Journal of Advanced Computer Knowledge and Algorithms
JACKA journal published by the Informatics Engineering Program, Faculty of Engineering, Universitas Malikussaleh to accommodate the scientific writings of the ideas or studies related to informatics science. JACKA journal published many related subjects on informatics science such as (but not limited to): Adversarial Machine Learning: Addressing security concerns and developing algorithms robust to adversarial attacks. Anomaly Detection Algorithms: Identifying unusual patterns or outliers in data. Automated Machine Learning (AutoML): Developing algorithms that automate the machine learning model selection and hyperparameter tuning. Automated Planning and Scheduling: Developing algorithms for autonomous decision-making and task scheduling. Bayesian Networks: Utilizing probability theory to model and analyze uncertain systems. Computer Vision: Developing algorithms for image and video analysis, enabling machines to interpret visual information. Constraint Satisfaction Problems (CSP): Designing algorithms to solve problems subject to constraints. Deep Learning: Developing algorithms for neural networks with multiple layers to model complex patterns. Distributed AI Algorithms: Implementing AI algorithms that can work across multiple interconnected devices or nodes. Ensemble Learning: Combining multiple models to improve overall system performance. Evolutionary Algorithms: Utilizing principles of natural selection for optimization and problem-solving. Experiential Learning Algorithms: Designing algorithms that improve performance through experience and learning. Expert Systems: Creating rule-based systems that emulate human expertise in specific domains. Explainable AI (XAI): Developing algorithms that provide transparency and explanations for AI decisions. Fuzzy Logic: Implementing logic that deals with uncertainty and imprecision in decision-making. Genetic Algorithms: Implementing algorithms inspired by genetic evolution for optimization tasks. Knowledge Representation and Reasoning: Creating structures and algorithms to represent and manipulate knowledge. Machine Learning Algorithms: Designing algorithms that enable systems to learn from data and make predictions. Multi-agent Systems: Designing algorithms for systems with multiple interacting agents. Natural Language Processing (NLP): Creating algorithms that understand and process human language. Neuroevolution: Combining evolutionary algorithms with neural networks for optimization. Optimization Algorithms: Developing algorithms focused on improving the performance, efficiency, or decision-making of systems by finding optimal solutions to problems. Pattern Recognition: Developing algorithms to identify patterns within data. Reinforcement Learning: Designing algorithms that learn through trial and error, often applied in decision-making systems. Robotics Algorithms: Designing algorithms for autonomous navigation, manipulation, and decision-making in robots. Semantic Web Technologies: Implementing algorithms for structuring and retrieving information on the web. Sentiment Analysis Algorithms: Analyzing text data to determine sentiment or emotion. Speech Recognition: Developing algorithms to convert spoken language into text. Swarm Intelligence: Developing algorithms based on collective behavior, as seen in swarms or colonies. Time Series Forecasting Algorithms: Predicting future values based on historical data patterns.
Articles
11 Documents
<![CDATA[Independent Campus Student Exchange Sentiment Analysis Using SVM ]]>
<![CDATA[Putri Irhami]]>;
<![CDATA[Eva Darnila]]>;
<![CDATA[Fadlisyah Fadlisyah]]>
<![CDATA[ Journal of Advanced Computer Knowledge and Algorithms Vol 1, No 2 (2024): Journal of Advanced Computer Knowledge and Algorithms - April 2024 ]]>
Publisher : <![CDATA[Department of Informatics, Universitas Malikussaleh ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.29103/jacka.v1i2.14902
<![CDATA[Support Vector Machine (SVM) is a machine learning method that is widely used for regression and classification problems, especially application review classification. Student exchange is one of the programs that universities must prepare. The student exchange program is intended to reduce the problem of disparities in educational facilities and infrastructure in Indonesia. The advantage of student exchange is that they can manage their time, have high awareness in communicating, are able to admit when they experience problems and need help, independent student exchange offers study options of up to 20 credits, both covering Higher Education Recipients courses and activities in the form of the Nusantara Module. Additionally, students are offered the option to register for a maximum of 6 credits of higher education online. The method used in this research is the SVM algorithm, the dataset used consists of 1000 comment reviews with a ratio of 70;30. This research was implemented in a web system using the Python programming language. Of the 300 test data implemented with 700 training data. The Support Vector Machine (SVM) algorithm in classifying review data obtained the highest accuracy in dividing training data & test data 70:30 at 85.00% then precision 28.33%, recall 33.33%]]>
<![CDATA[Cover, Editorial Board, Acknowledgement and Table of Contents ]]>
<![CDATA[JACKA JACKA]]>
<![CDATA[ Journal of Advanced Computer Knowledge and Algorithms Vol 1, No 2 (2024): Journal of Advanced Computer Knowledge and Algorithms - April 2024 ]]>
Publisher : <![CDATA[Department of Informatics, Universitas Malikussaleh ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.29103/jacka.v1i2.15900
<![CDATA[Implementation of the Profile Matching Algorithm to Identify Outstanding Students at Pesantren Modern Misbahul Ulum ]]>
<![CDATA[Moulana Luthfiah]]>;
<![CDATA[Bustami Bustami]]>;
<![CDATA[Fajriana Fajriana]]>
<![CDATA[ Journal of Advanced Computer Knowledge and Algorithms Vol 1, No 1 (2024): Journal of Advanced Computer Knowledge and Algorithms - January 2024 ]]>
Publisher : <![CDATA[Department of Informatics, Universitas Malikussaleh ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.29103/jacka.v1i1.14533
<![CDATA[This research aims to design a decision support system to identify outstanding students at the Misbahul Ulum Modern Islamic Boarding School. Although determining outstanding students is done by taking the average score from all aspects of the criteria, the evaluation process which is still carried out manually takes quite a long time. Therefore, a computerized decision support system is needed to help make decisions more efficiently and accurately. This system was developed using the PHP programming language. The aim of this research is to overcome obstacles in recognizing outstanding students at the Misbahul Ulum Modern Islamic Boarding School by applying the Profile Matching Algorithm. This method can provide the best data by comparing alternative values and predetermined criteria. This research contributes to the development of educational technology and facilitates the introduction of outstanding students at the Misbahul Ulum Modern Islamic Boarding School. By determining the criteria aspects, determining the weight value of each aspect, finding the GAP value, and carrying out a ranking process, this research produces objective results. Involving 252 students from grades 7-11, this research produced a final grade for each student. The calculation results show that Zikril got the highest score, while Asma Biwi got the lowest score.]]>
<![CDATA[Classification of Heart Disease Using Modified K-Nearest Neighbor (MKNN) Method ]]>
<![CDATA[Aulia Azzahra Ma'aruf Lubis]]>;
<![CDATA[Rozzi Kesuma Dinata]]>;
<![CDATA[Hafizh Al Kautsar Aidilof]]>
<![CDATA[ Journal of Advanced Computer Knowledge and Algorithms Vol 1, No 2 (2024): Journal of Advanced Computer Knowledge and Algorithms - April 2024 ]]>
Publisher : <![CDATA[Department of Informatics, Universitas Malikussaleh ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.29103/jacka.v1i2.15702
<![CDATA[Penyakit jantung memiliki banyak jenis dan gejala yang dialami. Penyakit jantung adalah sebuah kondisi ketika organ jantung tidak dapat bekerja sebagaimana fungsinya dengan baik. Jantung adalah organ penting dalam tubuh manusia yang dimana fungsinya adalah memompa darah ke seluruh tubuh. Karena itu dibutuhkannya diagnosa awal untuk pencegahan penyakit jantung dengan memanfaatkan system yang dapat dibuat untuk diagnosa awal pada gejala yang dialami. Yang pada penelitian ini akan menggunakan metode Modified K-Nearest Neighbor (MKNN) dalam mengklasifikasikan penyakit jantung berdasarkan kriteria atau gejala yang ada. Penelitian ini menggunakan 6 kriteria penyakit dan 3 kelas diagnosa penyakit jantung. Dengan melewati beberapa langkah pengerjaan yaitu menghitung jarak Euclidean, menghitung nilai validitas dan terakhir menghitung weight voting dengan mengandalkan nilai K yang telah ditentukan sejak awal perhitungan. Pada penelitian ini telah ditentukan nilai K=5 dan didapat hasil pengujian akurasi sebesar 85%, dengan recall 90% dan precision 85%.]]>
<![CDATA[Implementation of K-NN Algorithm to classify the Scholarship Recipients of Aceh Carong at Universitas Malikussaleh ]]>
<![CDATA[Riski Yanti]]>;
<![CDATA[Sujacka Retno]]>;
<![CDATA[Balqis Yafis]]>
<![CDATA[ Journal of Advanced Computer Knowledge and Algorithms Vol 1, No 1 (2024): Journal of Advanced Computer Knowledge and Algorithms - January 2024 ]]>
Publisher : <![CDATA[Department of Informatics, Universitas Malikussaleh ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.29103/jacka.v1i1.14534
<![CDATA[In an effort to increase the efficiency of the scholarship selection process, this research aims to implement the K-Nearest Neighbors (K-NN) algorithm in the classification of scholarship recipients. The research method involves collecting data on scholarship receipts from several previous years based on predetermined criteria such as father's job, mother's job, parent's income, number of parents working, father's last education, and mother's last education. Next, the K-NN algorithm is applied to classify prospective scholarship recipients based on the similarity of their profiles to students who have received previous scholarships. The results of this research indicate that the implementation of the K-NN algorithm in the classification of scholarship admissions at Malikussaleh Aceh Carong University can increase selection accuracy. The experimental results of the accuracy values obtained show that using the K-Nearest Neighbors algorithm with the Euclidean Distance approach and a value of K = 3 produces an algorithm accuracy level of 87.55%. Thus, the K-NN algorithm can be a useful method for scholarship selectors to support more precise and objective decision making.]]>
<![CDATA[Applying TF-IDF and K-NN for Clickbait Detection in Indonesian Online News Headlines ]]>
<![CDATA[Muhammad Athallah Afif]]>;
<![CDATA[Munirul Ula]]>;
<![CDATA[Lidya Rosnita]]>;
<![CDATA[Rizal Rizal]]>
<![CDATA[ Journal of Advanced Computer Knowledge and Algorithms Vol 1, No 2 (2024): Journal of Advanced Computer Knowledge and Algorithms - April 2024 ]]>
Publisher : <![CDATA[Department of Informatics, Universitas Malikussaleh ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.29103/jacka.v1i2.15810
<![CDATA[This research explores the application of TF-IDF (Term Frequency-Inverse Document Frequency) and K-Nearest Neighbor (K-NN) in constructing a clickbait detection system for Indonesian online news headlines. The TF-IDF method is employed to ascertain the significance of words in news headlines, utilizing a tokenization process to generate numeric representations. The TF-IDF matrix serves as features in the K-NN classification model, with k=1 determining the most similar class. Model evaluation yields outstanding results, achieving accuracy, precision, recall, and F1-Score all reaching 1.0. The confusion matrix unveils no misclassifications, affirming the model's adeptness in correctly classifying all samples.]]>
<![CDATA[Comparison of the Results of the K-Nearest Neighbor (KNN) and Naïve Bayes Methods in the Classification of ISPA Diseases (Case Study: RSUD Fauziah Bireuen) ]]>
<![CDATA[Riska Yolanda Putri]]>;
<![CDATA[Zara Yunizar]]>;
<![CDATA[Safwandi Safwandi]]>
<![CDATA[ Journal of Advanced Computer Knowledge and Algorithms Vol 1, No 1 (2024): Journal of Advanced Computer Knowledge and Algorithms - January 2024 ]]>
Publisher : <![CDATA[Department of Informatics, Universitas Malikussaleh ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.29103/jacka.v1i1.14535
<![CDATA[Acute Respiratory Infection or commonly called (ARI) is a disease caused by bacteria or viruses. (ARI) can attack all ages, especially children. This study aims to compare the accuracy of classification in (ARI) disease. The data used is data from patients affected by (ARI) disease at Fauziah Bireuen Hospital. K-Nearest Neighbors and Naïve Bayes can be used in the classification of (ARI) diseases. Measurement of accuracy using Confusion Matrix in the K-Nearest Neighbors method with the Eulidean Distance approach in the case of (ARI) disease classification obtained a percentage of precision of 91%, recall 84% and accuracy of 88%. While the Naïve Bayes method obtained a percentage of precision of 95%, recall 78% and accuracy of 86%. The results of the accuracy comparison of the two methods show that the K-Nearest Neighbors method has a higher accuracy rate than the Naïve Bayes method.]]>
<![CDATA[K-NN with Purity Algorithm to Enhance the Classification of the Air Quality Dataset ]]>
<![CDATA[Sujacka Retno]]>;
<![CDATA[Novia Hasdyna]]>;
<![CDATA[Balqis Yafis]]>
<![CDATA[ Journal of Advanced Computer Knowledge and Algorithms Vol 1, No 2 (2024): Journal of Advanced Computer Knowledge and Algorithms - April 2024 ]]>
Publisher : <![CDATA[Department of Informatics, Universitas Malikussaleh ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.29103/jacka.v1i2.15890
<![CDATA[The large number of attributes in a large dataset can cause a decrease in the level of classification accuracy. Attribute reduction can be a solution to improve classification performance, especially in the K-NN algorithm. This research discusses the classification results of K-NN with attribute reduction using Purity. Based on the results of testing carried out on the Air Quality Dataset, the level of accuracy obtained after attribute reduction was 70.71%, while the level of accuracy obtained before attribute reduction was 56.44%, the increase in accuracy obtained from testing this dataset was equal to 14.27%. The proposed Purity method for attribute reduction can increase the accuracy level of the K-NN classification process.]]>
<![CDATA[Classification of Receiving Electricity Subsidy Assistance in Blang Panyang Village Using the K-NN (K-Nearest Neighbor) Method ]]>
<![CDATA[Miftahul Jannah]]>;
<![CDATA[Cut Syahira Salsabila]]>;
<![CDATA[Nur Faiza]]>;
<![CDATA[Mutasar Mutasar]]>
<![CDATA[ Journal of Advanced Computer Knowledge and Algorithms Vol 1, No 1 (2024): Journal of Advanced Computer Knowledge and Algorithms - January 2024 ]]>
Publisher : <![CDATA[Department of Informatics, Universitas Malikussaleh ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.29103/jacka.v1i1.14530
<![CDATA[The electricity subsidy program is one of the poverty reduction programs by providing electricity subsidy assistance funds to poor and disadvantaged households paid by the Government of Indonesia to PT PLN (Persero). The government implements a targeted electricity subsidy policy, the subsidy must be truly enjoyed by the poor. The purpose of this research is to test the K-Nearest Neighbors algorithm in predicting the receipt of electricity subsidy assistance. In the dataset of beneficiaries used in this study, there are 45 records or tuples with four attributes (house condition, income, occupation and number of amperes). The prediction of new data categories is done by using the manual calculation stage of Euclidean Distance from three different K values. The results show that with K=15, K=30 and K=45 the new data (46) has an "Ineligible" category with an accuracy rate of 100%. Then with K=45, K=30 and K=45 the new data (D46) has a "Viable" category with an accuracy rate of 66.6%.]]>
<![CDATA[Cover, Editorial Board, Acknowledgement and Table of Contents ]]>
<![CDATA[JACKA JACKA]]>
<![CDATA[ Journal of Advanced Computer Knowledge and Algorithms Vol 1, No 1 (2024): Journal of Advanced Computer Knowledge and Algorithms - January 2024 ]]>
Publisher : <![CDATA[Department of Informatics, Universitas Malikussaleh ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.29103/jacka.v1i1.14574
