Claim Missing Document
Check
Articles

Found 17 Documents
Search

ANALISIS CLUSTERING TEKS TANGGAPAN MASYARAKAT DI TWITTER TERHADAP PEMBATASAN SOSIAL BERSKALA BESAR MENGGUNAKAN ALGORITMA K-MEANS Muhammad Nur Akbar; Darmatasia Darmatasia; Mustikasari Mustikasari; Muh Syahwal
Jurnal INSYPRO (Information System and Processing) Vol 6 No 1 (2021)
Publisher : Prodi Sistem Informasi UIN Alauddin

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (609.636 KB) | DOI: 10.24252/insypro.v6i1.23325

Abstract

Virus corona (COVID-19) ditetapkan sebagai pandemi oleh WHO (World Health Organization atau Badan Kesehatan Dunia) karena penyebarannya yang terus meningkat dan telah mencapai sebagian besar negara di dunia, termasuk Indonesia. Setiap negara dituntut dapat lebih agresif dalam mengambil tindakan pencegahan dan perawatan. Pemerintah Indonesia sendiri mengeluarkan kebijakan berupa wajib masker, jam malam, serta PSBB (Pembatasan Sosial Berskala Besar) guna menekan laju menyebaran COVID-19.  Namun kebijakan tersebut menuai tanggapan  pro dan kontra dari masyarakat khususnya melalui media sosial, di satu sisi PSBB dianggap mampu menekan laju penyebaran COVID-19 namun di sisi lain PSBB dianggap akan memperburuk kondisi perekonomian masyarakat, khususnya golongan menengah bawah. Penelitian ini bertujuan untuk mengelompokkan tanggapan masyarakat mengenai PSBB di twitter ke dalam beberapa cluster, tanggapan yang berada dalam satu cluster yang sama dianggap memiliki topik atau karakteristik pembahasan yang serupa dan sebaliknya, sehingga dapat memberi insight tambahan pada pihak pemerintah dalam mengevaluasi kebijakannya. Algoritma K-Means digunakan untuk mengelompokkan tanggapan yang memiliki kesamaan karakteristik sebab terbukti memiliki tingkat akurasi yang tinggi dengan waktu eksekusi yang relatif cepat karena bersifat linear. Penelitian ini menghasilkan 4 cluster berbeda dengan mengunakan metode Elbow dalam penentuan jumlah K pada algoritma K-Means dan nilai SSE (Sum of Square Error) sebagai parameter evaluasinya.   
Analisis Sentimen Pengguna Indihome dengan Metode Klasifikasi Support Vector Machine (SVM) Muhammad Nur Akbar; Nur Annisa Safitri Yusuf; Nasrullah Nasrullah; Mubarak Mubarak
Journal Software, Hardware and Information Technology Vol 2 No 1 (2022)
Publisher : Jurusan Sistem Informasi Universitas Islam Negeri Alauddin Makassar

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24252/shift.v2i1.18

Abstract

Indonesia Digital Home (IndiHome) is a communication service provider from PT Telekomunikasi Indonesia (Telkom) that provides several communication and data service packages such as internet, home telephone and cable television (Usee TV & IP TV) which implements copper and fiber pptic cable services. Currently, IndiHome is implementing a 100% fiber service replacement for all customers in order to produce high data speeds and more reliable services. However, the fact is that fiber optic services often receive complaints from customers due to weather and other factors. It was recorded that in 2020 internet users in Indonesia reached 196.7 million people or 73.7% million of the population and around 51.2% were social media users (Kompas.com, 2020). One of the social media with 6.43 million active users is Twitter. Twitter then became a medium for channeling opinions regarding a service, including the services provided by Indihome. Based on this, a method is needed, namely sentiment analysis to understand whether the opinion is negative or positive. The Support Vector Machine (SVM) is used to create a classification model for sentiment analysis of IndiHome service users' opinions on Twitter with an accuracy of 91.3%.
Penambangan Pengklasifiksi Fuzzy dengan Multiobjective Evolutionary Fuzzy Classifier Nur Salman; Mustikasari Mustikasari; Muhammad Nur Akbar
Journal Software, Hardware and Information Technology Vol 2 No 1 (2022)
Publisher : Jurusan Sistem Informasi Universitas Islam Negeri Alauddin Makassar

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24252/shift.v2i1.24

Abstract

Classification is one of the key issues in the field of data mining and knowledge discovery. This paper implements a method of constructing a fuzzy rule mining classifier, which is extended in the context of classification. There are three stages of this approach: fuzzy rule set extraction, second; a linguistic labeling process that assigns a linguistic label to each fuzzy set. Owing to many attributes in the database, the feature selection process is also carried out, reducing the complexity to build the final classifier. Third: incorporate strategies to avoid rule redundancy and conflict into process mining. We applied the application Multiobjective Evolutionary Fuzzy Classifier (MOFC), which produced a classifier with satisfactory classification accuracy compared to other classifiers such as C4.5. In addition, in terms of classification based on association rules, MOFC can filter the large of rules and be proven to be able to build compact fuzzy models while maintaining a very good level of accuracy and producing a much smaller set of rules. We examine the performance of fuzzy rule classifiers through computational experiments on three benchmark data sets in the UCI machine learning repository.
PENERAPAN ENSEMBLE CLASSIFIER DALAM PENENTUAN PERSETUJUAN KREDIT BANK Muhammad Nur Akbar; M. Hasrul H; A. Muh Nur Hidayat
Ainet : Jurnal Informatika Vol 2, No 2 (2020): September (2020)
Publisher : Universitas Muhammadiyah Makassar

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26618/ainet.v2i1.4937

Abstract

Perkembangan teknologi informasi dalam bidang usaha perbankan sangat diperlukan dalam rangka meningkatkan kualitas layanan dengan memberikan kenyamanan, keamanan, dan kemudahan salah satunya pada produk kredit yang terdapat pada suatu bank. Pihak perbankan dihadapkan pada suatu masalah yaitu harus meningkatkan volume pemberian kredit namun juga harus dapat mengurangi resiko kredit bermasalah. Untuk meminimalisir kredit bermasalah pihak bank dituntut melakukan analisis yang tepat. Hal tersebut dapat diatasi salah satunya dengan memanfaatkan teknologi informasi sebagai pendukung pengambilan keputusan, berupa aplikasi data mining. Ensemble Classifier dengan majority voting ditawarkan sebagai solusi dangan nilai rata-rata akurasi sebesar 88.75% yang telah diuji pada kasus imbalanced class. Setelah diimplementasikan pada data testing sebanyak 274 nasabah, diperoleh keputusan 109 nasabah dinyatakan SETUJU dan 165 nasabah dinyatakan di TOLAK.
System Security for Motorcycle with Arduino Andi Muhammad Nur Nurhidayat; Muhammad Nur Akbar
Ainet : Jurnal Informatika Vol 2, No 2 (2020): September (2020)
Publisher : Universitas Muhammadiyah Makassar

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26618/ainet.v2i2.4938

Abstract

Nowadays, vehicles are the main requirement for carrying out activities of daily life. One of the most widely used by Indonesians is motorbikes. The motorcycle safety system provided by motorcycle manufacturers has not been effective and still has many shortcomings. Besides, the rampant motorcycle thief and robbery cases have made vehicle owners have to pay attention to their motorbikes' safety. An integrated motorcycle safety system is much needed. The design of a motorcycle vehicle system that is integrated with the Arduino via Bluetooth media provides quite maximum test results. Based on the results of testing, the accuracy of the tool gives 99% results. This safety system can turn off and turn on the motorcycle stock contact automatically. When the motorbike is on, Bluetooth on the motorbike and Bluetooth on the helmet are separated beyond a distance of 10 meters, the stock contact on the motorcycle vehicle will turn off so that the motorbike cannot be turned on at the same time the alarm will flash. The motor can be turned on after Bluetooth on the helmet, and the motorbike is connected or reconnected. With this, the safety standards on motorbikes are getting better. Motorcycle users will feel safe after leaving their vehicle.
Analisis Sentimen Terhadap Jasa Ekspedisi Pos Indonesia Pada Sosial Media Twitter Menggunakan Naïve Bayes Classifier Muhammad Nur Akbar; Darmatasia Darmatasia; Yulia Ardana
Journal Software, Hardware and Information Technology Vol 2 No 2 (2022)
Publisher : Jurusan Sistem Informasi Universitas Islam Negeri Alauddin Makassar

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24252/shift.v2i2.34

Abstract

Nowdays, the rapid growth of information technology positively impacts companies engaged in industry, sales, and services, especially e-commerce. The increase in the number of transactions in various e-commerce impacts the increase in the use of expedition services. Pos Indonesia is the oldest expedition service provider in Indonesia and is required to be able to innovate in providing the best service for its customers. The importance of customers for a company depends on how the company builds customer relationships. A strong company will have good customer relations. Many customers have expressed their opinions regarding Pos Indonesia through Twitter. In this study, text mining techniques are used, namely sentiment analysis which helps analyze opinions, sentiments, evaluations, assessments, attitudes, and public emotions towards Pos Indonesia services. Naïve Bayes Classifier was chosen because it is simple, fast, and has high accuracy. The Naïve Bayes Classifier has successfully classified positive and negative sentiments on 100 tweets from Pos Indonesia customers with an accuracy of 87%.
Aplikasi Blended Learning Pusat Pengembangan dan Penyaluran Potensi Mahasiswa Menggunakan Progressive Web App Andi Irsandi Ramadani; Andi Muhammad Syafar; Muhammad Nur Akbar
INFORMATION MANAGEMENT FOR EDUCATORS AND PROFESSIONALS : Journal of Information Management Vol 6 No 2 (2022): INFORMATION MANAGEMENT FOR EDUCATORS AND PROFESSIONALS (JUNI 2022)
Publisher : Penelitian dan Pengabdian Masyarakat Universitas Bina Insani

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.51211/imbi.v6i2.1863

Abstract

Ada suatu waktu dimana setiap mahasiswa akan menyadari bahwa ternyata mereka harus lebih banyak belajar sendiri daripada belajar dari dosen di kampus. Karena kampus hanya mengajarkan pengantar setiap mata kuliah baik perkuliahan yang bersifat teoritis maupun praktis. Ada beberapa platform e-Learning untuk mengasah potensi akademik maupun teknis. Namun kebanyakan platform tersebut hanya menyediakan rekaman-rekaman pembelajaran. Sehingga dalam proses belajar, yang terjadi adalah komunikasi satu arah yang tidak memungkinkan para pelajar memberikan sanggahan, umpan balik, atau bahkan pertanyaan jika ada hal yang sulit dimengerti. Selain dari itu, yang sering menjadi kendala adalah ketersediaan platform e- Learning yang stabil, cepat, cross platform, dan user experience yang intuitif. Aplikasi blended learning merupakan platform yang memfasilitasi mahasiswa untuk mengembangkan atau menyalurkan bakat dan potensinya secara blended learning (online dengan tatap muka). Sistem ini dibuat menggunakan progressive web app (APP) sehingga web menjadi capable, reliable, dan installable. Hasil dari sistem ini adalah platform bagi mahasiswa untuk peningkatan sekaligus sebagai platform penyaluran potensi, minat, atau bakat. Dengan aplikasi ini, diharapkan mahasiswa mendapat ruang dan cara yang lebih baik untuk belajar.
Sentiment Analysis Terhadap Review Aplikasi Maxim di Google Play Store Menggunakan Support Vector Machine (SVM) Muhammad Nur Akbar; Nur Hasanahlmar'iyah Rusydi; M. Hasrul H.; Nurul Shaumi Ramadhanti; Erfiana
AGENTS: Journal of Artificial Intelligence and Data Science Vol 2 No 2 (2022): Maret - Agustus
Publisher : Prodi Teknik Informatika Universitas Islam Negeri Makassar

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (1507.007 KB) | DOI: 10.24252/jagti.v2i2.39

Abstract

Before selecting and installing applications on the Google Play Store, users often read reviews of other users. This makes user review analysis very attractive for app owners to make future decisions. One of them is the Maxim application, a new online transportation application that provides different services from similar applications. This study aims to analyze user reviews of the maxim application on the Google Play Store using sentiment analysis. The research data is taken from the Google Play Store website, while the data taken is in the form of a review text. This user review analysis uses the Support Vector Machine (SVM) method producing an accuracy of 79%.
ANALISIS CLUSTERING UNTUK SEGMENTASI PENGGUNA KARTU KREDIT DENGAN MENGGUNAKAN ALGORITMA K-MEANS DAN PRINCIPAL COMPONENT ANALYSIS Muhammad Nur Akbar; Azizah Salsabila; Aldi Perdana Asri; Muhammad Syawir
AGENTS: Journal of Artificial Intelligence and Data Science Vol 3 No 1 (2023): September - Februari
Publisher : Prodi Teknik Informatika Universitas Islam Negeri Makassar

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (832.897 KB) | DOI: 10.24252/jagti.v3i1.56

Abstract

Customer segmentation is a process used by companies to group customers based on common characteristics. The goal is to understand customer needs and preferences better so that companies can provide products and services that match customer needs. One way to segment customers is to use clustering algorithms, such as k-means. This algorithm groups data into adjacent clusters with randomly selected centroids. In the case of credit card customer segmentation, the k-means algorithm can be used to group customers based on characteristics such as number of transactions, amount of payments, and credit history. Thus, companies can better understand the needs and preferences of credit card customers and determine more effective marketing strategies. The advantages of the k-means algorithm and the clustering method are that the developed models can help companies determine more effective marketing strategies, easy-to-use algorithms with fast computation time and accurate results, and the PCA algorithm is also used to reduce dimensions and makes data visualization easier. Based on the test results and analysis of credit card customer data, the performance of the k-means algorithm is considered relatively good for segmentation with the number of clusters = 3 and the Davies Bouldin value = -0.778.
Implementasi Plugin Metode Penilaian Otomatis UTS dan UAS pada Mata Kuliah Akbar, Muhammad Nur; Meidinah, Nur; Nur Hidayat, Andi Muhammad
Jurnal INSYPRO (Information System and Processing) Vol 8 No 2 (2023)
Publisher : Prodi Sistem Informasi UIN Alauddin

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24252/insypro.v8i2.42682

Abstract

The era of the 4.0 industrial revolution, as it is today, is developing rapidly in the field of technology in various countries, including Indonesia itself. Nowadays, there is a lot of technological sophistication that supports the development of the information world. However, some teachers still rely on manual evaluation in the learning process. The lack of implementation of technological advances of course makes the assessment process inefficient. The method of automatic valuation is a solution that can be used to deal with the problem. By applying e-learning technology, the process of digitization will be faster. In this research, that is the main focus is the exam. Moodle as an e-Learning technology has many features that can be applied, one of which features automatic evaluation. With the implementation of automated assessment, the assessment process for mid-term and end-term exams becomes more efficient and effective. Automatic evaluations tend to provide more accurate and consistent results, eliminating the potential for human error in the evaluation process, so students get feedback faster after completing the quiz, which can help them understand and improve their understanding.