Claim Missing Document
Check
Articles

Found 6 Documents
Search
Journal : JOINCS (Journal of Informatics, Network, and Computer Science)

Face Detection Using Linear Discriminant Analysis (Lda) Method and Support Vector Machine (Svm) Fajar Hariadi; Riwa Rambu Hada Enda
JOINCS (Journal of Informatics, Network, and Computer Science) Vol 2 No 1 (2019): April
Publisher : Universitas Muhammadiyah Sidoarjo

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (914.042 KB) | DOI: 10.21070/joincs.v1i2.521

Abstract

Safety and comfort are basic needs that must be met by all humans. People use CCTV that is often used to monitor public areas that have many people. Initially images from CCTV cameras were only sent via cable to a certain monitor room and needed direct supervision by security personnel with still low image resolution. One method for identifying faces is the Linear Discriminant Analysis (LDA) method. LDA is a method to find a linear subspace that maximizes the separation of two classes of patterns according to Fisher Criterion (fisher criteria weight). This study aims to detect faces with the Linear Discriminant Analysis (LDA) method as extraction features and classify facial images using the Support Vector Machine (SVM) method. The conclusion of this study is that the results obtained from face detection get a fairly high percentage of 84.2% for detected faces and 15.8% for undetectable faces and the results obtained are influenced by a fairly good facial image and image cropping process good and unchanging face position which makes it easy to detect faces.
Comparative Analysis of Text Mining Results With Tf ldf Features and SQL Like Operator in Indonesian News Search Riwa Rambu Hada Enda; Fajar Hariadi
JOINCS (Journal of Informatics, Network, and Computer Science) Vol 3 No 1 (2020): April
Publisher : Universitas Muhammadiyah Sidoarjo

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (4619.229 KB) | DOI: 10.21070/joincs.v3i0.714

Abstract

Research on the implementation of text mining uses the TF IDF method to be used in the Information retrieval / Indonesian news search feature. The dataset used was sourced from NewsAPI and built a Codeigniter based website named "News Plus Six Dua". This study also uses the Vertor Space Model (VSM) method to overcome the weaknesses of the TF IDF method at the time of the sorting process. The results of this study explain that the search by the TF IDF method has higher accuracy when compared to SQL like operators. TF IDF produces a percentage of precision 100% and recall (sensitivity) 66.7% on searches with the keyword "Indonesian soccer schedule" while SQL like operators do not display search results or equal to 0%. But the TF IDF method has the disadvantage of running slower than SQL like operators. This has been tested using either the number of words or terms entered, the number of datasets, and the location of access. At the location of access, access via hosting is monitored faster when compared via localhost.
The Effect Of Load Balancing And Failover Of Two Wide Area Networks With Per Connection Classifier Method On Qos Throughput, Packet Loss, Qos Delay, And Qos Jitter Fajar Hariadi; Pingky Alfa Ray Leo Lede; Umbu Melvy Kalaway
JOINCS (Journal of Informatics, Network, and Computer Science) Vol 4 No 2 (2021): November
Publisher : Universitas Muhammadiyah Sidoarjo

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (935.175 KB) | DOI: 10.21070/joincs.v4i2.1502

Abstract

This study discussed the design and implementation of load balancing of two WANs so that the traffic load and bandwidth provided by both WANs could be used equally. The methods used in this research were observation and identification, network topology design, implementation, and testing. This study found that load balancing could divide the network load to all available gateways so that there was no overload on the network used by visitors. After implementing load balancing, the network throughput parameter increased 1.51% compared to the Indihome 1 network and 7.5% compared to the Indihome 2 network. The packet loss parameter increased by 99% compared to the Indihome 1 network and increased 90% compared to the Indihome 2 network. Network delay decreased by 1.33% compared to the Indihome 1 network, and decreased by 1.5% compared to the Indihome 2 network. The Network jitter parameter decreased by 0.93% compared to the Indihome 1 network and decreased by 1.54% compared to the Indihome 2 network.
The Effect Of Queue Tree On Packet Loss In Bandwitch Management Online Based School Exam Wellem Wila Dale; Fajar Hariadi; Raynesta Mikaela Indri Malo
JOINCS (Journal of Informatics, Network, and Computer Science) Vol 4 No 2 (2021): November
Publisher : Universitas Muhammadiyah Sidoarjo

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (1354.951 KB) | DOI: 10.21070/joincs.v4i2.1515

Abstract

Unequal bandwidth distribution between clients that are used for online tests will cause the quality connection among clients to differ. Especially when some client dominates the bandwidth. The other client will suffer to have a proper connection. Using Queue Tree bandwidth management with the type of Per Connection Queue (PCQ) the distribution of bandwidth can be done evenly among clients. The implementation of PCQ Queue Tree is able to reduce the length of ping time by 90% from 134 ms and 13 ms. Delay is decreased 47% from 17 ms to 9 ms with a very good category. Jitter decreased 41% from 17 ms to 7 ms with a very good category. Throughput increased 53% from 234.582 Kbps to 503.584 Kbps in the medium category. Meanwhile, in the packet loss parameter, there is no difference in conditions before and after the implementation of PCQ Queue Tree because there is no packet loss during the test, which means packet loss condition is in the very good category.
Expert System for Diagnosing Diseases in Tomato Plants Using a Website-Based Forward Chaining Method Katrina Rambu Bangi Roni; Fajar Hariadi
JOINCS (Journal of Informatics, Network, and Computer Science) Vol 5 No 2 (2022): November
Publisher : Universitas Muhammadiyah Sidoarjo

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21070/joincs.v5i2.1609

Abstract

Tomato plant is one of the plants in agriculture that is widely cultivated in Indonesia, one of which is in East Sumba Regency. The problem that is often faced is the lack of knowledge and facilities for tomato farmers which are still limited in dealing with diseases in these tomatoes. To diagnose tomato plant diseases that attack through the symptoms they cause, a tomato plant expert is needed to analyze and diagnose. Currently farmers often ignore tomato diseases due to lack of knowledge. They assume it's finally happening in the planting phase, tomato pest and disease experts are reduced, they don't know how to control and eliminate decreased productivity. In this case, to help farmers know the type of disease that attacks, an expert system is needed to help diagnose tomato plant diseases and provide the right solution in handling the disease using the Forward Chaining method. In the Forward Chaining method, the data search process starts from the premise to the final conclusion. Based on the test results, it can be seen that the probability value for the accuracy of the expert system produced is capable of having a percentage of 100%, from the calculation of the 8 types of the same disease, for inaccuracy so that this expert system can be stated to have been running well.
User And Bandwidth Management Using Mikrotik Hotspot At State Vocational School 5 Waingapu Fajar Hariadi; Deonisius Anamatalu
JOINCS (Journal of Informatics, Network, and Computer Science) Vol 6 No 2 (2023): November
Publisher : Universitas Muhammadiyah Sidoarjo

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21070/joincs.v6i2.1605

Abstract

The use of the internet in schools is very helpful for teachers, staff and students, the internet is an important need for most adult people, they even regard the internet as a basic daily need that must be met, such as at SMK Negeri 5 Waingapu. The use of the internet in schools is very helpful for teachers, staff and students. State Vocational High School 5 Waingapu uses an internet service provider (ISP) sourced from KOMINFO Bakti as internet access with a total bandwidth of 2 Mbps with an average user of 15 clients. The distance to access the internet network on WIFi is 20 meters, but there is still no good management of network usage so that when accessing the internet, many users (clients) often experience interference. One form of utilizing wireless technology is a hotspot. The development method used in this research is to develop a hotspot network at SMK Negeri 5 Waingapu using the Network Development Life Cycle (NDLC) model which is used as the overall development method in developing or designing computer network systems. By implementing bandwidth management using a hotspot router proxy at SMK Negeri 5 Waingapu and testing it using a speedtest before implementation, namely a download speed of 4.55 and an upload speed of 1.09. And after the implementation of bandwidth management for teacher user hotspots, namely for download bandwidth of 0.95 Mbps and upload bandwidth of 0.21 Mbps, then for hotspot user principals download bandwidth is 0.94 Mbps and upload bandwidth is 0.94 Mbps and hotspot user download bandwidth for students is 0.73 Mbps and upload of 0.94 Mbps.