1,721,018 research outputs found
Intelligent mining multi dimensional association rules from large inconsistent databases
Full text linkPrediction of Graduation Accuracy Using the K-Means Clustering Algorithm and Classification Decision Tree
No full textBecoming a scholar at the right time for students is a very meaningful award for them if it is supported by seriousness and perseverance in their studies. Here, sample data was taken from 131 randomly taken in testing. Where there are still students who are not detected by the study program in completing their lectures, so research is carried out on clustering and classification with decision trees in determining the level of accuracy of lectures by clustering data, determining the initial centroid value and the centroid point. The results found were that there were 78 people grouped in cluster 0 and 53 people grouped in cluster 1, where those with potential for punctuality for their studies were in cluster 0 so they were students who could finish within the specified time. Meanwhile, students grouped in cluster 1 illustrate that these students need coaching and guidance both in the study program and with their supervisors. In the classification taken from the results of data clustering, two classes were obtained, namely class a and class b, with 73 and 58 data respectively, so that the results between clustering and classification did not differ too much in the data to predict the accuracy of a student's graduation
Analyzing the use of Social Media by Fashion Designers with K-Means and C45
Full text linkSocial media is one part of digital marketing that is used for the development of marketing business products known as social-marketing. The use of social media as social marketing is still managed conventionally and has not implemented business social media. This study was conducted to analyze the clusters and classifications of the use of social media by fashion designers in West Sumatra in marketing their products. This analysis uses the k-Means algorithm and c45 uses the Rapidminer application for the fashion designer industry in West Sumatra. Data is collected from Instagram and Facebook of fashion designers. The data analyzed by K-Means resulted in 3 clusters of social media use, namely 3 less active clusters, 12 active clusters and 1 very active, then classification using the C45 method resulted in a decision tree that described the most and the least in using social media. This study resulted in grouping and classifying variables from whether or not the use of social media in social marketing for the fashion designer industry players in West Sumatra was good or not. The results of this study can be used as a reference for developing integrated marketing for West Sumatra fashion designers
Development of the Rough Set Method to Determine Lecturer Scholarship Opportunities
No full textCurrently, all groups can experience the development of artificial intelligence, this happens because artificial intelligence has experienced very significant changes. Artificial Intelligence (AI) consists of several branches, one of which is machine learning. Machine Learning (ML) technology is a branch of AI that is very interesting because it is a machine that can learn like humans. The method used here is the rough set method. In this research, a case will be raised to determine scholarship opportunities for lecturers based on predetermined criteria. To solve the problem above, machine learning was used using the Rough Set method, using Rosetta software. By the regulations determined by the scholarship provider, in this case, the institution concerned where the lecturer is registered as teaching staff to obtain a scholarship, criteria are needed to determine who will be selected to receive the scholarship. The distribution of scholarships is carried out to improve lecturer performance, as an achievement as well as an appreciation for the lecturer concerned for his long service to the institution
Accurately Determining Labor Test Results Using the Rough Set Method
No full textAn exam is something that must be done to test a person's ability or intelligence. The laboratory exam in the Computer Systems study program at Putra Indonesia University "YPTK" Padang consists of a digital systems exam, a fuzzy logic control exam, and a tool presentation. The Labor Exam must be passed by students who will take the comprehensive exam. In this study, laboratory exam data was taken for 20 students. So far, processing of student laboratory exam results has been done manually so it takes a long time to make decisions. To overcome this problem, a Rough Set method is used to determine laboratory test results. The Rough Set method is part of machine learning. This research produces 29 rules as knowledge, namely {Digital System} Or {A} = 3 rules, {Fuzzy Logic} Or {B} = 3 rules, {Tool Presentation} Or {C} = 3 rules, {Fuzzy Logic, Tool Percentage} Or {BC} = 6 rules, {Digital System, Fuzzy Logic} Or {AB} = 6 rules and {Digital System, Tool Percentage} Or {AC} = 8 rules. The Rough Set method can determine student laboratory exam results (pass or fail) accurately
Rought Set: Effective Method for Determining Scholarship Recipients
No full textEvery year, higher education institutions receive a KIP Tuition scholarship quota that has been determined by Ristek Dikti through LLDIKTI which is given during the new student admissions process. The process of determining recipients is carried out manually resulting in inaccurate scholarship recipients being selected and the selection results may not be the same based on those who participated in making the decision. This research is motivated by the need for an algorithm for determining prospective scholarship recipients that is appropriate and effective because the recipient selection process often takes a long time because many high school and equivalent students register so that they exceed the quota limit while the quota given is limited. This research aims to use a system for scholarship recipients and provide rules and knowledge, namely rough set Theory and adapted to the Rosetta application, using prospective student data during the selection process for new students who apply for the KIP Kuliah scholarship in the 2020/2021 academic year. The resulting decision is the KIP Opportunity which consists of 4 (four) attributes, including parents' income, housing status, dependents, and parental status. The results of this research using sample data from 12 people produced 6 (six) rules and knowledge of 26 rules. This research is very supportive in identifying the eligibility of KIP Kuliah recipients
Metode k-means clustering untuk mengukur tingkat kedisiplinan pegawai (studi kasus di pemerintah kabupaten padang pariaman)
Full text linkKnowledge Discovery In Database (KDD) is a process of converting raw data into useful data in the form of information. Data mining is a technique of digging up hidden or hidden valuable information in a very large data collection (database) so that an interesting pattern is found that was previously unknown. Clustering is a method in data mining in which data objects that have similarities or the same characteristics are grouped into one group and those that are different are grouped into another group. One aspect of discipline that can be used to evaluate employee performance is attendance. The k-means method is used to classify employee discipline levels and then describes the values that have been obtained to generate new knowledge regarding data patterns on employee discipline levels. The attendance data is clustered into 3, namely to measure low, medium, and high levels of discipline. After carrying out the calculation process, the 41 employee samples produced 3 iterations, and the final result was 3 clustering, namely cluster 1 of 10 employees with low discipline, cluster 2 of 7 employees with moderate discipline, and cluster 3 of 24 employees with high discipline. This is intended so that leaders can find out which employees have high, medium and low levels of discipline so that they can provide appreciation or rewards and sanctions in order to maintain and improve their discipline so that service to the community can be optimal and the vision and mission of the local government can be achieved.
Keywords: KDD, Data Mining, K-Means Clustering Method, DisciplineKnowledge Discovery In Database (KDD) merupakan sebuah proses merubah data mentah menjadi data yang bermanfaat berbentuk informasi. Data mining adalah suatu teknik menggali informasi berharga yang terpendam atau tersembunyi pada suatu koleksi data (database) yang sangat besar sehingga ditemukan suatu pola yang menarik yang sebelumnya tidak diketahui. Clustering adalah salah satu metode dalam data mining yang dimana objek data yang mempunyai kemiripan atau karakteristik yang sama akan dikelompokan menjadi satu kelompok dan yang berbeda di kelompokkan pada kelompok yang lainnya. Salah satu aspek kedisiplinan yang dapat digunakan untuk mengevaluasi kinerja pegawai adalah dengan kehadiran. Metode k-means digunakan untuk mengelompokan tingkat kedisiplinan pegawai kemudian mendeskripsikan nilai-nilai yang sudah didapatkan untuk menghasilkan sebuah knowledge baru mengenai pola data tingkat kedisplinan pegawai. Data absensi di clustering menjadi 3 yaitu untuk mengukur tingkat kedisiplinan rendah, sedang, dan tinggi. proses perhitungan dari 41 sampel pegawai menghasilkan 3 kali iterasi, dan di dapatkan hasil akhir 3 cluster yaitu cluster 1 sebanyak 10 pegawai dengan kedisiplinan rendah, cluster 2 sebanyak 7 pegawai dengan kedisiplinan sedang, dan cluster 3 sebanyak 24 pegawai dengan kedisiplinan tinggi, Penelitian ini bertujuan agar pimpinan dapat mengetahui pegawai mana yang memiliki tingkat kedisiplinan tinggi, sedang dan rendah sehingga dapat memberikan apresiasi atau penghargaan dan sanksi agar dapat mempertahankan dan meningkatkan kedisiplinannya sehingga pelayanan kepada masyarakat bisa optimal dan visi misi pemerintah daerah bisa tercapai.
Kata kunci: KDD, Data Mining, Metode K-Means Clustering, Kedisiplina
Texture and Flag Color Extraction in Backpropagation Neural Network Architecture
No full textA flag is a rectangular or triangular piece of cloth or paper used as a symbol of the state, association, body, etc. or as a sign. It is often also used to symbolize a country to show its sovereignty. Along with the large number of countries, the country's flag also has many varieties and colors. The use of computers as a human aid is expected to the extent that the computer's ability can replace the limitations that humans have. Humans can recognize an object by using their eyes and brain, but if the eyes and brain cannot work properly it will hamper human work. In this research, training will be conducted on the Back Propagation Neural Network Architecture. Characteristic data for image recognition is obtained by extracting texture features and RGB color features. So that the network can recognize the flags by matching the feature data obtained from the training carried out. Characteristic data obtained from 24 data consisting of 16 training images and 8 testing images. From the results of the image network training can be identified properly, the accuracy rate of object identification is 87.50%. GUI users are able to identify flag images based on RGB text and color features
Development of a Hybrid CNN-BiLSTM Architecture to Enhance Text Classification Accuracy
Full text linkIntroduction: Natural Language Processing (NLP) has experienced significant advancements to address the growing demand for efficient and accurate text classification. Despite numerous methodologies, achieving a balance between high accuracy and model stability remains a critical challenge. This research aims to explore the implementation of a hybrid architecture integrating Convolutional Neural Networks (CNN) and Bidirectional Long Short-Term Memory (BiLSTM) with FastText embeddings, targeting effective text classification.Methods: The proposed hybrid architecture combines the CNN\u27s ability to capture local patterns and BiLSTM\u27s temporal feature extraction capabilities, enhanced by FastText embeddings for richer word representation. Regulatory mechanisms such as Dropout and Early Stopping were employed to mitigate overfitting. Comparative experiments were conducted to evaluate the performance of the model with and without Early Stopping.Results: The experimental findings reveal that the model without Early Stopping achieved a remarkable accuracy of 99%, albeit with a higher susceptibility to overfitting. Conversely, the implementation of Early Stopping resulted in a stable accuracy of 73%, demonstrating enhanced generalization capabilities while preventing overfitting. The inclusion of Dropout further contributed to model regularization and stability.Conclusions: This study underscores the significance of balancing accuracy and stability in deep learning models for text classification. The proposed hybrid architecture effectively combines the strengths of CNN, BiLSTM, and FastText embeddings, providing valuable insights into the trade-offs between achieving high accuracy and ensuring robust generalization. Future work could further explore optimization techniques and datasets for broader applicability
Sistem Pendukung Keputusan Kelayakan Penerima Kartu Indonesia Pintar Kuliah Menggunakan Metode SAW
Full text linkSistem Informasi Manajemen (SIM) sendiri adalah sebuah sistem formal dan informal yang menyajikan informasi mengenai sejarah, situasi saat ini, dan proyeksi masa depan melalui komunikasi lisan dan tulisan, terkait dengan berbagai operasi perusahaan dan lingkungan di sekitarnya. Selain itu, Sistem Penunjang Keputusan (Decision Support System) menjadi komponen penting dalam mendukung pengambilan keputusan untuk menyeleksi penerima beasiswa Kip Kuliah pengelola yayasan Universitas Dehasen Bengkulu memerlukan pendekatan yang lebih sistematis. tujuan penelitian untuk mengembangkan sistem pendukung keputusan membantu yayasan dalam proses seleksi penerima beasiswa. mempertimbangkan kriteria-kriteria tertentu, sistem diharapkan memberikan rekomendasi yang lebih akurat, sehingga proses seleksi dapat berjalan. Manfaat dari penelitian ini membantu pengelola mengambil keputusan lebih tepat. Metode SAW terdiri dari penilaian atribut setiap alternatif dan direpresentasikan dalam matriks penilaian keputusan. Matriks digunakan untuk menentukan seluruh kriteria dan skor dari setiap alternatif. Metode SAW memerlukan normalisasi matriks keputusan (X) untuk dibandingkan dengan peringkat alternatif yang ada. Metode SAW atribut kriteria ke-untungan (benefit) dan kriteria biaya (cost), Perbedaan dari kedua kriteria ini adalah dalam pemilihan kriteria mengambil keputusan.Kesimpulannya, dengan adanya sistem pendukung keputusan ini diharapkan proses seleksi penerima beasiswa KIP-Kuliah di Universitas Dehasen Bengkulu dapat berjalan dengan lebih efisien dan menghasilkan keputusan yang lebih akurat.Sistem Informasi Manajemen (SIM) sendiri adalah sebuah sistem formal dan informal yang menyajikan informasi mengenai sejarah, situasi saat ini, dan proyeksi masa depan melalui komunikasi lisan dan tulisan, terkait dengan berbagai operasi perusahaan dan lingkungan di sekitarnya. Selain itu, Sistem Penunjang Keputusan (Decision Support System) menjadi komponen penting dalam mendukung pengambilan keputusan untuk menyeleksi penerima beasiswa Kip Kuliah pengelola yayasan Universitas Dehasen Bengkulu memerlukan pendekatan yang lebih sistematis. tujuan penelitian untuk mengembangkan sistem pendukung keputusan membantu yayasan dalam proses seleksi penerima beasiswa. mempertimbangkan kriteria-kriteria tertentu, sistem diharapkan memberikan rekomendasi yang lebih akurat, sehingga proses seleksi dapat berjalan. Manfaat dari penelitian ini membantu pengelola mengambil keputusan lebih tepat. Metode SAW terdiri dari penilaian atribut setiap alternatif dan direpresentasikan dalam matriks penilaian keputusan. Matriks digunakan untuk menentukan seluruh kriteria dan skor dari setiap alternatif. Metode SAW memerlukan normalisasi matriks keputusan (X) untuk dibandingkan dengan peringkat alternatif yang ada. Metode SAW atribut kriteria ke-untungan (benefit) dan kriteria biaya (cost), Perbedaan dari kedua kriteria ini adalah dalam pemilihan kriteria mengambil keputusan.Kesimpulannya, dengan adanya sistem pendukung keputusan ini diharapkan proses seleksi penerima beasiswa KIP-Kuliah di Universitas Dehasen Bengkulu dapat berjalan dengan lebih efisien dan menghasilkan keputusan yang lebih akurat
- …
