International Journal of Artificial Intelligence Research

Journal Information
EISSN : 2579-7298
Published by: STMIK Dharma Wacana (10.29099)
Total articles ≅ 56
Current Coverage
DOAJ
Filter:

Latest articles in this journal

Basiroh Basiroh, Shahab Wahhab Kareem
International Journal of Artificial Intelligence Research, Volume 5; doi:10.29099/ijair.v5i1.203

Abstract:
Nowadays technological developments are increasingly having a positive influence on the development of human life, including in the health sector. One of them is an expert system that can transfer an expert's knowledge into a computer application to simplify and speed up the diagnosis of a disorder or disease in humans. The purpose of this final project is to design an application to diagnose diseases that occur during pregnancy which is caused by the existence of these pregnancies to simplify and speed up the diagnosis of diseases experienced by pregnant women. This study uses the forward chaining method. By involving experts in this expert system analysis according to current needs. Users are given easy access to information on several types of pregnancy disorders and their symptoms, as well as consultation through several questions that the user must answer to find out the results of the diagnosis. While experts are facilitated in system management, both the process of adding, updating and, deleting data.
Mustofa Kamil, Ankur Singh Bist, Untung Rahardja, Nuke Puji Lestari Santoso, Muhammad Iqbal
International Journal of Artificial Intelligence Research, Volume 5; doi:10.29099/ijair.v5i1.173

Abstract:
The current situation of the Covid-19 pandemic is currently increasing public concern about the community. The government has especially recommended Stay at Home and the implementation of PSBB in various regions. One of the concerns is when the election of regional leaders to the general chairman. Even though there is already a safeguard regulation, this is not considered safe in the current Covid-19 pandemic. The solution in this research is the use of a blockchain-based E-voting system to help tackle election unrest during Covid-19. Where e-voting with blockchain technology can be carried out anywhere through the device without the need to be present in the voting booth, reducing data fraud, accurate and decentralized voting results that can be accessed by the public in real-time. The use of cryptographic protocols is applied for data transfer between system components as well as valid system security. This research method uses SUS trial analysis in a significant system of the Covid-19 pandemic situation. The implication that the SUS Score analysis shows 90 shows an acceptable E-voting system, meaning that the community can accept it because it brings positive and significant impacts such as effectiveness and efficiency.
, Yonatan Adiwinata, Desy Purnami Singgih Putri, Ni Putu Sutramiani
International Journal of Artificial Intelligence Research, Volume 5; doi:10.29099/ijair.v5i1.187

Abstract:
One of the main tasks of a lecturer is to give students an academic assessment in the learning process. The assessment process begins with reading or checking the answers of student assignments that contain a combination of very long sentences such as essay or report assignments. This certainly takes a lot of time to get the primary information contained therein. It is necessary to summarize the answers so that the lecturer does not need to read the whole document but is still able to take the essence of the response to the task. This study proposes the application of summarizing text documents of student essay assignments automatically using the Fuzzy C-Means method with the sentence weighting feature. The sentence weighting feature is used by selecting the sentence with the highest weight in one cluster, helping the system to get the primary information from a document quickly. The results of this study indicate that the system succeeds in summarizing text with an average evaluation of the values of precision, recall, accuracy, and F-measure of 0.52, 0.54, 0.70, and 0.52, respectively.One of the main tasks of a lecturer is to give students an academic assessment in the learning process. The assessment process begins with reading or checking the answers of student assignments that contain a combination of very long sentences such as essay or report assignments. This certainly takes a lot of time to get the primary information contained therein. It is necessary to summarize the answers so that the lecturer does not need to read the whole document but is still able to take the essence of the response to the task. This study proposes the application of summarizing text documents of student essay assignments automatically using the Fuzzy C-Means method with the sentence weighting feature. The sentence weighting feature is used by selecting the sentence with the highest weight in one cluster, helping the system to get the primary information from a document quickly. The results of this study indicate that the system succeeds in summarizing text with an average evaluation of the values of precision, recall, accuracy, and F-measure of 0.52, 0.54, 0.70, and 0.52, respectively.
Davin Wijaya, Jumri Habbeyb Ds, Samuelta Barus, Beriman Pasaribu, Loredana Ioana Sirbu,
International Journal of Artificial Intelligence Research, Volume 5; doi:10.29099/ijair.v4i2.169

Abstract:
Employee turnover is the loss of talent in the workforce that can be costly for a company. Uplift modeling is one of the prescriptive methods in machine learning models that not only predict an outcome but also prescribe a solution. Recent studies are focusing on the conventional predictive models to predict employee turnover rather than uplift modeling. In this research, we analyze whether the uplifting model has better performance than the conventional predictive model in solving employee turnover. Performance comparison between the two methods was carried out by experimentation using two synthetic datasets and one real dataset. The results show that despite the conventional predictive model yields an average prediction accuracy of 84%; it only yields a success rate of 50% to target the right employee with a retention program on the three datasets. By contrast, the uplift model only yields an average accuracy of 67% but yields a consistent success rate of 100% in targeting the right employee with a retention program.
Untari Novia Wisesty, Febryanti Sthevanie, Rita Rismala
International Journal of Artificial Intelligence Research, Volume 4, pp 127-134; doi:10.29099/ijair.v4i2.188

Abstract:
Early detection of cancer can increase the success of treatment in patients with cancer. In the latest research, cancer can be detected through DNA Microarrays. Someone who suffers from cancer will experience changes in the value of certain gene expression. In previous studies, the Genetic Algorithm as a feature selection method and the Momentum Backpropagation algorithm as a classification method provide a fairly high classification performance, but the Momentum Backpropagation algorithm still has a low convergence rate because the learning rate used is still static. The low convergence rate makes the training process need more time to converge. Therefore, in this research an optimization of the Momentum Backpropagation algorithm is done by adding an adaptive learning rate scheme. The proposed scheme is proven to reduce the number of epochs needed in the training process from 390 epochs to 76 epochs compared to the Momentum Backpropagation algorithm. The proposed scheme can gain high accuracy of 90.51% for Colon Tumor data, and 100% for Leukemia, Lung Cancer, and Ovarian Cancer data.
Ika Sari Damayanthi Sebayang, Muhammad Fahmia
International Journal of Artificial Intelligence Research, Volume 4, pp 75-85; doi:10.29099/ijair.v4i2.174

Abstract:
To determine the amount of dependable flow, a hydrological approach is needed where changes in rainfall become runoff. This diversification is a very complex hydrological phenomenon. Where this is a nonlinear process, with time changing and distributed separately. To approach this phenomenon, an analysis of the hydrological system has been developed using a model which is a simplification of the actual natural variables. The model is formed by a set of mathematical equations that reflect the behavior of parameters in hydrology. Modeling in this case uses artificial neural networks, multilayer perceptron combined with the backpropagation method is used to study the rainfall-runoff relationship and verify the model statistically based on the mean square error (MSE), Nash-Sutcliffe Efficiency (NSE) and correlation coefficient value (R2). Of the three models formed, model 3 provides optimum results with correlation levels using NSE per month as follows, in Cikapundung Sub-Basin NSE = 0,990703, R2 = 0,995008, and MSE = 0,00014443, while in Citarik Sub-Basin NSE = 0.9500, R2 = 0.97592, and MSE = 0.0010804 . From these results it can be seen that ANN has a fairly good ability to replicate random discharge fluctuations in the form of artificial models that have almost the same fluctuations and can also be applied in rainfall runoff modelization even though the results of the test results are not very accurate because there are still irregularities
Mardhiya Hayaty, Siti Muthmainah, Syed Muhammad Ghufran
International Journal of Artificial Intelligence Research, Volume 4, pp 86-94; doi:10.29099/ijair.v4i2.152

Abstract:
High accuracy value is one of the parameters of the success of classification in predicting classes. The higher the value, the more correct the class prediction. One way to improve accuracy is dataset has a balanced class composition. It is complicated to ensure the dataset has a stable class, especially in rare cases. This study used a blood donor dataset; the classification process predicts donors are feasible and not feasible; in this case, the reward ratio is quite high. This work aims to increase the number of minority class data randomly and synthetically so that the amount of data in both classes is balanced. The application of SOS and ROS succeeded in increasing the accuracy of inappropriate class recognition from 12% to 100% in the KNN algorithm. In contrast, the naïve Bayes algorithm did not experience an increase before and after the balancing process, which was 89%.
Adhi Prahara, Son Ali Akbar, Ahmad Azhari
International Journal of Artificial Intelligence Research, Volume 4, pp 107-116; doi:10.29099/ijair.v4i2.179

Abstract:
Road defect such as potholes and road cracks, became a problem that arose every year in Indonesia. It could endanger drivers and damage the vehicles. It also obstructed the goods distribution via land transportation that had major impact to the economy. To handle this problem, the government released an online complaints system that utilized information system and GPS technology. To follow up the complaints especially road defect problem, a survey was conducted to assess the damage. Manual survey became less effective for large road area and might disturb the traffic. Therefore, we used road aerial imagery captured by Unmanned Aerial Vehicle (UAV). The proposed method used texton combined with K-Nearest Neighbor (K-NN) to segment the road area and Support Vector Machine (SVM) to detect the road defect. Morphological operation followed by blob analysis was performed to locate, measure, and determine the type of defect. The experiment showed that the proposed method able to segment the road area and detect road defect from aerial imagery with good Boundary F1 score.
Ahmad Chusyairi, Pelsri Ramadar Noor Saputra
International Journal of Artificial Intelligence Research, Volume 5; doi:10.29099/ijair.v5i1.191

Abstract:
In Indonesia, public health services at the city or district level are carried out by regional public hospitals or “puskesmas” (health care centers), especially in Banyuwangi regency, East Java, Indonesia that has 45 health care centers spread throughout the villages. This research focused on the deaths of babies caused by diarrhea diseases, which are the second leading cause of death among children younger than 5 years globally. All of the health care centers need to be divided into 3 groups to find out which health care centers have the least, most moderate, and many diarrhea sufferers. Fuzzy C-Means algorithm is used to overcome this problem. The result from this research shown that 2 health care centers have the smallest member of diarrhea sufferers, 14 health care centers have a medium member of diarrhea sufferers, and the rest have a large number of diarrhea sufferers. From the result of this study, it can be a reference for the health department center in dealing with diarrheal diseases, accordingly, the infant mortality rate due to diarrheal diseases can be lowered to health care centers that have high diarrhea sufferers.
Ika Candradewi, Agus Harjoko, Bakhtiar Alldino Ardi Sumbodo
International Journal of Artificial Intelligence Research, Volume 5; doi:10.29099/ijair.v5i1.201

Abstract:
In the automation of vehicle traffic monitoring system, information about the type of vehicle, it is essential because used in the process of further analysis as management of traffic control lights. Currently, calculation of the number of vehicles is still done manually. Computer vision applied to traffic monitoring systems could present data more complete and update.In this study consists of three main stages, namely Classification, Feature Extraction, and Detection. At stage vehicle classification used multi-class SVM method to evaluate characteristics of the object into eight classes (LV-TK, LV-Mobil, LV-Mikrobis, MHV-TS, MHV-BS, HV-LB, HV- LT, MC). Features are obtained from the detection object, processed on the feature extraction stage to get features of geometry, HOG, and LBP in the detection stage of the vehicle used MOG method combined with HOG-SVM to get an object in the form of a moving vehicle and does not move. SVM had the advantage of detail and based statistical computing. Geometry, HOG, and LBP characterize complex and represents an object in the form of the gradient and local histogram.The test results demonstrate the accuracy of the calculation of the number of vehicles at the stage of vehicle detection is 92%, with the parameters HOG cellSize 4x4, 2x2 block size, the son of vehicle classification 9. The test results give the overall mean recognition rate 91,31 %, mean precision rate 77,32 %, and mean recall rate 75,66 %.
Back to Top Top