Spaces:
Running
Running
File size: 4,184 Bytes
54c8522 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 | """
Analyze PurposeOfHearing patterns to identify ripeness indicators.
This script examines the historical hearing data to classify purposes
as RIPE (ready for hearing) vs UNRIPE (bottleneck exists).
"""
import polars as pl
from pathlib import Path
# Load hearing data
hear_df = pl.read_csv("Data/ISDMHack_Hear.csv")
print("=" * 80)
print("PURPOSEOFHEARING ANALYSIS FOR RIPENESS CLASSIFICATION")
print("=" * 80)
# 1. Unique values and frequency
print("\nPurposeOfHearing Frequency Distribution:")
print("-" * 80)
purpose_counts = hear_df.group_by("PurposeOfHearing").count().sort("count", descending=True)
print(purpose_counts.head(30))
print(f"\nTotal unique purposes: {hear_df['PurposeOfHearing'].n_unique()}")
print(f"Total hearings: {len(hear_df)}")
# 2. Map to Remappedstages (consolidation)
print("\n" + "=" * 80)
print("PURPOSEOFHEARING → REMAPPEDSTAGES MAPPING")
print("=" * 80)
# Group by both to see relationship
mapping = (
hear_df
.group_by(["PurposeOfHearing", "Remappedstages"])
.count()
.sort("count", descending=True)
)
print(mapping.head(40))
# 3. Identify potential bottleneck indicators
print("\n" + "=" * 80)
print("RIPENESS CLASSIFICATION HEURISTICS")
print("=" * 80)
# Keywords suggesting unripe status
unripe_keywords = ["SUMMONS", "NOTICE", "ISSUE", "SERVICE", "STAY", "PENDING"]
ripe_keywords = ["ARGUMENTS", "HEARING", "FINAL", "JUDGMENT", "ORDERS", "DISPOSAL"]
# Classify purposes
def classify_purpose(purpose_str):
if purpose_str is None or purpose_str == "NA":
return "UNKNOWN"
purpose_upper = purpose_str.upper()
# Check unripe keywords first (more specific)
for keyword in unripe_keywords:
if keyword in purpose_upper:
return "UNRIPE"
# Check ripe keywords
for keyword in ripe_keywords:
if keyword in purpose_upper:
return "RIPE"
# Default
return "CONDITIONAL"
# Apply classification
purpose_with_classification = (
purpose_counts
.with_columns(
pl.col("PurposeOfHearing")
.map_elements(classify_purpose, return_dtype=pl.Utf8)
.alias("Ripeness_Classification")
)
)
print("\nPurpose Classification Summary:")
print("-" * 80)
print(purpose_with_classification.head(40))
# Summary stats
print("\n" + "=" * 80)
print("RIPENESS CLASSIFICATION SUMMARY")
print("=" * 80)
classification_summary = (
purpose_with_classification
.group_by("Ripeness_Classification")
.agg([
pl.col("count").sum().alias("total_hearings"),
pl.col("PurposeOfHearing").count().alias("num_purposes")
])
.with_columns(
(pl.col("total_hearings") / pl.col("total_hearings").sum() * 100)
.round(2)
.alias("percentage")
)
)
print(classification_summary)
# 4. Analyze by stage
print("\n" + "=" * 80)
print("RIPENESS BY STAGE")
print("=" * 80)
stage_purpose_analysis = (
hear_df
.filter(pl.col("Remappedstages").is_not_null())
.filter(pl.col("Remappedstages") != "NA")
.group_by(["Remappedstages", "PurposeOfHearing"])
.count()
.sort("count", descending=True)
)
print("\nTop Purpose-Stage combinations:")
print(stage_purpose_analysis.head(30))
# 5. Export classification mapping
output_path = Path("reports/ripeness_purpose_mapping.csv")
output_path.parent.mkdir(exist_ok=True)
purpose_with_classification.write_csv(output_path)
print(f"\n✓ Classification mapping saved to: {output_path}")
print("\n" + "=" * 80)
print("RECOMMENDATIONS FOR RIPENESS CLASSIFIER")
print("=" * 80)
print("""
Based on the analysis:
UNRIPE (Bottleneck exists):
- Purposes containing: SUMMONS, NOTICE, ISSUE, SERVICE, STAY, PENDING
- Cases waiting for procedural steps before substantive hearing
RIPE (Ready for hearing):
- Purposes containing: ARGUMENTS, HEARING, FINAL, JUDGMENT, ORDERS, DISPOSAL
- Cases ready for substantive judicial action
CONDITIONAL:
- Other purposes that may be ripe or unripe depending on context
- Needs additional logic based on stage, case age, hearing count
Use Remappedstages as secondary indicator:
- ADMISSION stage → more likely unripe (procedural)
- ORDERS/JUDGMENT stage → more likely ripe (substantive)
""")
|