Java Deep Learning Projects

Titanic survival prediction using MLP and LSTM networks

In this project, our main goal was to get familiar with Apache Spark ML library, followed by a basic introduction to machine learning, deep learning their types, architectures, and frameworks.

We could not achieve higher accuracy. Then, in Chapter 2, Cancer Types Prediction Using Recurrent Type Networks, we revisited the same project but using a robust recurrent LSTM network, which shows higher accuracy. The takeaway was learning how to prepare the dataset by considering most of the features and feeding into Spark and a DL4J-based MLP classifier.

In addition, this dataset is not so high dimensional, so applying DL methods is not a good idea. Therefore, I would recommend using other tree ...