Journal of Knowledge Engineering and Artificial Intelligence

Application of the Electre Method for Predicting Student Graduation

Ricky Novriansyah — 2022-09-27

The use of student data cannot be utilized optimally and efficiently, the university still has difficulty in analyzing data and evaluating student graduation rates. This is because there is no integration in a good database. The University needs to evaluate the graduation rate to improve and maintain and even improve the quality of graduates and accreditation of the University or Study program, so that it is necessary to maximize student graduation data and information as a basis for data sources to become a strategic information system for the University or Study Program. The ELECTRE method is one of the effective methods for MADM with qualitative and quantitative features. So the development of this method is to improve the ability to make such decisions. The basic concept of the ELECTRE method is to deal with outranking relationships by using paired comparisons between alternatives under each criterion separately. The application of the ELECTRE method in determining the prediction of the study period and the graduation predicate of students is able to produce predictive information on student graduation and is able to overcome existing problems, the prediction results of student graduation are very effective in helping universities predict student graduation data. The results of student graduation prediction testing using function testing of the system using blackbox testing get a value of 100% in accordance with the prediction application function that has been made.

Naive Bayes Method in Determining Diagnosis of Corn Plant Disease

Adhe Kurniawan — 2022-09-27

Corn is one of the leading agricultural commodities from the food crop sub-sector that is multi-purpose and has strategic value to be developed. At this time, corn is not only used for food but also for animal feed, and also fuel. The amount of production, productivity, and price of corn always fluctuates due to the influence of the ever-changing amount of demand and supply. The high demand for maize in the domestic market is an opportunity for Indonesia to balance the demand for and supply of maize. An expert system is a system that seeks to adopt human knowledge to computers so that computers can solve problems as is usually done by experts. A good expert system is designed to be able to solve a particular problem by imitating the work of experts. The naive Bayes method is a method used to predict probability. While the Bayes classification is a statistical classification that can predict the class of a probability member. For a simple Bayesian classification known as the Naïve Bayesian Classifier, it can be assumed that the effect of an attribute value of a given class is independent of other attributes. From the results of testing on the Expert System Application that was built, the expert system can solve the problem that it can display the diagnostic results quickly and precisely based on the symptoms entered by the user. Based on the tests carried out with the accuracy of the diagnosis obtained from the comparison between the results of expert diagnoses and system diagnoses with a percentage of 90%, the system can be clarified that it is feasible to use.

Implementation of Analytical Hierarchy Process in Selection of Prospective Nurses

Yuli Ismail — 2022-09-27

Nurses are the spearhead of whether or not health services are provided to patients because for 24 hours nurses always interact with patients. Good service is inseparable from the commitment of nurses to provide good service to patients. The decision support system (DSS) is the application of an information system that is intended only as a management tool in decision making. The decision support system is designed to produce a variety of alternatives offered to decision makers in carrying out their duties. AHP was developed to compile a problem into a hierarchy which is then weighted (determining priorities) based on the perception of decision makers to choose the best decision. In decision making, the thing that needs to be considered is at the time of data collection, where this data is expected to be close to the real value, for example, the degree of consumer interest can be done with a pairwise comparison approach. The process of recruiting prospective nurses depends on the number of nurses needed. If the nurses needed are 5 people, then prospective nurses named Nurse 1, Nurse 2, Nurse 3, Nurse 4, and Nurse 5 have a greater chance of being accepted as nurses.

Creditworthiness System Using Analytical Hierarchy Process Method

Sondang Nadeak — 2022-09-27

The role of savings and loans cooperatives is very important, so it is not surprising that cooperatives are always found where, especially in urban areas, even in remote areas, we can find cooperatives. Savings and loan cooperatives obtain capital from principal deposits and mandatory deposits of cooperative members. This makes owners sometimes cut off credit without paying attention to aspects of credit risk that should be consistently applied. The importance of quality of service to cooperatives greatly affects the progress of cooperative members, one of which is the speed and accuracy in making decisions to provide credit for Members or prospective borrowers. AHP is appropriate for use in decision making that involves comparing decision elements that are difficult to assess quantitatively. This is based on the assumption that the natural reaction of the human being when facing complex decision-making is to group the elements of the decision according to its characteristics in general. Based on the results of interviews with cooperatives, there are 5 criteria needed in determining the feasibility of providing credit, namely Character, Capacity, Capital, Condition of economy, and Collateral. The calculation results use AHP for each creditor, namely A2 creditors with a final value of 0.22785, A5 creditors with a final value of 0.21926, A3 creditors with a final value of 0.20459, A1 creditors with an end value of 0.1772, A4 creditors with an end value of 0.17111.

Selection of Scholarships for Underprivileged Students Using the TOPSIS

Iman Santoso — 2022-09-27

Scholarships that will be given to someone must have conditions, including, the scholarship recipient is a person who is less well off economically or financially, the prospective scholarship recipient has achievements, especially in the field of interest or has a contribution to social activities, but it all depends on the institution involved. give it. Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) with the concept that the best chosen alternative not only has the shortest distance from the positive ideal solution, but also has the longest distance from the negative ideal solution. The results of the calculation of the selection of scholarships for underprivileged students using the Technique for Order of Preference by Similarity to Ideal Solution method are obtained for rank 1 with a preference value of 0.7749 is Cassandra student, rank 2 with a preference value of 0.7587 is John's student, rank 3 with a preference value of 0.5569 is Gonzales students, rank 4 with a preference value of 0. 4788 is Maria's student, ranked 5th with a preference value of 0. 4547 is Moriera's student.

Predicting Student Preparation for Exams Using the Naive Bayes Method

Irkhas Saputra — 2022-09-27

The National Examination (UN) is a government policy in the field of education to determine the quality standards of educatioN. The function of the national examination is important to measure student competence and one of the considerations for selection to a higher level. The Naive Bayes algorithm is mostly used in spam message filtering, sentiment analysis, and recommendation systems. One of the main reasons for the use of this algorithm is due to its quick and easy implementation. The test was carried out using the WEKA application, using the confusion matrix method to determine the effectiveness of classification with model equations and can be calculated to find accuracy, sensitivity, specificity, positive predictive value (ppv), and negative predictive value (npv). The results of the evaluation and validation of the confusion matrix using training data and testing data showed the accuracy rate and error rate in the Naïve Bayes algorithm of 96.8254% and 3.1746%.

Determination of Exemplary Employees Using Simple Additive Weighting

Joko Setia — 2022-09-27

A human resource is someone who is willing, and able to contribute to the business in order to achieve the goals of the organization or company. One of the most important factors of a company is human resources. The SAW (Simple Additive Weighting) method is also used to find the optimal alternative from a number of alternatives with certain criteria. The essence of SAW is to determine the weight value for each attribute, then proceed with a ranking process that will select the alternatives that have been given. The variables used in this study are attendance data, responsibility, social interaction, performance, personality, work discipline in each division where the employee works. The results obtained information that of the 24 employees who deserve to be exemplary employees are 23 employees with the highest score. So that the employee is declared worthy as an exemplary employee with a score of 75.