Stock Prediction GOGOGO

• 

  Detect Outliers

Inspiration

Super interested in Machine Learning and data analysis

What it does

Forecasting stock growth

How we built it

• Analyse the dataset first
• Go with baseline code
• Try grouping features
• Feature Engineering by each team member
• Feature selection
• Model selection
• Grid search for tuning parameters

Challenges we ran into

• Analyse baseline code
• New ideas but hard to implement
• Sleep deprivation :[

Accomplishments that we're proud of

• Tried different models on a real-life problem
• Got a relatively good improvement for the baseline

What we learned

• Have a deeper understanding of data analysis
• Various classification machine learning algorithms for Stock Prediction

What's next for Stock Prediction GOGOGO

• Digging deeper into relationships from raw data
• Neural Networks
• For those tested models, find different ways to tune parameters

_More details are on Github

Built With

• machine-learning
• prediction
• python
• sklearn