The strategy involves the utilization of historical data spanning five years prior to the target election year. This is based on the premise that past patterns and trends within a reasonable timeframe may hold predictive power for future events, such as election outcomes.
Specifically, we aim to use security-related indicators, in conjunction with election data, to train a predictive model. The security indicators include diverse categories such as acts of vandalism, environmental offenses, anti-LGBT
flowchart TB
subgraph "Data Collection & Pre-processing"
Collect[Collect 5 years of security data] --> Preprocess[Preprocess & Clean Data]
end
subgraph "Model Building"
Preprocess --> Train[Train Predictive Model on 2017 Election Data]
Train --> Validate[Validate & Tune Model]
end
subgraph "Prediction & Evaluation"
Validate --> Predict[Predict 2022 Election Results]
Predict --> Evaluate[Evaluate Model Performance]
end