DataScience101

Monday. November 26, 2018 - 1 min

We formed a group who are interested in Data Science and interested to get internship in the same field. We meet every evening in the library to work on Kaggle Competition. We finished working on Movie Sentiment Analysis. Currently, we are working on Quora insincere question analysis. We are doing rigorous (theoretical and practical) study behind the machine learning algorithms.

Kaggle problems :-

Titanic
Movie Sentiment Analysis
Quora insincere question classification

What’s inside :-

Text Cleaning
Dataset preprocessing
Vector representation (TF-IDF, word2vec)
Traditional machine learning algorithms:
- Logistic Regression
- kNN
- SVM
- Naive Bayes
- Ensemble Learning (Gradient Boosting, Bagging)
- Perceptron
- MLP
Deep Learning
- RNN
- LSTM
- GRU
Data visualization
- ROC AUC curve
- Dataset distribution
- Word Cloud
- TSNE plot

Check out the project here.