34 research outputs found
Performance Analysis of Color based Image Retrieval
No full textRecently Content based image retrieval (CBIR) is an active research. This paper proposes a technique to retrieve images based on color feature and evaluate the retrieval system performance. In this retrieval system Euclidean distance and City block distance are used to measure similarity of images. This algorithm is tested by using Corel image database which is provided by James Wang. The performance of retrieval system is measured in terms of its recall and precision. The effectiveness of retrieval system is also measured based on Average Rank (AVRR) of all relevant retrieves images and Ideal Average Rank of relevant images (IAVRR). The experimental results show that city block has achieved higher retrieval performance than Euclidean distance.</jats:p
Rupiah Exchange Prediction of US Dollar Using Linear, Polynomial, and Radial Basis Function Kernel in Support Vector Regression
Get PDFAs a developing country, Indonesia is affected by fluctuations in foreign exchange rates, especially the US Dollar. Determination of foreign exchange rates must be profitable so a country can run its economy well. The prediction of the exchange rate is done to find out the large exchange rates that occur in the future and the government can take the right policy. Prediction is done by one of the Machine Learning methods, namely the Support Vector Regression (SVR) algorithm. The prediction model is made using three kernels in SVR. Each kernel has the best model and, the accuracy and error values are compared. The Linear Kernel has C = 7, max_iter = 100. The Polynomial Kernel has gamma = 1, degree = 1, max_iter = 4000 and C = 700. The RBF kernel has gamma = 0.03, epsilon = 0.007, max_iter = 2000 and C = 100. Linear kernels have advantages in terms of processing time compared to Polynomial and Radial Basis Function (RBF) kernels with an average processing time of 0.18 seconds. Besides that, in terms of accuracy and error, the RBF kernel has advantages over the Linear and Polynomial kernels with the value R2 = 95.94% and RMSE = 1.25%
SENTIMENT ANALYSIS OF ELECTRIC CARS USING RECURRENT NEURAL NETWORK METHOD IN INDONESIAN TWEETS
No full textSentiment analysis is computational research of the opinions of many people who are textually expressed against a particular topic. Twitter is the most popular communication tool among Internet users today to express their opinions. Deep Learning is a solution to allow computers to learn from experience and understand the world in terms of the hierarchy concept. Deep Learning objectives replace manual assignments with learning. The development of deep learning has a set of algorithms that focus on learning data representation. The recurrent Neural Network is one of the machine learning methods included in Deep learning because the data is processed through multi-players. RNN is also an algorithm that can recall the input with internal memory, therefore it is suitable for machine learning problems involving sequential data. The study aims to test models that have been created from tweets that are positive, negative, and neutral sentiment to determine the accuracy of the models. The models have been created using the Recurrent Neural Network when applied to tweet classifications to mark the individual classes of Indonesian-language tweet data sentiment. From the experiments conducted, results on the built system showed that the best test results in the tweet data with the RNN method using Confusion Matrix are with Precision 0.618, Recall 0.507 and Accuracy 0.722 on the data amounted to 3000 data and comparative data training and data testing of ratio data 80:2
User Experience Analysis of Employee Attendance List on Talent Application with Heuristic Evaluation Method
Get PDFThe development of digital technology has influenced human resource management systems, particularly in the management of employee attendance records. One of the most widely used applications in Indonesia is Mekari Talenta, a cloud-based HRIS platform with features ranging from online attendance, leave, overtime, to payroll integration. Despite its high rating on the Google Play Store, there are still a number of complaints regarding user experience, such as confusing navigation, an unintuitive interface, and difficulty in accessing certain features. This study aims to analyze the user experience on the Talenta application using the Heuristic Evaluation method based on Nielsen's 10 principles. Data collection was conducted through questionnaires and processed using SPSS for validity, reliability, and descriptive percentage analysis. The results of the study show that most of the Heuristic Evaluation principles scored in the "Good" category, especially in terms of visibility of system status, consistency and standards, and aesthetic and minimalist design. However, there are still weaknesses in terms of help and documentation as well as error prevention that need improvement. These findings recommend that developers improve the interface display, clarify the help documentation, and optimize the error prevention feature so that the application can provide a more optimal user experience. Further research is recommended using other evaluation methods, such as the User Experience Questionnaire (UEQ) or in-depth interviews, to obtain a more comprehensive picture
Arsitektur Convolutional Neural Network (CNN) Alexnet Untuk Klasifikasi Hama Pada Citra Daun Tanaman Kopi
No full textIndonesia is the fourth largest coffee producing country in the world. However, when compared to 3 other countries, Indonesia's coffee production is still relatively small. Many factors cause this to happen, including the number of farmers' coffee trees that are attacked by diseases. If the handling of this disease is slow, then the disease in one tree can be transmitted to other trees. This causes a decrease in Indonesian coffee productivity. In this study, the author implemented the Alexnet Convolutional Neural Network (CNN) architecture using the MATLAB programming platform for the identification of diseases in coffee plants through images. The total number of datasets used is 300 data which is divided into 3 classes, namely health, rust and red spider mite. The training process involving 260 training data resulted in an accuracy of 69.44-80.56%. The network testing process using 40 test data resulted in an accuracy of 81.6%. Based on the results of the study, it can be said that the Alexnet architecture is accurate for the classification of leaf pests on coffee plant
Trend Analysis in Sales Forecasting and Decision Support Systems AHP Method on the Selection of Types of Motorcycles PT. AHM
Get PDFDuring the current Covid-19 pandemic, it has a direct impact on decreasing motorcycle sales at PT. Astra Honda Motor in large enough numbers, resulting in losses for the company. This study aims to analyze the forecasting of motorcycle sales in the future and supported by a decision support system in selecting the type of motorbike that consumers are most interested in. The method used in this research is the Trend method and the Analytical Hierarchy Process (AHP) method. The results of this study are based on the mean square error analysis, sales forecasting method at PT. Astra uses exponential trend analysis. The results of forecasting sales for the next 2 periods using the exponential trend analysis method are 2,715,032 units and 2,671,937 units, the sales levels tend to be the same as sales in the 2020 period. Meanwhile, the results of the decision support system analysis use the Analytical Hierarchy Process method (AHP), in choosing the type of motorbike that consumers are most interested in is an automatic motorbike at 0.63 with the preferred alternative is the fuel consumption aspect of 0.36
APLIKASI DATA MINING MENGGUNAKAN ATURAN ASOSIASI DENGAN METODE APRIORI UNTUK ANALISIS KERANJANG PASAR PADA DATA TRANSAKSI PENJUALAN APOTEK
No full textTeknik analisis keranjang pasar adalah teknik untuk menemukan pola berupa produk-produk yang sering
dibeli bersamaan atau cenderung muncul bersama dalam sebuah transaksi. Penelitian ini bertujuan untuk
membuat sebuah aplikasi data mining menggunakan aturan asosiasi dengan metode apriori sebagai teknik
analisis keranjang pasarnya. Data yang diambil dalam penelitian ini adalah data transaksi penjualan disuatu
apotek di Perumnas 1. Hasil dari aturan asosiasi yang didapat yaitu berupa kombinasi dari jenis obat yang
sering dibeli oleh konsumen. Dari hasil tersebut diharapkan dapat membantu manajemen apotek untuk merancang
strategi pemasaran obat di apoteknya. Aplikasi ini dibuat menggunakan perangkat lunak Java dan
didukung oleh media penyimpanan database Microsoft Access.
Kata kunci : analisis keranjang pasar,aturan asosiasi, apriori, java
APLIKASI PENCARIAN BUKU BERBASIS WEB SEMANTIK UNTUK PERPUSTAKAAN SMK YADIKA 7 BOGOR
No full textAplikasi pencarian berbasis web semantik ini ditujukan untuk mempermudah
pengunjung perpustakaan di SMK Yadika 7 Bogor dalam melakukan pencarian buku.
Penelitian ini dilakukan dengan mengumpulkan data dan informasi dari perpustakaan
serta merangkum teori mengenai web semantik dan unsur-unsur yang dibutuhkan dalam
pembuatan aplikasi ini.
Pembuatan aplikasi ini dimulai dengan perancangan ontologi yang digunakan
dalam pembuatan struktur semantik perpustakaan. Selanjutnya dilakukan perancangan
query serta perancangan tampilan aplikasi. Setelah tahap perancangan, dilanjutkan
dengan tahap implementasi yaitu implementasi query dan implementasi interface.
Setelah aplikasi selesai dikerjakan, akan dilakukan ujicoba dan aplikasi akan dievaluasi
oleh pengguna
SISTEM MEDIA SOSIAL FORUM INTERAKSI MAHASISWA DENGAN PENAMBAHAN FITUR FOTO TAGGING
No full textTeknologi media sosial di Indonesia saat ini umumnya belum ada yang menyediakan penyatuan beberapa group akademik perguruan tinggi menjadi kesatuan group dengan adanya pengkategorian topik. Media sosial yang lahir di Indonesia juga masih jarang ditemukan adanya fitur yang menarik seperti foto tagging untuk menambah kenyamanan pengguna khususnya mahasiswa untuk saling berinteraksi. Hal ini mendorong Peneliti untuk merancang dan mengimplementasikan sistem media sosial forum interaksi mahasiswa dengan penambahan foto tagging. Fitur pertemanan dan pengkategorian jurusan mempermudah mahasiswa untuk berbagi informasi. Fitur foto tagging juga dapat mempermudah mahasiswa untuk saling berinteraksi. Selain itu Administrator dapat mengelola seluruh data mahasiswa yang terdaftar sebagai anggota maupun moderator forum. Sistem ini dibuat dengan menggunakan bahasa pemrograman jQuery, PHP dan database MySQL. Pada hasil pengujian blackbox dihasilkan semua fungsi dari tiap fitur berjalan baik, sedangkan dari hasil evaluasi pengguna dengan menggunakan kuesioner, responden rata-rata menyatakan sangat setuju bahwa sistem memang mudah digunakan, berkualitas, serta bermanfaat. Dengan demikian, peneliti menyimpulkan bahwa sistem media sosial forum ini mempermudah mahasiswa untuk saling berinteraksi dalam berdiskusi seputar informasi akademik maupun non akademik
Implementasi Metode Reccurrent Neural Network pada Text Summarization dengan Teknik Abstraktif
Get PDFText Summarization atau peringkas text merupakan salah satu penerapan Artificial Intelligence (AI) dimana komputer dapat meringkas text pada suatu kalimat atau artikel menjadi lebih sederhana dengan tujuan untuk mempermudah manusia dalam mengambil kesimpulan dari artikel yang panjang tanpa harus membaca secara keseluruhan. Peringkasan teks secara otomatis dengan menggunakan teknik Abstraktif memiliki kemampuan meringkas teks lebih natural sebagaimana manusia meringkas dibandingkan dengan teknik ekstraktif yang hanya menyusun kalimat berdasarkan frekuensi kemunculan kata. Untuk dapat menghasilkan sistem peringkas teks dengan metode abstraktif, membutuhkan metode Recurrent Neural Network (RNN) yang memiliki sistematika perhitungan bobot secara berulang. RNN merupakan bagian dari Deep Learning dimana nilai akurasi yang dihasilkan dapat lebih baik dibandingkan dengan jaringan saraf tiruan sederhana karena bobot yang dihitung akan lebih akurat mendekati persamaan setiap kata. Jenis RNN yang digunakan adalah LSTM (Long Short Term Memory) untuk menutupi kekurangan pada RNN yang tidak dapat menyimpan memori untuk dipilah dan menambahkan mekanisme Attention agar setiap kata dapat lebih fokus pada konteks. Penelitian ini menguji performa sistem menggunakan Precision, Recall, dan F-Measure dengan membandingan hasil ringkasan yang dihasilkan oleh sistem dan ringkasan yang dibuat oleh manusia. Dataset yang digunakan adalah data artikel berita dengan jumlah total artikel sebanyak 4515 buah artikel. Pengujian dibagi berdasarkan data dengan menggunakan Stemming dan dengan teknik Non-stemming. Nilai rata-rata recall artikel berita non-stemming adalah sebesar 41%, precision sebesar 81%, dan F-measure sebesar 54,27%. Sedangkan nilai rata-rata recall artikel berita dengan teknik stemming sebesar 44%, precision sebesar 88%, dan F-measure sebesar 58,20 %.AbstractText Summarization is the application of Artificial Intelligence (AI) where the computer can summarize text of article to make it easier for humans to draw conclusions from long articles without having to read entirely. Abstractive techniques has ability to summarize the text more naturally as humans summarize. The summary results from abstractive techinques are more in context when compared to extractive techniques which only arrange sentences based on the frequency of occurrence of the word. To be able to produce a text summarization system with an abstractive techniques, it is required Deep Learning by using the Recurrent Neural Network (RNN) rather than simple Artificial Neural Network (ANN) method which has a systematic calculation of weight repeatedly in order to improve accuracy. The type of RNN used is LSTM (Long Short Term Memory) to cover the shortcomings of the RNN which cannot store memory to be sorted and add an Attention mechanism so that each word can focus more on the context.This study examines the performance of Precision, Recall, and F-Measure from the comparison of the summary results produced by the system and summaries made by humans. The dataset used is news article data with 4515 articles. Testing was divided based on data using Stemming and Non-stemming techniques. The average recall value of non-stemming news articles is 41%, precision is 81%, and F-measure is 54.27%. While the average value of recall of news articles with stemming technique is 44%, precision is 88%, and F-measure is 58.20%
