<![CDATA[The temperature sensor in an air conditioning (AC) system plays a key role in maintaining performance and efficiency when the cooling function is activated. However, this sensor can sometimes fail or experience problems, leading to inaccurate temperature readings and, consequently, improper system operation. Therefore, having an early detection mechanism is very important to identify potential issues. This study attempts to apply the Local Outlier Factor (LOF) algorithm to improve anomaly detection capabilities in AC temperature sensor data. LOF is a density-based method that identifies data points with density significantly different from their neighbors. The dataset used in this study consists of temperature readings collected from AC sensors over a specific period. The process includes data preprocessing, LOF score calculation, and evaluation using metrics such as accuracy, precision, recall, and F1-score. The results show that LOF can detect anomalies with fairly high accuracy, making it a promising solution for automatic AC monitoring systems, allowing faulty sensors to be detected without the need for manual inspection.]]>
Copyrights © 2026
