Karya tulis ilmiah (KTI) merupakan karya ilmiah yang ditulis dengan mengikuti kaidah ilmiah. Kaidah ilmiah sebagai syarat utama dalam penulisan sebuah karya dimaksudkan agar karya yang dihasilkan dapat dipertanggung jawabkan secara ilmiah. Tujuan kegiatan pelatihan penulisan karya tulis ilmiah yaitu menumbuhkan minat, semangat, serta ide kreatif dan inovatif dari siswa-siswi kelas X dan XI SMAN 5 Samarinda untuk menghasilkan sebuah karya ilmiah yang sesuai dengan kaidah penulisan. Berdasarkan hasil pelaksanaan kegiatan pelatihan dapat disimpulkan bahwa kegiatan berjalan dengan baik dan mendapat dukungan penuh dari pihak sekolah. Seluruh peserta pelatihan mengikuti kegiatan hingga akhir dengan tingkat kehadiran sebesar 100%. Peserta kegiatan antusias untuk bertanya, mengeksplorasi ide, serta mengemukakan pendapat. Dengan demikian, kedepannya diharapkan adanya kegiatan lanjutan dengan melibatkan guru pendamping untuk mengoptimalkan perannya dalam penyusunan karya tulis ilmiah bagi peserta didik.Kata Kunci: kaidah ilmiah, KTI, peserta didik. Training on Writing Scientific Papers to Encourage Quality Improvement of High School Level Students ABSTRACT The scientific paper is an essay written by following scientific rules that are the main requirement so that the resulting essay can be justified scientifically. The purpose of the training is to increase the interest, enthusiasm, creative, and innovative ideas from students of class X and XI of SMAN 5 Samarinda to create a scientific paper that is following the rules. Based on the implementation of the training, it can be concluded that it is run well and received support from the school. All participants follow this activity until the end with an attendance rate of 100%. They are enthusiastic to ask, explore, and express their ideas and opinions. Then, in the future, it is expected that there will be further activities involving the teachers to optimization the role of assistants to create their student’s scientific papers.Keywords: scientific paper,scientific rules, students.

Decision tree is a algorithm used as a reasoning procedure to get answers from problems are entered. Many methods can be used in decision trees, including the C5.0 algorithm and Classification and Regression Tree (CART). C5.0 algorithm is a non-binary decision tree where the branch of tree can be more than two, while the CART algorithm is a binary decision tree where the branch of tree consists of only two branches. This research aims to determine the classification results of the C5.0 and CART algorithms and to determine the comparison of the accuracy classification results from these two methods. The variables used in this research are the average monthly income (Y), employment (X1), number of family members (X2), last education (X3) and gender (X4). After analyzing the results obtained that the accuracy rate of C5.0 algorithm is 79,17% while the accuracy rate of CART is 84,63%. So it can be said that the CART method is a better method in classifying the average income of the people of Teluk Baru Village in Muara Ancalong District in 2019 compared to the C5.0 algorithm method.   Keywords: C5.0 Algorithm, CART, Classification, Decision Tree.

Potential Villages (PODES) provide data on the existence, availability and development of the potential of each government administrative area. In order to make it easier for governments to make policies for a region, it is necessary to group the village and sub-districts. Cluster analysis is an analysis that aims to group objects based on the information that found in the data. One of the cluster analysis methods is the divisive analysis, which is a hierarchical grouping method with a top-down approach, where all objects are placed in one cluster and then sequentially divided into separate groups. This research aim to group villages or sub-districts in Kutai Kartanegara based on the determinants of village backwardness and obtaining the silhouette coefficient value from the optimal cluster analysis using the divisive analysis algorithm. The data used is the 2018 PODES data in Kutai Kartanegara and used 15 variables from natural and environmental factors, facilities infrastructure and access factors as well as socio-economic factors of the population. The results of the optimal cluster formed in the grouping of villages or sub-districts in Kutai Kartanegara using the divisive analysis method are 2 clusters. Cluster 1 consisting of 230 villages or sub-districts and cluster 2 consisting of 2 sub-districts. Silhouette coefficient value for data validation from clustering village or sub-districts in Kutai Kartanegara using the divisive analysis method produces 2 clusters is 0,744 which states that the cluster structure formed in this grouping is a strong structure.

Classification is a technique to form a model of data that is already known to its classification group. The model was formed will be used to classify new objects. Fisher discriminant analysis is multivariate technique to separate objects in different groups. Naive Bayes is a classification technique based on probability and Bayes theorem with assumption of independence. This research has a goal to compare the level of classification accuracy between Fisher's discriminant analysis and Naive Bayes method on the insurance premium payment status customer. The data used four independent variables that is income, age, premium payment period and premium payment amount. The results of misclassification using the APER (Apparent Rate Error) indicate that the naive Bayes method has a higher level of accuracy is 15,38% than Fisher’s discriminant analysis is 46,15% on the insurance premium payment status customer.

Cluster analysis has the aim of grouping several objects of observation based on the data found in the information to describe the objects and their relationships. The grouping method used in this research is the Fuzzy C-Means (FCM) and Subtractive Fuzzy C-Means (SFCM) methods. The two grouping methods were applied to the people's welfare indicator data in 42 regencies/cities on the island of Kalimantan. The purpose of this study was to obtain the results of grouping districts/cities on the island of Kalimantan based on indicators of people's welfare and to obtain the results of a comparison of the FCM and SFCM methods. Based on the results of the analysis, the FCM and SFCM methods yield the same conclusions, so that in this study the FCM and SFCM methods are both good to use in classifying districts/cities on the island of Kalimantan based on people's welfare indicators and produce an optimal cluster of two clusters, namely the first cluster consisting of 10 Regencies/Cities on the island of Kalimantan, while the second cluster consists of 32 districts/cities on the island of Borneo.

The hybrid method is a method of combining two forecasting models. Hybrid method is used to improve forecasting accuracy. In this study, the Time Series Regression (TSR) linear model will be combined with the Autoregressive Integrated Moving Average (ARIMA) model. The TSR linear model is used to obtain the model and residual value, then the residual value of the TSR linear model will be modeled by the ARIMA model. This combination method will produce a hybrid TSR linear-ARIMA model. The case study in this research is stock closing price (daily) of PT. Telkom Indonesia Tbk. The stock closing price (daily) of PT. Telkom Indonesia Tbk in 2020 showed an decreasing and increasing trend pattern. The results of this study obtained the best model of hybrid TSR linear-ARIMA (2,1,1) with the proportion of data training and testing is 70:30. In the best model, the MAD value is 56.595, the MAPE value is 1.880%, and the RMSE value is 78.663. It is also found that the hybrid TSR linear-ARIMA model has a smaller error value than the TSR linear model. The results of forecasting the stock price of PT. Telkom Indonesia Tbk for the period 02 January 2021 to 29 January 2021 formed a decreasing trend pattern.

This research focuses on identifying factors affecting mathematics scores in junior high school students during the COVID-19 pandemic in Samarinda, Indonesia. We estimate factors of five domains of the mathematics score for e-learning in junior high school students during the COVID-19 pandemic. The factors are sociodemographic (gender and degree level), students’ habits of accessing the Internet outside online learning, tutoring program, learning platforms (video conferencing, online classes, and social media), and skills of digital literacy (internet searching, knowledge assembly, and content evaluation). A total of 160 students in the junior high school in Samarinda were selected accidentally. An online survey was conducted based on a structured questionnaire. The dummy variables multiple linear regression model was used in this research to identify factors affecting mathematics scores in the junior high school students during the COVID-19 pandemic in Samarinda. The novelty of this research is in finding the factors that have a significant positive effect on mathematics scores, namely gender (female), tutoring programs, video conferencing platforms, internet searches, and content evaluation. While the factors that have a significant negative effect on the value of mathematics are the habit of accessing the Internet outside online learning and knowledge assembly. Based on these findings, we recommend the teachers to use video conferencing as the best practice for teaching mathematics. As research implications, both teachers and students can discuss improving the students’ digital literacy skills.

No abstract