The performance of the organizations or companiesare based on the qualities possessed by their employee. Both of good or bad employee performance will have an impact on productivity and the impact of profits obtained by the company. Support Vector Machine (SVM) is a machine learning method based on statistical learning theory and can solve high non-linearity, regression, etc. In machine learning, the optimization model is a part for improving the accuracy of the model for data learning. Several techniques are used, one of which is feature selection, namely reducing data dimensions so that it can reduce computation in data modeling. This study aims to apply the method of machine learning to the employee data of the Bank Rakyat Indonesia (BRI) company. The method used is SVM method by increasing the accuracy of learning data by using a feature selection technique using a wrapper algorithm. From the results of the classification test, the average accuracy obtained is 72 percent with a precision value of 71 and the recall value is rounded off to 72 percent, with a combination of SVM and cross-validation. Data obtained from Kaggle data, which consists of training data and testing data. each consisting of 30 columns and 22005 rows in the training data and testing data consisting of 29 col-umns and 6000 rows. The results of this study get a classification score of 82 percent. The precision value obtained is rounded off to 82 percent, a recall of 86 percent and an f1-score of 81 percent.

Gambar merupakan salah satu alat bukti tindak kejahatan. Gambar dapat memberikan banyak informasi dianataranya tentang pelaku, jam kejadian, cara terjadi dan lainnya. Ada banyak alat yang telah dihasilkan dengan menggunakan Node MCU ESP32 CAM. Penggunaan ESP 32 CAM karena memiliki fitur kemampuan menangkap gambar dengan modul kamera yang telah terpasang. Penggunaan ESP 32 CAM banyak digunakan pada proyek Internet of Things (IoT) karena memiliki modul Wifi yang terpasang onboard. Pada proyek penangkap gambar yang mendeteksi manusia maka ESP 32 CAM membutuhkan modul tambahan sebuah Sensor PIR Motion. Ada banyak jenis sensor PIR Motion dengan jarak jangkau yang beragam dan waktu respon yang berbeda. Pada penelitin ini dilakuakan pengujian beberapa jenis sensor PIR Motion untuk mendapatkan data PIR Motian dengan jangkauan terjauh dan respon tercepat. Dikembangkan rancangan alat penangkap gambar berbasis ESP 32-CAM. Hasil penelitian menunjukkan bahwa sensor PIR Motion terbaik ialah seri HC-SR501 dan rancangan rangkaian alat penangkap gambar dapat bekerja pada ruang gelap dan memiliki sumber tegangan sendiri dari batrai yang dapat diisi ulang.Pictures are one of the evidence of a crime. Images can provide a lot of information including about the perpetrator, the time of the incident, how it happened and others. There are many tools that have been generated using the Node MCU ESP32 CAM. The use of ESP 32 CAM because it has the ability to capture images with a camera module that has been installed. The use of ESP 32 CAM is widely used in Internet Of Things (IoT) projects because it has a Wifi module installed onboard. In the project of capturing images that detect humans, the ESP 32 CAM requires an additional module, a PIR Motion Sensor. There are many types of PIR Motion sensors with varying ranges and different response times. In this study, several types of PIR Motion sensors were tested to obtain Motian PIR data with the farthest range and fastest response. ESP 32-CAM based image capture tool was developed. The results show that the best PIR Motion sensor is the HC-SR501 series and the design of the image capture device can work in a dark room and has its own voltage source from a rechargeable battery.

Bagi perusahaan yang bergerak dibidang sektor jasa, seperti perusahaan tour and travel pengolahan data sangatlah penting untuk mengetahui karakteristik atau minat wisatawan dalam berwisata. Sulitnya memprediksi kebutuhan atau minat wisatawan, merupakan kendala yang dihadapi perusahaan tour and travel sehingga manajemen harus dapat mengambil keputusan yang tepat dan cepat, guna memberikan pelayanan yang baik serta kepuasan kepada customer. Keputusan yang diambil harus mempertimbangkan dengan baik berdasarkan data-data yang dimiliki terutama yang berkaitan erat dengan karakteristik data traveller. Analisis data untuk mencari pola karakteristik data dari traveler memutuhkan metode yang baik dan hasil yang akurat , sehingga hasil dari analisis tersebut dapat menemukan informasi yang berguna bagi perusahaan. Metode pengeolahan data yang sering digunakan ialah data mining. K-Nearest Neighbour (K-NN) merupakan salah satu algoritma data mining yang cara kerjanya menerapkan pembelajaran pada data. Tujuan dari penelitian ini akan melakukan analisis data travellers atau wisatawan pada perusahaan tour and travel dengan menggunakan algoritma K-NN  untuk mencari minat atau karakteristik wisatawan dalam memilih objek wisata. Pada hasil kualifikasi metode K-NN dan pengujian dengan metode confusion matrix nilai akurasi yang didapat 84%,  nilai presisi 88%, nilai recall 85 dan nilai f1-score 85%. Pada hasil kualifikasi menunjukan bahwa objek wisata yang cenderung dominan menjadi minat wisatawan ada pada objek wisata pantai.Berdasarkan hasil analisis dengan menggunakan kedua metode tersebut bahwa karakteristik wisatawan cendrung memilih paket nomor 3 yakni Reguler Tour Lombok Packagage, dan objek wisata yang dominan minat wisatawan yakni objek wisata pantai dan disusul oleh objek wisata gili.

Indonesia is one of the countries with the most favorite tourist objects in the world. There are many cultures, customs and natural beauty that are still preserved. One type of tourism that is closely related to history and ancestral heritage is cultural heritage. Cultural heritage has a high historical value and witnesses the civilization of life on earth. One of the districts in West Nusa Tenggara, namely Central Lombok Regency, has 150 cultural heritages spread across various sub-districts. 7 of these cultural heritages have been recorded at the national and even international levels. However, due to limited sources of information and access to cultural heritage sites, tourists and local people do not know the locations and history of the cultural heritage sites. So that research is carried out with the aim of obtaining valid data and information to be promoted in the form of videos that can be watched by anyone, anywhere and anytime via YouTube and can be used as a medium of learning about the life history of the ancestors of the Sasak tribe in Central Lombok Regency.In the research conducted, the researchers collected data and information by means of literature study, observation and direct interviews to cultural heritage locations and the Tourism Office of Central Lombok Regency. Method of analysis with SWOT (Strengths, Weaknesses, Opportunities, Threats). The design method uses the multimedia production process flow method with Adobe Audition software as audio editing and Adobe Premiere Pro 2020 as video editing. And the questionnaire test is used to obtain responses from the respondents by calculating the Likert scale

Foreign tourists entering Indonesia in 2017 and 2018 have increased. From the data obtained on the website of the Ministry of Tourism (Kemenpar) the number of foreign tourists in 2017 was 14,039,799, while in 2018 there were 15,806.1, with a comparison of the number of tourists from the two years, the percentage increase in tourists was 12.58%. The data analysis approach using a classification model is a data analysis approach by studying the data and making predictions with the new data. in the classification model, there are many algorithms that can be applied in data analysis, one of which is the Decision Tree algorithm. This study aims to analyze the pattern of tourist visits based on the objects visited by the number of tourists visiting certain tourist objects. From the modeling using the Decesion Tree C4.5 Algorithm and the scenario of splitting the data into three parts, the highest accuracy value was obtained for splitting data of 80:20 for train and testing data and max depth 7, which obtained an accuracy of 94% for train data and 92% for data. testing. Modeling with the Boostrap Aggregating Method, the accuracy score obtained on training data is 93% and testing data is 92. percent. 3 accuracy results from using bagging reduce the accuracy of the C4.5 algorithm on the data training side from 94% to 93 percent, while the accuracy of testing data is still the same, namely 92%.