Portfolio

Natural Language Processing

Final Project: News Articles Classification Using LSTM Method

News Classification: An Indonesian news classification system using LSTM method trained with 10 thousand news articles consisting of 5 categories (Entertainment, Sports, Showbiz, Technology, and Headlines).

Model Deploypment: LSTM for News Classification

After I completed the modeling process for classifying news articles, I refined the model by implementing a deployment model using the Python library Flask.

Cyberbullying Detection On Instagram: TF-IDF and Machine Learning Models

Problem Statement: Cyberbullying is a problem often faced by internet users, especially children and adolescents. Cyberbullying is an act of violence committed online, through media such as text messages, social media, or chat applications. It can cause serious consequences for victims, such as emotional distress, stress, and even suicide.

Gools: Using machine learning techniques to automatically identify and manage cyberbullying content.

Email Spam Detection: TF-IDF and Machine Learning Models

To predict whether a message is spam, I first vectorize the text message into a format that a machine learning algorithm can understand using TF-IDF. Then I trained a machine learning model consisting of Naive Bayes and Random Forest to learn to distinguish between normal and spam messages.

Data Science

DQLab in Telco: Churn Customer Prediction

After cleaning the data, it is time to create a model that can predict customer churn, so as to reduce the number of customers who switch (churn) using machine learning algorithms. Finally, with the trained model, I can identify customers who are likely to churn.