Pradnya Dhuhita, Windha Mega
Unknown Affiliation

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search

MobileNet V2 Implementation in Skin Cancer Detection Pradnya Dhuhita, Windha Mega; Ubaid, Muhammad Yahya; Baita, Anna
ILKOM Jurnal Ilmiah Vol 15, No 3 (2023)
Publisher : Prodi Teknik Informatika FIK Universitas Muslim Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33096/ilkom.v15i3.1702.498-506

Abstract

Skin cancer is one of the most worrying diseases for humans. In Indonesia alone, skin cancer occupies the third position after cervical cancer and breast cancer. Currently, doctors still use the biopsy method to diagnose skin cancer. It is less effective because this method requires the performance of an experienced doctor, takes a long time, and is a painful process. Because of that, we need a way in which skin cancer can be classified using dermoscopic images to help doctors diagnose skin cancer earlier. Researchers proposed to classify skin cancer into seven classes, namely actinic keratoses, basal cell carcinoma, benign keratosis-like lesions, dermatofibroma, melanoma, melanocytic nevus, and vascular lesions. The method used in this study is a convolutional neural network (CNN) with the MobileNet V2 architecture. The dataset used is the HAM10000 dataset, with a total of 10015 images. In this study, a comparison was made between data augmentation, learning rate, epochs, and different amounts of data. Based on the test results, the highest accuracy results were obtained, namely 79%. The best model is implemented into a mobile application.
Analysis Of Lstm-Adamax Performance In Bitcoin Price Prediction Using RSI & MACD Indicators Pradnya Dhuhita, Windha Mega; Sumbadri, Yoga Imam
The Indonesian Journal of Computer Science Research Vol. 5 No. 1 (2026): Januari
Publisher : Hemispheres Press

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59095/ijcsr.v5i1.247

Abstract

One digital asset that is difficult to predict due to its extreme volatility is Bitcoin. Bitcoin's value has been predicted using various techniques, ranging from technical analysis to artificial intelligence-based models. Long Short-Term Memory (LSTM) is an artificial neural network architecture capable of recognizing patterns in historical data and is often used for time series data prediction. This study explores the application of an optimized LSTM model with the Adamax algorithm combined with the technical indicators RSI (Relative Strength Index) and MACD (Moving Average Convergence Divergence) to predict Bitcoin prices based on historical data. The LSTM-Adamax model demonstrated strong performance, achieving an RMSE of 435.9, MAE of 284.5, and R² of 0.99947, indicating high accuracy and robustness in capturing price patterns. A comparative evaluation between the model with and without these indicators revealed a slight performance improvement when the technical indicators were used. The model was successfully implemented as a web application using Streamlit, allowing users to upload Bitcoin price data, configure prediction parameters, and visualize the results in real-time. The application also communicates the predicted price movement direction (up or down) and its magnitude. In conclusion, the integration of LSTM-Adamax with RSI and MACD proved effective in predicting Bitcoin prices based on time-series data, providing reliable predictions and user-friendly implementation through a web interface.