This research applying predictive data mining on CPO (Crude Palm Oil) ABSTRAKPenelitian ini menerapkan data mining pada prediksi harga CPO (Crude Palm Oil) dengan membandingkan pemodelan optimasi seleksi fitur algoritma genetika dan algoritma greedy pada metode neural network (NN). Prediksi harga CPO dilakukan untuk memenuhi kebutuhan investor kelapa sawit, melalui analisa masalah fluktuasi harga CPO time series yang tidak pasti. Guna mempermudah dalam melakukan perhitungan, langkahlangkah dari algoritma Genetika dan algoritma Greedy diimplementasikan dengan program komputer Rapid Miner Studio. Adapun tujuan penelitian ini yaitu mengetahui perbandingan akurasi dengan parameter evaluasi RMSE yang dihasilkan dan waktu eksekusi program yang dibutuhkan oleh algoritma Genetika dan algoritma Greedy dalam menyelesaikan masalah prediksi harga CPO. Hasil pengujian akurasi menunjukkan bahwa penggunaan metode NN optimasi algoritma Genetika secara umum memberikan nilai RMSE yang lebih baik (0,096) dibandingkan algoritma Greedy-forward selection (0,111) dan algoritma Greedy-backward selection (0,101). Akan tetapi jika ditinjau dari waktu eksekusi program yang dibutuhkan dalam menyelesaikan masalah prediksi harga CPO, maka algoritma Genetika membutuhkan waktu yang lebih lama dari pada algoritma Greedy.
Oyster Mushroom is one of the high-value consumption mushrooms. The development of mushroom experienced rapid growth marked by the increasing number of farmers breeding mushroom is directly proportional to number of mushroom food business. The process of mushroom cultivation in mushroom’s greenhouse depends on physical factors such as temperature, humidity, light, pH of planting medium, and air aeration. Watering activity is done if the condition and humidity of mushroom's greenhouse was dry, three times a day at the morning, afternoon and evening. If the conditions of temperature and humidity fluctuate in the current seasons, it is not enough to be sprayed. High temperature and humidity fluctuations disturbed harvest. The results of this research are realtime temperature and humidity control system, esp8266 12-F and DHT22 sensor to read humidity value that can do watering and keep humidity greenhouse. The best humidity limit is 65% and maximum is 80%. If the humidity condition reaches 65%, then esp8266 will activate the relay to turn on the DC pump machine and nozzle spray. If humidity reached 80%, then esp8266 will turn off the DC pump. 80% humidity limit setting to avoid mushrooms from getting too wet. Humidity conditions monitored with android smartphone in realtime.
The occurrence of a tire leak in the middle of a trip can often be caused by many things, such as being hit by a sharp object, the age of the tire is too old, leaking in the former patches, or it can be caused by others. These factors make vehicle users panic and think to immediately find the nearest tire repair location. Almost everyone has a smartphone where there is a Google Map application by utilizing GPS technology so as not to get lost when driving, with GPS technology can be estimated the distance between one location and another. The purpose of this study is to produce a GIS mobile application that provides a tire patch location search feature with information, so that it is expected to be useful for vehicle users if they experience a tire accident or erupt on the road. The workshop information that is presented is the opening / closing time of the workshop, there are nitrogen added facilities or not, tubeless tire patches, can repair motorcycle tire patches or can also tire the car and the contact of the workshop owner, so you can directly contact the nearest tire patch. The results of testing the user's response to the use of the application indicate that the application is useful.
The tire repair location is one of the determining factors sought by motorized vehicle users if they are experiencing a tire leak in the middle of a trip. This makes vehicle users panic and think to immediately look for the nearest tire repair location. Banjarbaru City has 5 sub-districts and 12 villages, and is the center of the South Kalimantan Province administration. Given the large number of settlements, especially now that there are many housing complexes in Banjarbaru, this automatically makes the number of new roads opened by the local government. Actually this can be an opportunity for tire repair entrepreneurs to open tire repair locations in several places, because nowadays there are still very few workshops or tire repair locations in the new residential area.Location Based Service System, or better known as LBS, combines the process of a mobile service with the geographical position of its users. The important point is when the target position, where a target might be an LBS user or another entity that is incorporated in a service. This research leads to android based applications and services that provide information of tire repairlocation, to make it easier for motorbike riders to find tire repair locations based on available facilities.
Pola konsumsi masyarakat akan produk hasil ternak semakin meningkat, tak terkecuali konsumsi ayam potong (broiler). Tingginya tingkat konsumsi masyarakat akan ayam potong otomatis akan memicu para produsen ternak untuk meningkatkan produksinya. Produksi ayam potong dapat diprediksi, salah satunya menggunakan Fuzzy Time Series. Perkembangan metode peramalan data time series yang cukup pesat mengakibatkan terdapat banyak pilihan metode yang dapat digunakan untuk meramalkan, salah satunya yaitu metode Fuzzy Time Series Markov Chain Model. Konsep Markov Chain digunakan dalam proses prediksi jumlah produksi ayam potong dengan menggunakan matriks transisi. Dengan Fuzzy Time Series Markov Chain Model, yang merupakan gabungan dari konsep metode fuzzy time series dengan model Markov, diharapkan dapat mendatangkan hasil analisis prediksi yang lebih akurat. Penelitian ini diharapkan dapat membantu proses analisis prediksi, sehingga proses analisis prediksi dapat dilakukan secara lebih efisien, teliti, mudah, dan praktis.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.