Agi Prasetiadi
Institut Teknologi Telkom Purwokerto

Published : 5 Documents Claim Missing Document
Claim Missing Document

Found 5 Documents

Navigating Bitcoin Panic-Selling using Linear Approach Agi Prasetiadi
JURNAL INFOTEL Vol 12 No 4 (2020): November 2020

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (1192.126 KB) | DOI: 10.20895/infotel.v12i4.543


COVID-19 affects significant human activity around the globe, including Bitcoin prices. The Bitcoin price is well known for its volatility, so it is not a big shocker when the panic-selling occurs during the pandemic. However, the mechanism to cope with these breakouts, especially the bearish one, is contentious. The experts give numerous pieces of advice with different conclusions in the end. It is also the same with Machine Learning. Various kernels show different results regarding how the price will move. It depends on the window size, how the data is being preprocessed, and the algorithm used. This paper inspects the best combination that various machine learning can offer with a linear approach to navigate the price prediction based on its depth interval, window size until the algorithms themselves. This paper also proposed a new approach to seeing the prediction range called s-steps ahead prediction using a linear model. The result shows that simple machine learning can herd 99.715% profit even during the bearish breakout.
Classification Based on Configuration Objects by Using Procrustes Analysis Ridho Ananda; Agi Prasetiadi
JURNAL INFOTEL Vol 13 No 2 (2021): May 2021

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (707.825 KB) | DOI: 10.20895/infotel.v13i2.637


Classification is one of the data mining topics that will predict an object to go into a certain group. The prediction process can be performed by using similarity measures, classification trees, or regression. On the other hand, Procrustes refers to a technique of matching two configurations that have been implemented for outlier detection. Based on the result, Procrustes has a potential to tackle the misclassification problem when the outliers are assumed as the misclassified object. Therefore, the Procrustes classification algorithm (PrCA) and Procrustes nearest neighbor classification algorithm (PNNCA) were proposed in this paper. The results of those algorithms had been compared to the classical classification algorithms, namely k-Nearest Neighbor (k-NN), Support Vector Machine (SVM), AdaBoost (AB), Random Forest (RF), Logistic Regression (LR), and Ridge Regression (RR). The data used were iris, cancer, liver, seeds, and wine dataset. The minimum and maximum accuracy values obtained by the PrCA algorithm were 0.610 and 0.925, while the PNNCA were 0.610 and 0.963. PrCA was generally better than k-NN, SVM, and AB. Meanwhile, PNNCA was generally better than k-NN, SVM, AB, and RF. Based on the results, PrCA and PNNCA certainly deserve to be proposed as a new approach in the classification process.
JATISI (Jurnal Teknik Informatika dan Sistem Informasi) Vol 7 No 3 (2020): JATISI (Jurnal Teknik Informatika dan Sistem Informasi)
Publisher : Lembaga Penelitian dan Pengabdian pada Masyarakat (LPPM) STMIK Global Informatika MDP

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35957/jatisi.v7i3.461


Voice is one of the characteristics possessed by humans. One of the human abilities to identify someone by seeing and hearing. Through this ability one can easily distinguish gender, based on physical and voice categories. voices that previously can be recognized easily by humans, for now can also be recognized by computers, which aims to separate sounds based on gender. The final result of this research is to know the average error of data objects that have been inputted with the algorithm used in this study is the Fast Fourier Transform. The results of the implementation process of the Fast Fourier Transform Algorithm show whether or not the average sound error is influenced by the high or low amplitude of the sound being tested and from the results obtained the average error amplitude of the female voice is lower than the average error amplitude. Male voice. Of the two sources of human voices used, the lowest average error was produced in the 7th cluster for Female 3 votes with an average error result of 132.840, while the results of trials conducted with 3 sound sources resulted in the lowest average error in the second cluster. -3 for Female 3 votes with an average error result of 212,976, and for the four sound sources used it produces the lowest average error in the 3rd cluster on Mela's voice with an average error result of 217,462.
Identifikasi Spesies Reptil Menggunakan Convolutional Neural Network (CNN) Olvy Diaz Annesa; Condro Kartiko; Agi Prasetiadi
Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi) Vol 4 No 5 (2020): Oktober 2020
Publisher : Ikatan Ahli Informatika Indonesia (IAII)

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (1430.695 KB) | DOI: 10.29207/resti.v4i5.2282


Reptiles are one of the most common fauna in the territory of Indonesia. quite a lot of people who have an interest in knowing more about this fauna in order to increase knowledge. Based on previous research, Deep Learning is needed in particular the CNN method for computer programs to identify reptile species through images. This reseacrh aims to determine the right model in producing high accuracy in the identification of reptile species. Thousands of images are generated through data augmentation processes for manually captured images. Using the Python programming language and Dropout technique, an accuracy of 93% was obtained by this research in identifying 14 different types of reptiles.
Perbandingan Performa Antara Algoritma Naive Bayes Dan K-Nearest Neighbour Pada Klasifikasi Kanker Payudara Annisa Nugraheni; Rima Dias Ramadhani; Amalia Beladinna Arifa; Agi Prasetiadi
Indonesian Journal of Data Science, IoT, Machine Learning and Informatics Vol 2 No 1 (2022): February
Publisher : Research Group of Data Engineering, Faculty of Informatics

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.20895/dinda.v2i1.391


Breast cancer is the second most common cause of death from cancer after lung cancer is in the first place. Breast cancer occurs when cells in breast tissue begin to grow uncontrollably and can disrupt existing healthy tissue. Therefore, there is a need for a classification to distinguish breast cancer patients and healthy people. Based on previous research, the Naïve Bayes and K-Nearest Neighbor algorithms are considered capable of classifying breast cancer. In the research process using the breast cancer dataset from the Breast Cancer Coimbra dataset in 2018 UCI Machine Learning Repository with a total of 116 data, while for the calculation of the feasibility of the method using the Confusion Matrix (Accuracy, Precision, and Recall) and the ROC-AUC curve. The purpose of this study is to compare the performance of the Naïve Bayes and K-Nearest Neighbor algorithms. In testing using the Naïve Bayes algorithm and the K-Nearest Neighbor algorithm, there are several test scenarios, namely, data testing before and after normalization, model testing based on a comparison of training data and testing data, model testing based on K values ​​in K-Nearest Neighbors, and model testing. based on the selection of the strongest attribute with the Pearson correlation test. The results of this study indicate that the Naïve Bayes algorithm has the highest average accuracy of 69.12%, healthy precision 64.90%, pain precision 83%, healthy recall 88%, sick recall 61.11% and AUC 0.82 which is included in the good classification category. Meanwhile, the highest average results of the K-Nearest Neighbor algorithm are 76.83% for accuracy, 76% healthy precision, 80.21% pain precision, 74.18% for healthy recall, 80.81% sick recall and 0.91 AUC which is included in the excellent classification category.