Cracking Kaggle Competitions

• This repository contains lots of Data Cleaning, Feature Engineering, Exploratory Data Analysis(EDA), and Modeling techniques for Classical Machine Learning competitions on Kaggle. It also contains Various Text Processing techniques, Different uses and architectures of RNN using Pytorch and other approaches for NLP competitions on Kaggle.

• Each competition has its own approach to solve it to get the best results.

Advice

If you want to learn something new/advanced i advice you to have a look on this competition:

• Used Cars Price Prediction (https://www.kaggle.com/avikasliwal/used-cars-price-prediction) I believe it shows my new skills.

Competitions in progress

Classification

• None.

Regression

• None.

NLP

• None.

Cracked Competitions

Classification

• Titanc Disaster (https://www.kaggle.com/c/titanic) Score: 77% (Higher is better)

  • Kaggle Notebook: (https://www.kaggle.com/kickitlikeshika/titanic-disaster)

• Forest Cover Type (https://www.kaggle.com/c/forest-cover-type-prediction) Score: 79% (Higher is better)

  • Kaggle Notebook: (https://www.kaggle.com/kickitlikeshika/extra-trees-79-accuracy)

• IEEE-CIS Fraud Detection (https://www.kaggle.com/c/ieee-fraud-detection) Score: 92% (Higher is better)

  • Kaggle Notebook: (https://www.kaggle.com/kickitlikeshika/fraud-detection-with-92-acc)

• Instant Gratification (https://www.kaggle.com/c/instant-gratification) No Score.

  • Kaggle Notebook: None.

• Categorical Feature Encoding (https://www.kaggle.com/c/cat-in-the-dat) Score: 76% (Higher is better)

  • Kaggle Notebook: (https://www.kaggle.com/kickitlikeshika/easy-76)

• Amazon Employee Access Challenge (https://www.kaggle.com/c/amazon-employee-access-challenge/overview) Score: 89% (Higher is better)

  • Kaggle Notebook: (https://www.kaggle.com/kickitlikeshika/employee-access-eda-data-cleaning)

Regression

• House Prices Prediction (https://www.kaggle.com/c/house-prices-advanced-regression-techniques) Score: 0.12 (Lower is better)

  • Kaggle Notebook: (https://www.kaggle.com/kickitlikeshika/easy-and-effective-approach-to-predict-house-price)

• Fish Market (https://www.kaggle.com/aungpyaeap/fish-market) No Score.

  • Kaggle Notebook: (https://www.kaggle.com/kickitlikeshika/fish-market-prediction)

• Bike Sharing Demand (https://www.kaggle.com/c/bike-sharing-demand) Score: 0.417 (Lower is better)

  • Kaggle Notebook: (https://www.kaggle.com/kickitlikeshika/data-exploration-with-41)

• TMDB Box Office Prediction (https://www.kaggle.com/c/tmdb-box-office-prediction) Score: 2.127 (Lower is better)

  • Kaggle Notebook: (https://www.kaggle.com/kickitlikeshika/advanced-eda-feature-engineering)

• Used Cars Price Prediction (https://www.kaggle.com/avikasliwal/used-cars-price-prediction) No Score.

  • Kaggle Notebook: (https://www.kaggle.com/kickitlikeshika/advanced-techniques-to-get-the-best-accuracy)

• Tabular Palyground Series Feb 2021 (https://www.kaggle.com/c/tabular-playground-series-feb-2021) Score: 0.84235 (Lower is better)

NLP

• Quora Insincere Questions Classification (https://www.kaggle.com/c/quora-insincere-questions-classification). Score: .70 (Higher is better)

  • Kaggle Notebook: None.

• Natural Language Processing with Disaster Tweets (https://www.kaggle.com/c/nlp-getting-started). Score: .798 (Higher is better)

  • Kaggle Notebook: None.

• Twitter Tweets Data (https://www.kaggle.com/saadbinmanjuradit/twitter-tweets-data). No Score.

  • Kaggle Notebook: None.

• CommonLit readability Prize (https://www.kaggle.com/c/commonlitreadabilityprize) (top 9%) Score: 0.460 (Lower is better)

  • Kaggle Notebook: None.

• Sentiment Analysis on Movie Reviews: https://www.kaggle.com/c/sentiment-analysis-on-movie-reviews (Top 1%, 5th place) Score: 0.69644 (Higher is better)

  • Kaggle Notebook: https://www.kaggle.com/kickitlikeshika/bert-for-sentiment-analysis-5th-place-solution

• Google QUEST Q&A Labeling: https://www.kaggle.com/c/google-quest-challenge

  • Kaggle Notebook: None.

Hints

• You can get the data/csv files from the links of the competitions above.