Predictive_Maintenance_RandomForestClassifier, Esquemas y mapas conceptuales de Diagnóstico

¡Descarga Predictive_Maintenance_RandomForestClassifier y más Esquemas y mapas conceptuales en PDF de Diagnóstico solo en Docsity!

PREDICTIVE MAINTENANCE- RANDOMFORESTCLASSIFIER

KAROL STEFANY RANGEL GIRALDO

ING.LUIS HERNÁNDEZ CONTRERAS

UNIVERSIDAD DE CÓRDOBA

INGENIERÍA MECÁNICA

MONTERÍA

RandomForestClassifier is a machine learning model based on decision trees. It belongs to a group of models called ensemble methods, which means it combines the results of many individual trees to make a more accurate and reliable final prediction. Each tree is trained on a random sample of the data, and the model chooses the class that gets the most votes from the trees. This approach reduces the risk of overfitting, improves accuracy, and works well even with noisy data or many input features. Another advantage is that it does not require a lot of data preprocessing, making it a practical and powerful tool for classification tasks 1) Initial Data Analysis The dataset contained information from machines, including variables such as product type, air temperature, process temperature, rotational speed, torque, and tool wear. The goal of the analysis was to predict the failure type that a machine might experience during its operation. The target column, called “Failure Type,” included several possible categories: No Failure, Heat Dissipation Failure, Power Failure, Tool Wear Failure, Overstrain Failure, and Random Failures. After reviewing the dataset, I found that there were no missing values, and the data was ready to be processed. 2) Data Visualization To better understand the data, I created a graphic that showed how often each failure type appeared. Most of the data belonged to the “No Failure” category, while the other failure types were very rare. This indicated that the dataset was unbalanced. Since the model would see many examples of “No Failure” and very few of the others, it could have trouble learning how to recognize the rare failure types.