metadata
dataset_info:
features:
- name: date
dtype: string
- name: total_claims
dtype: int32
- name: claims_approved
dtype: int32
- name: claims_rejected
dtype: int32
- name: total_premium_collected
dtype: float64
- name: total_claims_paid
dtype: float64
- name: loss_ratio
dtype: float64
- name: combined_ratio
dtype: float64
- name: fraud_detected
dtype: int32
- name: fraud_amount_saved
dtype: float64
- name: avg_processing_time_hours
dtype: float64
- name: customer_satisfaction_score
dtype: float64
- name: new_policies
dtype: int32
- name: policy_renewals
dtype: int32
- name: country
dtype: string
splits:
- name: train
num_bytes: 104140
num_examples: 1000
download_size: 50916
dataset_size: 104140
configs:
- config_name: default
data_files:
- split: train
path: data/train-*