarding criminal acts is always there every day. Indonesia's population is increasing and the background of perpetrators who are unemployed is often one of the reasons why the police find it difficult to resolve criminal acts that occur due to limited human resources. To overcome this problem, information is needed that provides areas in Indonesia where criminal acts frequently occur so that the police can make decisions to allocate human resources to protect those jurisdictions from criminal acts that occur. Using data on criminal offenses and the employment of criminal offenders, namely not working from 2021, data was taken from the National Police Criminal Investigation Unit's Pusiknas Annual Journal. The data will be clustered using data mining techniques using the K-Means and K-Medoids algorithms. These 2 algorithms produced 2 clusters with the smallest Davies Bouldin index value found in the K-Means algorithm with a value of 0.272. With the research results which produced 2 clusters, it can be concluded that there are categories of high crime and low crime.

Recognizing Hotel Visitors Preferences Based on Service Consumption Level Using K-Means Method

Sun, 31 Dec 2023 00:00:00 +0000

Consumer segmentation is an old issue that remains interesting to study today, given the magnitude of the benefits obtained when consumers can be segmented properly. Marketing cost efficiency is one of the great benefits of this process. Likewise, the effectiveness of marketing activities to maintain customer retention. It is because companies can better identify consumers. Based on the hotel service consumption level, this research could identify consumer clusters based on hotel consumer preferences. Thus, hotel management could target specific types of service promotion better and on target. This research built a hotel visitor clustering model using the K-Means Clustering method to determine customer segments based on the level and type of hotel service consumption. The K-Means model was built based on hotel visitor consumption data for each type of service. Furthermore, the hotel visitor clusters formed were identified by their characteristics. Four consumer clusters were obtained based on the silhouette score analysis, which described the characteristics of consumers in each cluster.

Digitalisasi Prasasti dan Pelinggih Desa Baturan Gianyar Berbasis Augmented Reality Based Marker

Sun, 31 Dec 2023 00:00:00 +0000

Digitalization of the inscriptions and shrines of Batuan Gianyar Village based on Augmented Reality (AR) Markers is an innovative project that aims to preserve, promote and revive the cultural and historical heritage of Baturan Village, Gianyar, Indonesia, through the application of Augmented Reality technology. Baturan Village is known to have inscriptions and pelinggih which have high historical value. This research uses AR Marker technology to connect the physical world with digital content. The inscriptions and shrines of Baturan Village are marked with AR markers which allow users to access additional information about these objects via mobile devices. This AR application provides an immersive and interactive experience, allowing users to explore the history and meaning of inscriptions and shrines. The method used in developing this AR application is R & D. The results of this research after testing loading time using four different smartphones, Xiaomi Note8 Pro and iPhone 13, have a faster average loading time response compared to other smartphones.