Murat Bayraktar

I joined an industrial research project focused on developing an Energy optimization toolchain. I was responsible for designing a pipeline that pre-processes and detects anomalies by applying unsupervised machine learning algorithms such as (Isolation Forest/DBSCAN/Agglomerative clustering). Also, I tried other approaches by combining LSTM cells with an AutoEncoder neural network learning from synthesised labels.Lastly, I deployed the app with Streamlit UI to make the whole prediction process user friendly