No abstract

No abstract

Corn is one of the staple foods consumed by many people after rice plants, especially in Indonesia. High consumer demand requires corn production in large quantities to meet these needs. However, corn production is not always in large quantities due to several factors, namely diseases in corn plants. Unhealthy corn plants can reduce the amount of production. Healthy and unhealthy corn plants can be identified manually, but this method is not efficient, so in this study, it is proposed to classify corn diseases using the Random Forest, Neural Network, and Nave Bayes methods. The dataset used is a collection of corn leaf images taken from farmers’ fields in the Madura Region with four target classes, namely healthy, gray leaf spot, blight, and common rust. Based on the test results, the classification using the Neural Network method provides a better accuracy value than the other two methods in classifying corn leaf datasets, namely the AUC value reaches 90.09%, classification accuracy is 74.44%, f1 score is 72.01%, precision is 74.14% and recall by 74.43%.

Salt is one of Indonesia’s major commodities. However, the quality of industrial salt in Indonesia is still an obstacle, so the need for industrial salt still relies on imported salt, especially from Australia. Quality improvement is done through purification using the recrystallization method. The use of a method that is still simple results in the salt being produced still has an as-is quality. Quality is shown from the appearance of salt physically and chemically. Good salt is shown by the crystal form which is smooth and has clear white color. Therefore, good knowledge of salt quality must be known early, in addition to being able to meet the Indonesian National Standard (SNI),in this way salt farmers will more easily improve the quality of salt produced and can differentiate salt designation based on its quality category. This study takes the theme of how to make decisions to determine the quality of salt, so that a decision support system will be built to assist in determining good salt quality by using the Simple Additive Weighting (SAW) method. This method can support the decision making of salt quality determination based on the weight of each attribute. Morever, the total score of the end result can produce a good alternative decision in accordance with specified criteria, so that it will produce salt quality.

Thesis topic is an inseparable part in the world of tertiary education. Determining the thesis topic becomes a problem for students. The determination of the thesis topic leads to the trend of the topic in the development of computer science. The determination of the topic of thesis for students often ignores their ability to process. Ideally in determining the topic of the thesis, the record of student grades can be an important variable in deciding topics for students, where the student’s grade record is contained in the transcript. Therefore, this study uses the Support Vector Machine (SVM) method in recommending thesis topics by classifying selected subject groups that have been taken by students. The Support Vector Machine method is a classification method of supervision because it requires testing data and training data as a training process at the time of prediction. Support Vector Machine provides an optimal model, which provides a solution with a maximum margin to determine the distance of data to the hyperplane. The test results show an accuracy of 80%.