mcp-servers/nadlan-mcp
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Change default IQR multiplier to 1.0 for more aggressive outlier filtering

Browse Source 
- Change ANALYSIS_IQR_MULTIPLIER default from 1.5 to 1.0 in config.py
- Add iqr_multiplier parameter to all filtering & statistics functions
- Allow runtime override via MCP tools (get_valuation_comparables, get_deal_statistics)
- Update CLAUDE.md docs with new default & override examples

Rationale: k=1.0 catches more suspicious deals (e.g. 43% below median) while
still preserving legitimate edge cases via hard bounds. Users can override
per-call for more conservative filtering (k=1.5) if needed.

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>
This commit is contained in:
Nitzan Pomerantz
2025-11-28 22:40:29 +02:00
parent 1bc94e39ba
commit 8e591423da
6 changed files with 43 additions and 17 deletions
Download Patch File Download Diff File Expand all files Collapse all files
nadlan_mcp/config.py
+1 -1
View File
@@ -61,7 +61,7 @@ class GovmapConfig:
default_factory=lambda: os.getenv("ANALYSIS_OUTLIER_METHOD", "iqr")
)
analysis_iqr_multiplier: float = field(
default_factory=lambda: float(os.getenv("ANALYSIS_IQR_MULTIPLIER", "1.5"))
default_factory=lambda: float(os.getenv("ANALYSIS_IQR_MULTIPLIER", "1.0"))
)
analysis_min_deals_for_outlier_detection: int = field(
default_factory=lambda: int(os.getenv("ANALYSIS_MIN_DEALS_FOR_OUTLIER_DETECTION", "10"))
nadlan_mcp/fastmcp_server.py
+14 -5
View File
@@ -772,11 +772,12 @@ def get_valuation_comparables(
max_floor: Optional[int] = None,
radius_meters: int = 100,
max_comparables: int = 50,
iqr_multiplier: Optional[float] = None,
) -> str:
"""Get comparable properties for valuation analysis.
This tool provides detailed comparable deals filtered by your criteria.
Automatically applies IQR outlier filtering (k=1.5) to remove statistical outliers
Automatically applies IQR outlier filtering (k=1.0 default) to remove statistical outliers
and improve data quality. The response includes metadata about filtering so you can
inform users about removed outliers.
@@ -794,6 +795,7 @@ def get_valuation_comparables(
max_floor: Maximum floor number
radius_meters: Search radius in meters (default: 100, larger than find_recent_deals to get more comparables)
max_comparables: Maximum number of deals to return (default: 50, optimized for MCP token limits)
iqr_multiplier: Override IQR multiplier for outlier detection (default: 1.0). Lower = more aggressive filtering
Returns:
JSON string containing:
@@ -803,7 +805,7 @@ def get_valuation_comparables(
- total_deals_before_filtering: Count before filtering
- outliers_removed: Number of deals filtered out
- filtering_method: Method used (e.g., "iqr")
- iqr_multiplier: IQR multiplier used (e.g., 1.5)
- iqr_multiplier: IQR multiplier used (e.g., 1.0)
"""
log_mcp_call(
"get_valuation_comparables",
@@ -820,6 +822,7 @@ def get_valuation_comparables(
max_floor=max_floor,
radius_meters=radius_meters,
max_comparables=max_comparables,
iqr_multiplier=iqr_multiplier,
)
try:
# Get all deals for the address with higher limits for valuation
@@ -876,10 +879,13 @@ def get_valuation_comparables(
):
deals_before_outlier_filter = len(filtered_deals)
filtered_deals, outlier_report = filter_deals_for_analysis(
filtered_deals, config, metric="price_per_sqm"
filtered_deals, config, metric="price_per_sqm", iqr_multiplier=iqr_multiplier
)
effective_k = (
iqr_multiplier if iqr_multiplier is not None else config.analysis_iqr_multiplier
)
logger.info(
f"After outlier filtering ({config.analysis_outlier_method}, k={config.analysis_iqr_multiplier}): "
f"After outlier filtering ({config.analysis_outlier_method}, k={effective_k}): "
f"{len(filtered_deals)} deals (removed {deals_before_outlier_filter - len(filtered_deals)} outliers)"
)
else:
@@ -946,6 +952,7 @@ def get_deal_statistics(
property_type: Optional[str] = None,
min_rooms: Optional[float] = None,
max_rooms: Optional[float] = None,
iqr_multiplier: Optional[float] = None,
) -> str:
"""Calculate statistical aggregations on deal data for an address.
@@ -958,6 +965,7 @@ def get_deal_statistics(
property_type: Filter by property type (e.g., "דירה", "בית")
min_rooms: Minimum number of rooms
max_rooms: Maximum number of rooms
iqr_multiplier: Override IQR multiplier for outlier detection (default: 1.0). Lower = more aggressive filtering
Returns:
JSON string containing statistical metrics (mean, median, percentiles, etc.)
@@ -969,6 +977,7 @@ def get_deal_statistics(
property_type=property_type,
min_rooms=min_rooms,
max_rooms=max_rooms,
iqr_multiplier=iqr_multiplier,
)
try:
# Get all deals for the address
@@ -998,7 +1007,7 @@ def get_deal_statistics(
)
# Calculate statistics
stats = client.calculate_deal_statistics(deals)
stats = client.calculate_deal_statistics(deals, iqr_multiplier=iqr_multiplier)
# Normalize response structure to match other tools
return json.dumps(
nadlan_mcp/govmap/client.py
+5 -2
View File
@@ -873,7 +873,9 @@ class GovmapClient:
)
# Statistics methods (delegate to statistics module)
def calculate_deal_statistics(self, deals: List[Deal]) -> DealStatistics:
def calculate_deal_statistics(
self, deals: List[Deal], iqr_multiplier: Optional[float] = None
) -> DealStatistics:
"""
Calculate statistical aggregations on deal data.
@@ -881,11 +883,12 @@ class GovmapClient:
Args:
deals: List of Deal model instances
iqr_multiplier: Override IQR multiplier for outlier detection (optional)
Returns:
DealStatistics model with comprehensive metrics
"""
return statistics.calculate_deal_statistics(deals)
return statistics.calculate_deal_statistics(deals, iqr_multiplier=iqr_multiplier)
def _calculate_std_dev(self, values: List[float]) -> float:
"""
nadlan_mcp/govmap/outlier_detection.py
+12 -3
View File
@@ -173,7 +173,10 @@ def apply_hard_bounds_deal_amount(
def filter_deals_for_analysis(
deals: List[Deal], config: Optional[GovmapConfig] = None, metric: str = "price_per_sqm"
deals: List[Deal],
config: Optional[GovmapConfig] = None,
metric: str = "price_per_sqm",
iqr_multiplier: Optional[float] = None,
) -> Tuple[List[Deal], Dict[str, Any]]:
"""
Filter deals to remove outliers based on configuration.
@@ -192,6 +195,7 @@ def filter_deals_for_analysis(
config: Configuration object (optional, uses global if not provided)
metric: Which metric to apply statistical outlier detection to
Options: "price_per_sqm", "deal_amount"
iqr_multiplier: Override IQR multiplier (optional, uses config value if not provided)
Returns:
Tuple of:
@@ -239,6 +243,11 @@ def filter_deals_for_analysis(
filters_to_remove[i] = True
# Step 3: Apply statistical outlier detection to specified metric
# Use override value if provided, otherwise use config
effective_iqr_multiplier = (
iqr_multiplier if iqr_multiplier is not None else config.analysis_iqr_multiplier
)
if config.analysis_outlier_method == "iqr":
# Extract values for the specified metric
if metric == "price_per_sqm":
@@ -260,7 +269,7 @@ def filter_deals_for_analysis(
value_indices = []
if values:
statistical_outliers = detect_outliers_iqr(values, config.analysis_iqr_multiplier)
statistical_outliers = detect_outliers_iqr(values, effective_iqr_multiplier)
for i, is_outlier in enumerate(statistical_outliers):
if is_outlier:
filters_to_remove[value_indices[i]] = True
@@ -302,7 +311,7 @@ def filter_deals_for_analysis(
"outlier_indices": outlier_indices,
"method_used": config.analysis_outlier_method,
"parameters": {
"iqr_multiplier": config.analysis_iqr_multiplier
"iqr_multiplier": effective_iqr_multiplier
if config.analysis_outlier_method == "iqr"
else None,
"metric": metric,
nadlan_mcp/govmap/statistics.py
+3 -2
View File
@@ -152,7 +152,7 @@ def _calculate_basic_stats(deals: List[Deal]) -> Dict:
def calculate_deal_statistics(
deals: List[Deal], config: Optional[GovmapConfig] = None
deals: List[Deal], config: Optional[GovmapConfig] = None, iqr_multiplier: Optional[float] = None
) -> DealStatistics:
"""
Calculate statistical aggregations on deal data with optional outlier filtering.
@@ -164,6 +164,7 @@ def calculate_deal_statistics(
Args:
deals: List of Deal model instances
config: Configuration object (optional, uses global config if not provided)
iqr_multiplier: Override IQR multiplier (optional, uses config value if not provided)
Returns:
DealStatistics model with comprehensive metrics, including:
@@ -203,7 +204,7 @@ def calculate_deal_statistics(
):
# Filter deals for analysis (primarily targeting price_per_sqm outliers)
filtered_deals, report_dict = filter_deals_for_analysis(
deals, config, metric="price_per_sqm"
deals, config, metric="price_per_sqm", iqr_multiplier=iqr_multiplier
)
# Create OutlierReport model