search

Niketkumardheeryan / ML-CaPsule

ML-capsule is a Project for beginners and experienced data science Enthusiasts who don't have a mentor or guidance and wish to learn Machine learning. Using our repo they can learn ML, DL, and many related technologies with different real-world projects and become Interview ready.
MIT License
382 stars 318 forks source link

Data Preprocessing Enhancements for Improved Model Performance #609

Open FreeSpirit11 opened 4 months ago

FreeSpirit11 commented 4 months ago

Description

To improve the performance and accuracy of the Bitcoin price prediction model, the following enhancements are proposed for the data preprocessing step:

  1. Use More Features:

    • Include additional relevant features such as trading volume, moving averages, or technical indicators (e.g., RSI, MACD).

  2. Feature Engineering:

    • Create new features such as percentage changes, rolling averages, or volatility measures to capture more information relevant to the prediction task.

  3. Data Normalization:

    • Ensure all features are properly scaled before training to improve model convergence and performance.

Implementation Details

  1. Include More Features:

    • Add features like Volume, MA7, MA14, MA21, RSI, and MACD.

  2. Feature Engineering:

    • Calculate and add percentage changes for the Close price.
    • Compute rolling averages and volatility measures over different windows (e.g., 7-day, 14-day).

  3. Data Normalization:

    • Use MinMaxScaler or StandardScaler from sklearn to normalize all features.
FreeSpirit11 commented 4 months ago

Hi @Niketkumardheeryan , I have raised this issue. Please assign it to me under GSSOC.

Niketkumardheeryan commented 4 months ago

@FreeSpirit11 go for it

FreeSpirit11 commented 4 months ago

HI @Niketkumardheeryan , I have started working on this issue. Please assign it level 2 or level 3 .