Data-Science
1. Machine Learning
A kaggle student dataset is chosen and all the following concepts of machine learning are covered,
Data preprocessing, Data splitting, Data imbalance, Feature reduction using PCA, Unsupervised learning of K-means clustering, Supervised learning of Logistic regression, Ensemble methods of Random forest, GRID search hyperparameter tuning.
2. Deep Learning
An IMDB Dataset is chosen and the following concepts are covered,
Sentimental classification using neural networks of different types such as SimpleRNN, LSTM, Convolution neural network, Transfer learning in neural net. Tensorflow-keras is used to cover all the above concepts and it was found that 2 layer LSTM variant had better performance than the rest and transfer learning process helped modelling using less volume datasets.
3. R Core programming
All the concepts and syntaxes of R programming are practised here,
Datatypes, Vectors, Matrix, Dataframe - Dplyr, R-SQL, sourcing files and URLs, Webscrapping Visualisations using ggplot, usage of statistical functions, Performing basic linear & logistic regression, K-means clustering.