Data Science

All about data science

StakeHolders

• 1. Data Scientist

     1. Prove/disprove hypothesis.

     2. Information & Data Gathering

     3. Data debating.

     4. Algorithms & ML Models.

     5. Communication.

1. Data Engineer

   1. 1. Build Data Driven Platforms.

      2. Operationlize Algorithms & Machine Learning models.

      3. Data Integration.

      4. Monitoring.

2. Data Visualization

• • 1. Story Telling

    2. Build Dashboard & Other Data Visualizations.

    3. Provide insight through visual means.

3. Process Owner

• 1. Project Management

  2. Manage StakeHolder Expectations.

  3. Maintain a Vision.

  4. Facilitate.

  5. Evengilize.

What is Data Science

• 1. require Math & Statistics

  2. is Hypothesis Driven

  3. require a range of competencies/roles.

Machine Learning Types

• 1. Supervised Learning

     1. Model trained on historical data. Resulting model can be used to predict to make predictions on new data.

     2. Use Case: Predicting a value based on patterns discovered in previous data.

     3. Regression

        1. Linear Regression

     4. Classification

• • • 1. Decision Trees

• 2. Unsupervised Learning

• • 1. Algorithms finds trends and patterns in data, without prior training on historical data.

    2. Use Case: Describing your data based on statistical analysis.

    3. Clustering

       1. K Means

• 3. Reinforcement Learning

• • 1. Model uses a feedback loop to iteratively improve its performance.

    2. Use Case: Learning how to best solve a problem based on trial and error. Often used in game programming.

References

1. GOTO 2020 A Practical-ish Introduction to Data Science - Mark West