Abstrak

It is very difficult to determine the eligibility of recipients of housing assistance for poor families, especially in gampongs, based on the number of gampong residents who apply. the large number of people who applied for assistance made it difficult for village officials and decision-makers to make decisions. The existence of a model analysis in the eligibility of recipients of housing assistance so that there is no subjectivity in the provision of assistance. The purpose of this research is to identify beneficiaries of housing assistance for underprivileged communities who meet the requirements using the Analytical Network Process (ANP) and Borda models. The variables used are parents' work (C1), number of parental dependents (C2), sources of income costs (C3), housing conditions (C4), status (C5) and education (C6). The results of selecting the ANP model were that Aminah got a score of 1.445, Hasanah (A2) 1.415, Baihaki (A3) 1.6148, Fakri (A4) 1.53. The recommendation from the ANP model analysis is baihaki with a value of 1.6148. while from the borda model the total values are A1:355, A2:355, A3:212, A4:381 A5:254, A6:209, with a weight value of 0.201, A2 weight: 0.201, A3 weight 0.120, A4 weight 0.216, A5 weight 0.144 and the weight of A6 0.118. The conclusion from the research results gives priority to the community for the highest value in obtaining housing assistance. this study which received priority baihaki with a value of 1.6148 for the results of the ANP model and the results of the value of the model group was A3 with a value of 0.272.

Pneumonia is an illness that affects practically everyone, from children to the elderly. Pneumonia is an infectious disease caused by viruses, bacteria, or fungi that affect the lungs. It is quite difficult to recognize someone who has pneumonia. This is because pneumonia has multiple levels of classification, and so the symptoms experienced may vary. The multilayer perceptron approach will be used in this study to categorize Pneumonia and determine the level of accuracy, which will contribute to scientific development. The Multilayer Perceptron is employed as the classification method with hyperparameter learning rate and momentum, while SURF is used to extract the features in this classification. Based on the experiments that have been carried out, in general, the learning rate value is not very influential in the learning process, both at the momentum values of 0.1, 0.3, 0.5, 0.7, and 0.9. The best desirable accuracy value for momentum 0.1 is at a learning rate of 0.05. The best desirable accuracy value for momentum 0.3 is at a learning rate of 0.09. The most desirable accuracy value for momentum 0.3 is at a learning rate of 0.05 and 0.07. At a learning rate of 0.03 the highest ideal accuracy value is obtained. The best desirable accuracy value for momentum 0.9 is at a learning rate of 0.09. this research should be redone using the number of hidden layers and nodes in each hidden layer. The addition of a hidden layer, as well as variations in the number of nodes in the hidden layer, will affect computation time and yield more optimal accuracy results.

The dangers of health problems in dental disease are common for children and adults. Many dental problems get priority treatment based on data from Riskesdas, about 67.6% of the Indonesian population suffers from dental and oral problems. This affects other parts of the organ that are interrelated. Therefore, this study formulates how to solve the determination of dental disease, by applying the UDB model in machine learning. The purpose of this study was to determine the application of machine learning Binary Decision Tree (BDT) in the classification of classified dental diseases identified by decision trees in determining the results of dental disease predictions including groups and how to solve them. The research methodology in the first stage of data collection was carried out directly with the dental clinic at Cut Meutia Lhokseumawe Hospital. Then input the dental disease data along with the dental disease symptom data. The final stage is dividing the attribute values in viewing the value at a predetermined branch which is then in the form of a decision tree as a reference for the final prediction. The results of the assessment have each value indicating a high level of accuracy, with an accuracy of 92 percent and an inaccuracy of 8 percent of the 40 data points tested. Furthermore, the conclusion of this study can produce an appropriate classification of dental disease and is able to produce accurate results seen from a small error rate

Clustering mahasiswa kedalam keminatan keahlian merupakan salah satu upaya yang perlu dilakukan oleh pihak jurusan untuk menjamin mahasiswa memperoleh pendidikan yang sesuai dengan keahliannya. Saat ini, terdapat banyak metode clustering yang sudah dikembangkan oleh pakar. Umumnya metode clustering mampu mengelompokkan objek-objek yang memiliki tingkat kesamaan ciri yang tinggi, tetapi tidak mampu membatasi jumlah objek yang boleh masuk kedalam suatu kelompok. Kasus klasterisasi mahasiswa kedalam keminatan keahlian merupakan kasus clustering yang membatasi jumlah objek yang boleh masuk kedalam suatu kelompok. Dengan kondisi tersebut, metode clustering yang ada tidak dapat digunakan untuk kasus ini. Peneliti mencoba melihat kasus ini dari sudut pandang optimasi, yaitu bagaimana mengoptimalkan pembentukan kelompok keminatan mahasiswa dengan tingkat ketidaksesuaian bakat yang rendah. Untuk penyelesaian kasus ini, peneliti menggunakan algoritma genetika sebagai metode untuk penyelesaian masalah. Algoritma genetika dibagi kedalam beberapa jenis, yaitu: algoritma genetika dengan prinsip elitisme dan non elitisme, algoritma genetika dengan persentase mutasi 0.01, 0.03 dan 0.05. Berdasarkan penelitian yang dilakukan, diperoleh bahwa algoritma genetika mampu melakukan clustering mahasiswa kedalam keminatan keahlian yang disediakan oleh jurusan. Algoritma genetika dengan prinsip elitisme mampu menemukan solusi optimum yang lebih baik sebesar 39% dibandingkan dengan algoritma genetika non elitisme. Algoritma genetika dengan persentase mutasi 0.05 menghasilkan solusi optimum terbaik, namum memiliki konsumsi waktu yang paling besar dibandingkan dengan persentase 0.01 dan 0.03.