cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
Mohammad Sani Suprayogi
Contact Email
yogie@usm.ac.id
Phone
-
Journal Mail Official
santi@usm.ac.id
Editorial Address
-
Location
Kota semarang,
Jawa tengah
INDONESIA
Jurnal Transformatika
Published by Universitas Semarang
ISSN : 16933656     EISSN : 24606731     DOI : -
Core Subject : Science,
Computer Science & IT
Transformatika is a peer reviewed Journal in Indonesian and English published two issues per year (January and July). The aim of Transformatika is to publish high-quality articles of the latest developments in the field of Information Technology. We accept the article with the scope of Information Systems, Web Technology, Computer Networks, Artificial Intelligence, and Multimedia.
Arjuna Subject : -
Articles 14 Documents
 1   2 
Search results for , issue "Vol. 21 No. 2 (2024): Januari 2024" : 14 Documents clear
GLOBAL THRESHOLDING IMPLEMENTATION FOR NOISE HANDLING IN DIGITAL IMAGE RECOGNITION Purwanto, Dannu; Agustiyar, Agustiyar
Jurnal Transformatika Vol. 21 No. 2 (2024): Januari 2024
Publisher : Jurusan Teknologi Informasi Universitas Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26623/transformatika.v21i2.8713

Abstract

Text recognition (OCR - Optical Character Recognition) is a research field that is gaining widespread attention due to its wide application in image and document processing. Although OCR technology has achieved a high level of success, the main challenge faced is the presence of noise in text image, noise causes decreased text recognition results, noise causes miss classification. Therefore needed noise handling text recognition.  The aim of this research is to provide valuable insight into the techniques and approaches used in the context of noise treatment using global threshold methods. The method used starts from an input digital image, then preprocessing is carried out by converting the image into a gray scale image, then a threshold is applied to the image, then recognition is carried out. From 6 experiments, the best results were obtained for character recognition with a threshold value (t) of 65 and a character recognition accuracy percentage of 94.29%. T value determined manually and static for separates the all object and the background, while in reality the lighting or contrast always varies. Suggestions for further research include developing an adaptive thresholding method approach to obtain threshold values automatically and optimally. So that if faced with varying lighting conditions or contrast, better results can be obtained.
Random State Parameter Undersampling untuk Penanganan Data dengan Kelas Tidak Seimbang pada Algoritme Random Forest Setiaji, Galet Guntoro; Suntoro, Joko; Rifa'i, Ahmad
Jurnal Transformatika Vol. 21 No. 2 (2024): Januari 2024
Publisher : Jurusan Teknologi Informasi Universitas Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26623/transformatika.v21i2.8901

Abstract

Algoritme Random Forest (RF) sangat populer digunakan pada metode klasifikasi karena waktu learning yang cepat, mampu melakukan pembobotan pada variabel, dan kinerja yang sangat baik pada dataset berukuruan besar, namun algoritme RF mempunyai performa yang buruk saat menangani data dengan kelas tidak seimbang. Data dengan kelas tidak seimbang adalah jumlah data pada kelas tertentu lebih banyak dibandingkan dengan jumlah data pada kelas lainnya. Undersampling (US-RF) adalah salah satu metode yang digunakan untuk penanganan data dengan kelas tidak seimbang, namun metode undersampling akan memilih dan mereduksi data secara acak pada kelas mayoritas sehingga berakibat hilangnya data yang berpotensi berguna. Untuk menghindari hilangnya data yang berpotensi berguna tersebut karena dipilih secara acak, maka akan diterapkan penetapan nilai random state pada metode undersampling. Metode yang diusulkan diberi nama random state parameter undersampling Random Forest (RSUS-RF). Dalam penelitian ini akan dibandingkan antara metode RF, US-RF dan RSUS-RF. Hasil penelitian menunjukkan nilai rata-rata akurasi metode RSUS-RF lebih tinggi dibandingkan dengan metode RF dan US-RF dengan nilai rata-rata akurasi metode RSUS-RF sebesar 0.8259, sedangkan nilai rata-rata akurasi metode RF dan metode US-RF sebesar 0.8035 dan 0.7945. Serta terdapat perbedaan secara signifikan diantara ketiga metode tersebut ketika diuji menggunakan Friedman Test dengan nilai p-value adalah 0.005. 
DECISION SUPPORT SYSTEM PEMBUKAAN LOKASI BARU JASA SERVIS MOTOR BERBASIS PROFILE MATCHING Nuryanto, Imam; Setiawan, Aries; Farida, Ida; Wibowo, Sasono; Widjajanto, Budi; Prihandono, Adi
Jurnal Transformatika Vol. 21 No. 2 (2024): Januari 2024
Publisher : Jurusan Teknologi Informasi Universitas Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26623/transformatika.v21i2.8166

Abstract

As the volume of motorbikes increases, the number of motorbike service outlets also increases. However, not all service places get customers equally, from the customer's side various variables really determine visits such as location, level of spare parts availability, level of service to consumers and price. Likewise, in terms of service owners, when opening a new service location, they need to pay attention to various variables such as proximity to residential areas, number of competitors, capacity of passing vehicles, and proximity to spare parts suppliers. To collaborate several influential variables to produce a decision regarding the right place, a method is needed that is capable of carrying out calculations to produce a ranking of locations that will have an influence. One method that can be offered is profile matching. This method performs by finding the difference between the weight value determined at the beginning and the input value for each location object. The ranking results of all location objects can be used as alternative locations for appropriate service locations..
EXPERT SYSTEM FOR DIAGNOSING DISEASES IN DOGS BASED ON WEB USING DEMPSTER-SHAFER METHOD Nalle, Olsen Junior; Sina, Derwin Rony; Ledoh, Juan Rizky Mannuel
Jurnal Transformatika Vol. 21 No. 2 (2024): Januari 2024
Publisher : Jurusan Teknologi Informasi Universitas Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26623/transformatika.v21i2.7425

Abstract

In the study titled " Expert System For Diagnosing Diseases In Dogs Based On Web Using Dempster-Shafer Method " it was found that keeping dogs provides positive benefits to human mental health. Despite dogs having rapid adaptability, high loyalty, as well as significant development and intelligence, proper care is essential to maintain their health and prevent diseases. The limitation in the number of veterinary doctors poses a challenge in handling animal diseases, particularly in dogs. Hence, this research utilizes an Expert System with the Dempster-Shafer method to diagnose canine diseases. Out of 45 test cases used to evaluate the system, it achieved an accuracy rate of 88.89% in comparison to veterinary doctors' diagnoses. Some cases showed discrepancies due to related symptoms, accounting for 11.11%, with more than one disease type within the system and having a confidence level below 70%.
Kombinasi Analytical Hierarchy Process dengan Weighted Product untuk Penerima Beasiswa Prestasi Sistem Pendukung Keputusan Fathony, Zamzam; Saputra, Eko Rachmat Slamet .H; Frobenius, Arvin Claudy
Jurnal Transformatika Vol. 21 No. 2 (2024): Januari 2024
Publisher : Jurusan Teknologi Informasi Universitas Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26623/transformatika.v21i2.6815

Abstract

Scholarships are an appreciation given by universities in the form of educational assistance, one of which is for prospective students who have achievements in non-academic fields. Mercu Buana University Jakarta provides as many as 20 quotas per year for merit scholarships. The decision-making process for merit scholarship recipients is still focused on manual calculation using the average value method process. Based on these problems, a study was conducted to design a decision support system using analytical hierarchy process and weighted product methods. Variables used in achievement weighting, level, test scores. The process of weighting the AHP method produces an achievement priority value of 0.260, a level of 0.633, and a test score of 0.106 and the results on the consistency criteria matrix are 0.033. The results of the WP ranking are the scores on the achievement criteria, namely -0.260, the level is 0.633 and the test score is 0.106. The results on the user acceptance test are 84.4%, it can be concluded, functionality can be accepted by users
Prediksi Kepuasan Mahasiswa Terhadap Pelayanan Akademik Menggunakan Model Decision Tree Zaman, Badroe; huizen, lenny margaretta; Ardima, Muhammad Basyier
Jurnal Transformatika Vol. 21 No. 2 (2024): Januari 2024
Publisher : Jurusan Teknologi Informasi Universitas Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26623/transformatika.v21i2.8214

Abstract

Perguruan Tinggi merupakan sebuah lembaga pendidikan dimana didalamnya mempunyai tugas dalam pelayanan akademik. Kepuasan mahasiswa dalam memperoleh pelayanan akademik  merupakan hal yang sangat penting dalam menilai sebuah Perguruan Tinggi. Tujuan dari penelitian ini adalah agar dapat mengetahui bagaimana tingkat kepuasan mahasiswa program studi Teknik Informatika dalam hal memperoleh pengajaran oleh dosen, mengenai sarana dan prasarananya. Metode klasifikasi dan prediksi yang digunakan pada penelitian ini diambil dari salah satu model Decision Tree yaitu algoritma C4.5. Algoritma C4.5 berfungsi untuk mengekspolari data, menemukan hubungan tersembunyi antara sejumlah calon variabel input dengan sebuah variabel target. Hasil pengukuran yang didapat adalah nilai akurasi sebesar 94,23%. Nilai recall dari setiap kelas sebesar 94,12% untuk kelas Ya dan 100% untuk kelas Tidak. Sedangkan nilai presisi setiap kelas adalah sebesar 100% untuk kelas Ya dan 25% untuk kelas Tidak.
Identifikasi Penyakit Jantung Menggunakan Machine Learning: Studi Komparatif sintiya, endah septa; Rizdania, Rizdania; Afrah, Ashri Shabrina
Jurnal Transformatika Vol. 21 No. 2 (2024): Januari 2024
Publisher : Jurusan Teknologi Informasi Universitas Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26623/transformatika.v21i2.7144

Abstract

Heart disease is the number one cause of death globally. This condition is followed by an unhealthy lifestyle. Heart disease prediction needs to be done considering the importance of health. The presence of machine learning has made it easier for humans to make early detection of patterns approaching heart disease. This study compares 6 machine learning methods for disease classification with KNN, Naïve Bayes, Decision tree, Random forest, logistic regression, and SVM. The final classification obtained ranking accuracy with the highest value in the KNN method with precision, accuracy, re-call, fi-score tests. It is hoped that these results can be applied to real case studies of heart disease.
An Examination of Negative Correlations Using Pearson Correlation Analysis to Optimize the Diversification of Cryptocurrency Portfolios Widodo, Edi Widodo,; Rahmawati, Eka Putri; Bilqist, Chay Shona
Jurnal Transformatika Vol. 21 No. 2 (2024): Januari 2024
Publisher : Jurusan Teknologi Informasi Universitas Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26623/transformatika.v21i2.8095

Abstract

The purpose of this study is to employ the Pearson correlation approach in order to assess the association between different types of cryptocurrencies. The dataset included in this research comprises daily peak price information for 10 distinct categories of cryptocurrencies with the biggest market capitalizations from October 1, 2017 to December 31, 2022. Assessing and computing the correlation between cryptocurrency pairs with the Pearson correlation coefficient is the objective. The information utilized in this study was acquired from the website www.coinmarketcap.com. Pairs of stablecoins and crypto coin assets have the largest negative correlation, according to the findings of this study, in contrast to pairs of crypto currency assets. The pair ETH-BNB has the strongest positive correlation with a value of 0.948, while the pair LTC-USDT has the most negative correlation at -0.347. In order to replicate the impact of the negative correlation on trading activities, an exchange simulation was performed between the LTC and USDT pairings. Based on the outcomes of the simulation, the asset rise resulting from the exchange of the LTC and USDT pair from January 1, 2022 to December 31, 2022 was 12.09 percent. During the same time period, the asset's value would have declined by -48.69 percent if LTC was held. Conversely, an expansion of the time period from October 1, 2017 to December 31, 2022 yields an asset gain of 251,047.85 percent as a consequence of the exchange between LTC and USDT. Those individuals interested in reducing risk and diversifying their portfolios with cryptocurrency investments may find this information highly beneficial. The results of this research offer significant contributions to the current body of literature on bitcoin investment and offer investors valuable information
Sistem Pendukung Keputusan Penyesuaian Nutrisi Makanan Berdasar Rekam Medis Pasien Berbasis Forward Chaining Setiawan, Aries; Setijaningsih, Retno Astuti; Ratnawati, Juli; Agiwahyuanto, Faik; Farida, Ida; Ashari, Ayu; Prasetya, Jaka
Jurnal Transformatika Vol. 21 No. 2 (2024): Januari 2024
Publisher : Jurusan Teknologi Informasi Universitas Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26623/transformatika.v21i2.8211

Abstract

Fulfilling nutrition for patients certainly does not only pay attention to the last illness they suffered, but nutritionists also need to pay attention to the patient's medical history. Providing certain foods to support the body's recovery after treatment for certain diseases may not necessarily be in accordance with the history of previous illnesses. Fulfilling nutritional intake according to certain disease conditions is not easy, especially if the patient has a medical history with a variety of diseases, so nutritionists need to be more selective in providing nutritional intake from a number of alternative foods that will be provided. A management decision system based on artificial intelligence is able to choose a food balance that is balanced with the various complaints experienced by patients. The method used in the food management information system for medical records uses the forward chaining method, namely by determining forward, in this case, food nutritional information that is suitable for the patient, by reading the facts that have been arranged as a representation to produce a conclusion. The accuracy value resulting from comparing manual nutrient selection and using forward chaining was 86%
Analisa Forensik Kontainer Podman Terhadap Backdoor Metasploit Menggunakan Checkpointctl Sya'bani, Hafiidh Akbar; Umam, Chaerul; Handoko, L Budi
Jurnal Transformatika Vol. 21 No. 2 (2024): Januari 2024
Publisher : Jurusan Teknologi Informasi Universitas Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26623/transformatika.v21i2.8109

Abstract

Container systems are type of virtualization technology with isolated environment. The isolated environment in container system does not make cyber attacks impossible to occur. In this research, containers in which a cyber incident occurred were forensically tested on the container's memory to obtain digital evidence. The forensic process is carried out using standards from NIST framework with the stages of collection, examination, analysis and reporting. The forensic process begins by performing a checkpoint on the container to obtain information from the container's memory. In Podman the checkpoint process is carried out on one of the containers and will produce a file in .tar.gz form, where this file contains the information contained in the container. After the checkpoint process is complete, forensics is then carried out by reading the checkpoint file using a tool called checkpointctl. Forensic results showed that the container was running a malicious program in the form of a backdoor with a PHP extension.

Page 1 of 2 | Total Record : 14

 1   2 

Filter by Year

2024 2024


Reset
Filter By Issues
All Issue Vol. 23 No. 2 (2026): January 2026 Vol. 23 No. 1 (2025): July 2025 Vol. 22 No. 2 (2025): January 2025 Vol. 22 No. 1 (2024): July 2024 Vol. 21 No. 2 (2024): Januari 2024 Vol 21, No 1 (2023): July 2023 Vol. 20 No. 2 (2023): January 2023 Vol. 20 No. 1 (2022): July 2022 Vol. 19 No. 2 (2022): January 2022 Vol. 19 No. 1 (2021): July 2021 Vol. 18 No. 2 (2021): January, 2021 Vol. 18 No. 1 (2020): July 2020 Vol. 17 No. 2 (2020): January 2020 Vol 17, No 1 (2019): July 2019 Vol. 17 No. 1 (2019): July 2019 Vol 16, No 2 (2019): January 2019 Vol 16, No 1 (2018): July 2018 Vol 15, No 2 (2018): January 2018 Vol 15, No 1 (2017): July 2017 Vol 14, No 2 (2017): January 2017 Vol 14, No 1 (2016): July 2016 Vol 13, No 2 (2016): January 2016 Vol 13, No 2 (2016) Vol 13, No 1 (2015): July 2015 Vol 12, No 2 (2015): January 2015 Vol 12, No 1 (2014): July 2014 Vol 11, No 2 (2014): January 2014 Vol 11, No 1 (2013): July 2013 Vol 10, No 2 (2013): January 2013 Vol 10, No 1 (2012): July 2012 Vol 9, No 2 (2012): January 2012 Vol 9, No 1 (2011): July 2011 Vol 8, No 2 (2011): January 2011 Vol 8, No 2 (2011) Vol 8, No 1 (2010): July 2010 Vol 7, No 2 (2010): January 2010 Vol 7, No 1 (2009): July 2009 Vol 6, No 2 (2009): January 2009 Vol 6, No 2 (2009) Vol 6, No 1 (2008): July 2008 Vol 5, No 2 (2008): January 2008 Vol 5, No 1 (2007): July 2007 Vol 4, No 2 (2007): January 2007 Vol 4, No 1 (2006): July 2006 Vol 2, No 2 (2005): January 2005 Vol 2, No 1 (2004): July 2004 Vol 1, No 2 (2004): January 2004 Vol 2, No 1 (2004) More Issue