cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
Rudy Herteno
Contact Email
rudy.herteno@ulm.ac.id
Phone
+6282250380732
Journal Mail Official
rudy.herteno@ulm.ac.id
Editorial Address
Jalan Ahmad Yani KM. 36, Kalimantan Selatan
Location
Kota banjarmasin,
Kalimantan selatan
INDONESIA
Journal of Data Science and Software Engineering
Published by Universitas Lambung Mangkurat
ISSN : 27755320     EISSN : 27755487     DOI : https://doi.org/10.20527/jdsse.v1i01.13
Core Subject : Science,
Computer Science & IT
Journal of Data Science and Software Engineering adalah jurnal yang dikelola oleh program studi Ilmu Komputer Universitas Lambung Mangkurat untuk mempublikasikan artikel ilmiah mahasiswa tugas akhir. Terbit tiga kali dalam setahun.
Arjuna Subject : Ilmu Komputer (semua kategori) - Ilmu Komputer (Lain-Lain)
Articles 5 Documents
 1 
Search results for , issue "Vol 1 No 02 (2020)" : 5 Documents clear
PERFORMANCE ANALYSIS OF CLASSIFIER ON FACEBOOK DATA USING UNIGRAM & BIGRAM COMBINATIONS Yudha Sulistiyo Wibowo; Mohammad Reza Faisal; Ahmad Rusadi; Dodon T Nugrahadi; Muhammad Itqan Mazdadi
Journal of Data Science and Software Engineering Vol 1 No 02 (2020)
Publisher : Fakultas MIPA Universitas Lambung Mangkurat

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (240.626 KB)

Abstract

Penelitian ini pada analisis sentimen menggunakan metode random forest sebagai klasifikasi. TF-IDF adalah fitur berbobot dan fitur kombinasi dari N-Grams adalah unigram dan bigram sebagai kata fitur. Dalam penelitian ini TF-IDF digunakan untuk fitur ekstraksi, tes ini menggunakan data Komentar Facebook tentang Berita Olahraga. Dalam studi ini, dataset digunakan sebanyak 1000 data terbagi menjadi 2, yaitu data pengujian dan pelatihan data. Mencapai kinerja akurasi tinggi hasil dalam fitur unigram dengan akurasi 83,67% dari 2757 fitur, Bigram menghasilkan 58% dengan fitur sebanyak 8457.
Penerapan Long Short Term Memory RNN untuk Prediksi Transaksi Penjualan Minimarket Patrick Ringkuangan; Fatma Indriani; Muhammad Itqan Mazdadi; Irwan Budiman; Andi Farmadi
Journal of Data Science and Software Engineering Vol 1 No 02 (2020)
Publisher : Fakultas MIPA Universitas Lambung Mangkurat

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (475.488 KB)

Abstract

This study aims to determine whether it can build a prediction of sales of goods at the Lapan-Lapan Mart by using the Long Short Term Memory Recurrent Neural Network method that can be used to predict the sale of goods. In this study, the data was taken from the Lapan-Lapan Mart, together with data on 10 different items sold every day. The data is then compiled for the level of sales to be weekly and a total of 52 data is obtained for each item so that the total data is amounted to 520. To get the weight in the LSTM calculation, there are two processes, namely forward and backward . the weight will be used to make predictions using the basic formula of the LSTM.Based on the research that has been done, it is known that the highest accuracy of using MAD (Mean Absolute Deviation) is 91 gr (11.61803507) indomie goods and 1.8kg of lemon daia (2.077000464) for the lowest MAD
IMPLEMENTASI ALGORITMA C5.0 UNTUK MEMBENTUK POLA POHON KEPUTUSAN DIAGNOSA PENYAKIT DIABETES MELLITUS Muhammad Latief Saputra; Irwan Budiman; Radityo Adi Nugroho; Dwi Kartini; Muliadi
Journal of Data Science and Software Engineering Vol 1 No 02 (2020)
Publisher : Fakultas MIPA Universitas Lambung Mangkurat

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (464.45 KB)

Abstract

This study applies the C5.0 algorithm to form a decision tree pattern for diagnosing diabetes mellitus. C5.0 algorithm is a decision tree based classification algorithm. This algorithm focuses on the acquisition of information gain on all attributes. The data used is a diabetes mellitus dataset obtained from the Kaggle database website. Data preprocessing is done and data sharing is done 4 times with the distribution of training data 60% 70% 80% and 90%. Data sharing uses stratafied random sampling methods so that the distribution of training and testing data is in accordance with its portion. Calculation of accuracy performance using confusion matrix. Classification performance using C5.0 algorithm. With 90% training data get 72.73% accuracy of rules generated as many as 70 rules. With 80% training data the accuracy value is 74.03%. The rule is 64 rules. With 70% training data get an accuracy value of 76.52% of the rules generated 59 rules. With 60% training data get an accuracy value of 74.59% of the rules generated as many as 53 rules. From all the experiments that have been done, the best accuracy is found in experiments with 70% training data.
IMPLEMENTASI METODE CONVOLUTIONAL NEURAL NETWORK UNTUK PREDIKSI HARGA SAHAM LQ45 Aris Pratama; Dwi Kartini; Akhmad Yusuf; Andi Farmadi; Irwan Budiman
Journal of Data Science and Software Engineering Vol 1 No 02 (2020)
Publisher : Fakultas MIPA Universitas Lambung Mangkurat

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (195.714 KB)

Abstract

Stock are securities of ownership of a company. Investments in the stock market on average can produce a return rate of 10-30% per year, this amount is about two to three times higher than the rate of return on deposits or savings in banks which are only 5-10 % every year. One problem is the stock price is fluctuating or changing due to certain factors. This study compares several window size data with different amounts of data, aiming to find window size data with a more accurate amount of data for stock price predictions. Convolutional neural network algorithm with window size data of 7 days, 14 days, 21 days and 28 days in the amount of data 1 year and 2 years for stock price predictions. The results of this study are the convolutional neural network algorithm with a data window size of 7 days at the amount of data 2 years is more accurate than the window size data and the amount of other data. Because the smallest error result is 0.000201587.
IMPLEMENTATION OF LOAD BALANCE EQUAL COST MULTI PATH (ECMP) BETWEEN ROUTING PROTOCOL BORDER GATEWAY PROTOCOL (BGP) AND OPEN SHORTEST PATH FIRST (OSPF) USING DUAL CONNECTION Aji Triwerdaya; Dodon Trianto Nugrahadi; Muhammad Itqan Masdadi; Irwan Budiman; Ahmad Rusadi Arrahimi
Journal of Data Science and Software Engineering Vol 1 No 02 (2020)
Publisher : Fakultas MIPA Universitas Lambung Mangkurat

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (601.437 KB)

Abstract

Currently, Internet is needed by everyone to lighten their work, then a method has been developed to be able to access the internet using 2 ISPs (Internet Service Providers), namely using load balance. This method can perform bandwidth management so that it can balance the bandwidth of 2 ISPs. To support this method, Load Balance Equal Cost Multi Path (ECMP) is used. Another innovation that continues to be developed routing, the process of exchange data packets between different IP networks and to identify the best route to each connected network, that can make routing better by using dynamic routing types, to unify the network if a change occurs of topology by exchanging new topology information with each other on a network using the Open Shortest Path First (OSPF) routing or using the Border Gateway Protocol (BGP). OSPF is an open source routing protocol that is often used[4] and OSPF is a link-state in the routing algorithm. This routing use the Dijkstra or SPF (Short Path First) algorithm to calculate the shortest path from each route. Coinciding with the increase in routers in an area, the information that routers in the same area must have at the same time will increase, then the Border Gateway Protocol (BGP) is the new routing protocol[7]. BGP is a vector-path protocol where each router decides locally the "best AS" line per destination. The local preference attribute is used to set the policy for outgoing traffic. Testing is done by comparing the performance of an ECMP network using OSPF routing and an ECMP network using BGP routing[3]. Testing is done by measuring based on the throughput and data delay parameters using 16, 32, 48 routers. the topology is divided into 3 areas, namely area 1 for user load balance, area 2 for ISP 1 and area 3 for ISP 2. Throughput is used to measure routing performance on the TCP transport protocol and UDP transport protocol. Then, data delay is for measuring the performance of routing on the TCP and UDP transport protocol with the addition of variations. The testing that have been carried out show that the network throughput with OSPF routing (764.13 bps) has a lower performance than the network with BGP routing (818.81 bps) when sending TCP and UDP data, and network delay with OSPF routing (85.61 ms) has a significant increase than the network with BGP routing (89.23 ms) when sending TCP and UDP data.

Page 1 of 1 | Total Record : 5

 1 

Filter by Year

2020 2020


Reset
Filter By Issues
All Issue Vol 3 No 03 (2022) Vol 3 No 02 (2022) Vol 3 No 01 (2022) Vol 2 No 03 (2021) Vol 2 No 02 (2021) Vol 2 No 01 (2021) Vol 1 No 02 (2020) Vol 1 No 01 (2020)